Even though it's clear the 5DSR did well, you chose to give it a consistent advantage by not zooming further to get an actual 1:1 scale between the two images.
Imhotep397 I zoomed in 200% for both. A percentage value is the same relative amount. Also I wasn't comparing resolution but comparing lens capabilities between the two systems. Thanks for watching my video and for the comment.
I've shot architecture exclusively for 15 years. I have an Arca Swiss w/100mp phase back w the Rodenstock 32, 40, and a sony a7rii w the 24tse. To claim the canon glass is superior is just plain wrong. Especially to come to this conclusion so decisively on one location test. You need at least an interior and an exterior shot to get a better idea of what these cameras can do. I will admit 35mm has come along way, and depth of field is better. When your tech camera is properly tuned the Rodenstock will out resolve the canon any day. But again I will admit with the new cameras the gap isn't what it used to be. I've been digging into this deeper lately. The a7rii is sharper than the 5dsr in my testing. There is also something with the way medium format renders space that is really nice. But again that isn't so much a technical issue.
Michael Robinson this was not a one time one location test his has been a work in progress for more than a year. Compare the diffraction for yourself at equivalent apertures and you'll see the same results.
Your video showed one scene and you seemed to make this judgment based on that one image from each camera, so one would assume you took one shot. I've got numerous images shot on both cameras and I can't tell you that the Canon is sharper because it isn't. Equivalent apertures mean nothing in the real world.. I only shoot at f/8 or f/11 whatever is best for the situation and the lens. Yes the Canon will have better depth of field. But the area that is in focus is sharper on the Rodenstock lenses 100% of the time. If you put the 24mm on a 5DIII and take a shot, then put that same 24mm on the 5DSR and take the same shot the 5DIII will appear to be sharper. Now if you shrink that 5DSR file down to match the 22MP of the 5DIII the 5DSR file will appear sharper. So appearing sharper in Capture One is one thing but to truly judge these you need to view these files at the same resolution.
Ive watched many of your videos on medium format and having shot canon for years and Nikon and now shooting only phase one you are not doing something right...? Im shooting an older P65+ back and it littlery blows any full frame available today paired with any lens out of the water. I think you are focusing too much on perceived sharpness vs actual sharpness / detail. The term is Acutance. Something no modern full frame can even come close to matching on even a 10 year old medium format back. Have you done any prints? This method also holds true with 120 film. Film can look very soft on screen at 100% but when you print it out its much clearer than any full frame digital cameras. This is the case with Digital Medium Format. I respect the work put into these videos but pixel peeping on a monitor means nothing. Print the work out and then you will see the truth.
Andrew Thompson thank you for the comment and I really appreciate you watching my videos. I've only ever used and tested medium format digital and I can very confidently say there isn't a single medium format lens that can outperform the best available from full frame. I can prove my claims and I demonstrate much of it in my videos. Consider lenses like Zeiss milvus, Otus, Sigma Art, latest canon, Nikon and Sony primes. There are allot of better lenses available from full frame. Your points around sharpness and how well your P65 performs, if you can demonstrate and prove your claims then I'd be very interested.
Andrew Thompson also a quick note yes I do print images for my clients and generally print pretty big. This is one image that was shot on a 5DSR then split into three floor to ceiling panels. This is only one from the series. You can see from the link below I shot the printed image with an iPhone so allot of the detail is lost and doesn't look quite as good as when you're up close to it but you get the idea. 1drv.ms/f/s!AqMU3QFCgyGIoKF_b2IBwZ_N3e37wg
Well, the files I scanned from the 4 x 5 film by the Epson V800 look sharper than your prints. I do agree with Andrew Thomson, I've seen quite a bit of digital prints in the National Gallery and they didn't impress me. My 12 x 16in prints look way sharper and very smooth with no grain whatsoever.
You need to zoom in to the same level on each. The closer zoom makes the images on the right look softer and less sharp. Cant really compare accurately unless they are zoomed in equally. Images always look softer and softer the more you zoom in. Thanks!
The zoom was the same, it was 200% for both. A percentage of something is always the same relative amount of something else. For example, 25% of something is always a quarter of whatever the whole figure is. On that basis the comparison is fair. Also, we're not comparing resolution, it's very clear that 100mp is more than 50mp. Diffraction and lens performance was being compared and the differences are clear, the Canon is better. Thank you for the comment, Richard, appreciate it.
Sonder Creative That's not what he meant. To do a proper comparison you have to normalize the files. Which means either upscale the 5DSR image to match the size of the Phase One or bring the Phase One to the size of the 5D file. That's the only way to properly compare different sensors.
Hey, that was an interesting video, but for the sharpness part, is it possible to see a version of the PhaseOne photos sharpened and down sampled? For sure it was softer and a lot of those megapixels were lost, but I think it would be more fair to down sample it to 50MP (or 42MP) to see if there's any gain in sharpness overall. When comparing noise performance between my A7RII and an A7S, I down sampled all the A7RII files to 12MP to see the differences between the two.
Tomasz Wachla I zoomed in 200% for both. A percentage on one is the relative same distance for another. Also we already know the Phase one has more resolution we're looking at lens performance not resolution differences.
I am curious which Linhof Technical camera was used? The 679CS? The Techno Field Camera? Or the Technikardan? When I first got started in the technical camera systems I was using the Technikardan with a kapture group sliding back adapter and a Leaf Aptus 22, with various lenses, but the main one being the 35mm Schneider Digitar. The tolerances on these lenses is SO SMALL that it's nearly impossible to achieve critical focus with any of these front/rear standard cameras, which is why you don't see any professional architectural photographer using them, at least I never have. I moved to a Canon system with the TSE lenses for a while, but the quality was nagging at me. I finally researched and found the technical cameras that come with the helical focus mounts for 4x5 digital lenses. Alpa, Arca Swiss, and Cambo. I setteld on the Cambo WRS 1250, a phase IQ260 and the same lens you used in this test, the Rodenstock 40mm. I've had this setup for many years now and it's results still blow my socks off, I shoot it at f5.6 or f8 and occasionally have to do multiple focal zones to stack later, the results from this camera setup are astounding, and Ive heard even better things about the Alpa once you've shimmed them correctly. I also have a 5ds and just got the a7riii, but those still can't hold a candle to the Phase. I appreciate the thorough review, but fear that that linhof might get you close to focus but not quite there, I had many shoot where I had the same.
seeing now that, a phase one specialist, setup and took the shots. I also had someone setting up my initial camera system on the Technikardan back in the day, which performed terribly, I do not recommend any of the Linhof's under any circumstances, Phase should fire whoever setup the camera. That all being said, I am very intently watching all the new cool APS-C and FF options coming out and hoping one day I can sell my expensive system and get a light and more affordable system.
Architects are more interested in composition than the IQ at 100% pixel peeping, I've found. Especially now when most images are on Instagram and viewed on mobile devices and the occasional magazine, the Canon system is more than adequate for quality.
I do some work as a first assistant for a high profile architectural and commercial photographer in Melbourne, Australia. On our shoots we will often use the Phase One IQ4 150MP with the blue ring Schneider optics (the best of these is probably the 80mm), but we also use a Canon 5DSR or Sony A7Riii for smaller shoots or personal projects. I can say with absolute confidence that none of the Full Frame cameras that I have used are as good as the Phase One. @SonderCreative I can understand how you may have come to this conclusion using the technical camera that you did, as this can take a lot of experience to correctly calibrate and set up for the sharpest possible images. This is one of the reasons that we use the Phase One XF system, along with the Schneider lenses, as they are all perfectly calibrated and provide by far the best resolved and highest quality image. When the images go to print, the print technicians at the marketing agencies that we work with will almost without fail be able to notice when we have used the the smaller sensor sizes of the Canon or Sony. Image quality and detail is more than just about the megapixels, it has a lot to do with the size of each pixel on the sensor, and it’s ability to capture light. In many of your comments you talk about conducting controlled tests. How can you do this with one system that has been calibrated by someone I assume to be an amateur (you) in using medium format technical cameras, and another that has been expertly calibrated in factory for each lens and sensor combination. I urge you to use a fully integrated system like the Phase One XF, if you really want to compare apples to apples with a fully integrated Canon or Sony full frame system.
Blue ring lenses are fantastic because they've been updated. The video was more about architectural photography and unfortunately, those type of lenses aren't really geared towards that kind of work. Having said that Phase One have now produced more up to date lenses for architectural photographers so it could be interesting to see how they perform. There is also the fact that lenses like Otus lenses don't exists for medium format so if we're discussing optics from an overall perspective, full-frame still has the advantage. The problem is that full-frame lenses are limited by their cameras and for that reason, overall Phase One will produce better images in most comparisons. This specific comparison I conducted was not about the full Phase One system but more about a particular niche area.
@@Sondercreative Phase One have also released the XT technical camera system that is a fully integrated system like the XF. It's what we have had they joy of using for most of our architectural shoots for the few months that it was out before our city got locked-down sue to coronavirus. Before that we used a Cambo Wide RS system that was also very well calibrated, and I still feel that that produced much better images than our Canon 5DSR with the 24mm TS. It would be quite an interesting comparison video for you to do if you were to compare the new Phase One XT technical camera system (I think they call it a 'field camera') and the Canon equivalent. I did think that your videos are very informative and impressive, however I would just caution you to make sure that you're comparing apples to apples before you go and provide the very wide claim that full frame produces better results than medium format. Hope you're staying safe and well during this whole pandemic.
Charles Beggs yes I meant the XT, I’d love to review and compare that so I’ll see if I can get my hands on a system. Thank you very much for the comment and the kind words, really appreciate it.
I am considering switching from canon to Sony, to the ARiii, and using the canon TS lenses still on the Sony, with the adapter, what are your opinions on this? Thanks!
Luke Lozano I don’t personally recommend that switch. Your tilt shift lenses will be softer in the corners and also in terms of sheer usability Sony is a bit behind. In any case try it first see if you like the system and then decide for yourself. Don’t just go with my recommendation or someone else’s. You need to love or hate it yourself.
yes, you are right about d.o.f., but THIS is why these cameras have an Automated Exposure averaging and Focus stacking features, so that any problems with this can be sorted out, ie the focus stack ensures focus, by automatically merging images in a series of shots, giving a perfectly focused image; the exposure averaging does the same, but for noise, and diffraction, allowing you to shoot at the best aperture, NO sensor noise ( no need for filters, or long exposure times, as the automation takes 50 or so shots at a faster speed, no hot pixels).
I don't recall those features being available in the IQ3 at the time of recording this video. I'm aware they're available now in the IQ4 and they're great. Having said that, you can focus stack and merge images quite effectively with pretty much any camera. Although having it built in as a feature is useful.
You're welcome, thank you for watching. Hmm, so not including shifts, then wide open (unfair comparison) the tilt shift wins. Having said that at f8 which is more useful for allot of landscape and architecture, the Otus is sharper but the difference is not massive. At f16 the diffraction is a little worse in the Otus so they are about the same in performance. Both are definitely better than the Rodenstock :).
A problem is that while you can count lines per mm from a scale, apparent sharpness and DR are visual impressions, not metrics per so. So for comparison of quality as viewed, two images should be viewed at the same apparent size, not the same % magnification from sensor output. That's my understanding, open to being corrected. Did I ask if your system displays true resolution at all magnifications? My older version of Photoshop only does at 50 or 100%, the rest are interpolated.
@@Sondercreative I respectfully disagree. It s like comparing a 4x6 print to an 8x10 print. So you enlarge the 4x6 to a 8x12 and the 8x10 to a 16x20? The objects in the image must be the same size. Also your test catered to the Canon's smaller sensor which I'm not complaining because this is my setup and I'm very pleased with the results. I say this because you are comparing the "corners" of a round room. Do you realize the substantial increase in depth of field ratio? I seriously believe it would have been a different story if the wall was flat. The corners would be so much farther from the sensor of the Phase One
The 5DsR stills holds up today. Solid camera and my camera of choice. Canon have always done well with highlight recovery for FF. I also agree Usman, recovered shadows rarely look good with out a fair bit of work done after.
im gonna need you to zoom in the same in both images cause if you tell me it loks sharper when its x¿zoomed into the full wall vs a single brick, well its not easy to compare
LeoVomend thanks for the comment. I recommend you download the images because unfortunately it won't be possible for me to update the videos. You'll also be able to see more clearly if you download and compare yourself. Hope that helps.
Might I suggest to compare pictures at the same printsize. Either blow up the 5dsr or reduce the 100mp to the size of the 5dsr. At least for me that would have made more sense for a sharpness comparison of the two cameras and to point out differences between both.
Michael Schwarz one of the key points of my conclusion was to point out the fact that the lenses for medium format are ineffective and regardless of how many pixels or how good they make their sensors the bottleneck is the lens. Had I adjusted the exposures that point would have been less effective. Thank you for watching and I appreciate you leaving a comment.
Bill Nichols did a similar test but he was using a phase one 40 mp back and his final result showed that the phase one results were much better than the 50mp canon so how do you justify your results even though you were using a phase one 100MP
I wonder if the front standard of the Linhof was out of square ? Years ago when I was shooting 5x4 super wide lens like the Schneider 58XL were tricky to use because the back focus distance was so short. Anything out of alignment on the camera standard and track would cause softness. Also the focus screen depth had to be properly shimmed and checked with a dial caliper or what was in focus on the screen was not the same as on the image recording plane. Also the working aperture of f11-16 meant you had to watch for camera shake outside and shooting from building balconies etc. Finally, view camera wide angle lens are not flat plane designs. The point of focus in the middle of the frame is different to that at the edge due to a convex image plane. Wide angles designed for SLRs are complex inverted telephoto designs to clear the mirror and often have field flattener rear optics.
I forgot something important: if the technical camera was not properly adjusted for the phase one back, using a shim kit, your images will look much worse than they could have. Ask anyone who has ever used an Alpa camera or something similar. The difference is like night and day. It's very likely that the used linhof doesn't provide the extreme mechanical precision needed for optimal results.
burak uestuen as mentioned in the video the camera was set up and used by a Phase one specialist so that argument doesn't work. Yes it is compatible and yes they do perform at the highest level when not shifted in any direction. For the images in the comparison the camera was not shifted in any direction.
Sonder Creative , there is a good reason why phase one is working together with alpa not linhof. I'm 100% sure, that if this test would have been done with the right gear and the proper knowledge, the results would have been totally different. To me it's obvious, that you are not the right person for this test. No wonder you come to wrong conclusions. When you have more experience in these matters, you will understand it. Don't believe me, ask Ming thein. He's famous for his knowledge of camera technology.
A Phase one specialist did this comparison with me I didn't take the shots. Personal attacks, the last resort for someone who doesn't have a valid point or no evidence for their claims. Also you're missing one fundamental point of this comparison and that it's dedicated to architectural photography. The point remains that regardless of what excuse you want to make the full frame system is better with the lens performing better. That's the bottom line and that's the practical side of things. Stop making excuses for the poor performance
Burak, Sonder Creative has mentioned multiple times that he did NOT take the P1 photos; a Phase One specialist took them. Did you not hear or read that? Look up cognitive dissonance. I am not a psychologist, but it might just apply in your case.
I had the same issue when I first started converting my 4x5 architectural workflow over to digital and a salesman at digital transitions in NYC setup this linhof with sliding back system that was just useless, I didn't bite then and finally through my own research wound up with a Cambo system that is calibrated so perfectly, it's as easy as shooting with a point and shoot, and I am getting phenomenal results. I hope one day that everything catches up to the quality needed for big prints, but not seeing it yet, unfortunately. I was knee deep in Cambo when I went and demo'd an alpa system they had shimmed up to perfection and it was mindblowing, was too committed to Cambo to jump ship though.
It must be some fundamental wrong with the comparison. I have a Sony A7R (36Mpix) and a Phase One XF with IQ180 (80Mpix) and it is a huge deferens in resolution and sharpness, if you will call it so, between Sony-files and PhaseOne-files. Phase One have the double resolution and it is impossible not to see that if you compare pictures. But with 80 or 100 Mpix you must focus exactly, hold the camera still - Phase One have a built in seismometer to control that there is no vibrations - and you must have a lens that can perform resolution for 100 Mpix. It would be interesting if you do this test again with a Schneider Kreuznach 35mm LS f/3.5 lens or 45mm LS f/3.5 lens on an XF boy. Or, use an Phase One A-series IQ3 100MP System (made for technical cameras) with Alpa body and a ALPAR f/4.0 35mm - made for 100 Mpix.
Joa Silver there's a significant difference between the original A7R and the current one and the 5DSR when it comes to image quality. The original A7R had some major flaws with the shutter slap and caused allot of issue with image quality. Try it for yourself with a 5DSR and more importantly, the latest 24mm tilt shift from canon. Also part 3 will go further into Phase One lenses that are rated up to and even beyond 100mp.
Joe is correct. I have been doing a ton of research on technical cameras with digital backs. It comes down to the fact that the digital sensor is very sensitive to the angle that light hits it. All camera manufacturers have addressed this in various ways, but largely through retrofocus design, which moves the lens further from the sensor and allows light to come in at a straighter angle. The digital mf backs require a camera like the XF to truly take advantage of the sensor, for this very reason. This also applies to full frame digital sensors and angle of light coming in from the lens. This is why super wide rangefinder lenses perform so poorly on the Sony A7 series. So yes, your comparison exposed the flaw of the technical camera + medium format back, however you gloss over the specific setup you use and dismiss medium format for architecture in general. the Phase One backs would substantially outperform the full frame Canon or Sony in a more apples to apples comparison. What I mean by that is the tilt shift lens was essentially at a proper distance from the sensor, while the mf lens was not. Also, as Joe pointed out, the Phase One A series would have given you better results. It doesn't have the tilt motions you desire, but the body is shimmed to have sub millimeter control of the distance between the lens and the back, to get as much out of these lenses as possible. It also, apparently, helps to use lenses with wider image circles, beyond what the sensor actually requires. If you can't live without tilt/shift, then keep doing what you're doing. But don't lead everyone on the internet who hasn't done the research to think medium format lenses are crap, they are certainly not.
Chris Duesing this whole review was specific for architectural photography. Tilt shift lenses are almost a requirement for that. Consider the context before making your statements. In general the sensors on medium format are better but the lenses from full frame are better. If you’re not going to be using a camera with movements for this comparison for medium format then I would use the Otus lens instead which is significantly beyond both of these lenses.
You have learned to make very authoritative and educated sounding videos, its too bad they aren't based in facts. When that is pointed out in comments you ignore basically every detail of the counter argument. I suspect that is why your video is creating so much controversy. You have managed to make a very interesting and correct point and yet at the same time draw some very wrong general conclusions. Nothing about your analysis reflects on the quality of medium format lenses. It is the shortcoming of digital sensors that leads to soft corners. You ignored that counter argument and spoke down to me in generalities.
Chris Duesing Tone of voice doesn’t get passed through text form and I only talk about the points and the evidence I present. I’m being specific and to the point, please don’t take that as condescension or me trying to be rude. The individual must always be respected but I don’t need to respect your opinions or ideas if you don’t present evidence. You’ve only made claims, claims are easy, anyone can do that. Your point about digital sensors having limitations and soft corners is strange. You do realise both are digital sensors that I’m comparing right?
Perhaps the Rodenstock is not that well suited for the Phase One back. And as some people point out, there might be some lens correction in place with the full-frame examples (I don't know how Capture One is handling this). Sharpness is not always the main signifier, and sometimes you can get more data to work on out of a less "sharp" lens - applying algorithms to a seemingly "softer" image might produce a superior result in the end. But under practical considerations I would also go with a full-frame digital camera (the Sony A7RII in this case) and be fine …
Christian Augustin you make some very good points and I agree to some extent went. Two things, for architectural photography this is the system to use so if it's not compatible then canon is still better. If we're comparing just lenses there is also the Zeiss 28mm Otus which is much sharper than the canon. I'd say the Zeiss is significantly better than the Rodenstock lens. Ultimately full frame still has better lenses available. Thank you for the comment.
I think the Rodenstock was originally made for large format photography on film (it looks like such a beast), and there it might’ve been actually a great lens. But when it comes to specialized lenses with extrem high resolution - current full-frame lenses have the advantage of a broader audience (and therefor get more love than large-format lenses).
I shoot architecture on the A7r with Canon glass. I have have a Canon 5dII and a Phase P20. I can tell you that the 2004 Phase beats the other cameras hands down. Hands down. For me it's the colour and the graceful transition between tones. If I had the MF lens options, I'd use it in my work without hesitation. Also, why do you have only one LCC capture on screen? It needs to be done for every body/lens/aperture/ISO/shift combo.
Joseph Asghar you're comparing it to the original A7R which was known to be notoriously bad. I'm comparing two very different cameras, you can't use your experience with another couple of cameras as a way of determining my findings, that's illogical. The shutter shake on the A7R was really bad and even on DXO mark the same lenses that perform amazingly well on the mark II perform really bad on the original A7R. The same LCC was sufficient in clearing up all of the images however I do agree with you about that point.
Sonder Creative The difference between the two Sony cameras is 5% of linear resolution. Shutter shake is irrelevant to my work, on a gitzo with a manfrotto 410. Your misunderstanding of the purpose and workings of LCC calibration negates your review - it fundamentally affects the image if not specific to the image.
Joseph Asghar the LCC was sufficient, is has absolutely no impact on resolution and detail. Also the difference between both cameras is not 5% look it up. Try comparing the cameras I did and then we can talk until then you don't have a valid point.
And you would be very wise to do so Joseph. Here's an article I wrote about the journey from 35mm digital to medium format a few years back: blog.cambo.com/2012/06/28/a-new-camera-the-odyssey-from-35mm-to-medium-format-digital-photography-by-richard-osbourne/
Thank you Usman for another great video. I actually watched for a 3rd time, but this time I desided to actually download the images you generously provided. I must say that I notice that the focal plane of the two cameras-lenses doesn't match. Suprisingly, is the Canon which seems to have more areas out of focus (most of the right part of the photo, look at "Yellow Sunrise" sign, and the upper part of the left part as well. Maybe you had a slight tilt on the lesn?). I did what the other comments suggested and match the resolution of the photos, and I do mostly agree with you fundings, in areas where both cameras are in focus the canon seems more contrast with almost the same resolution but the phase one seems more detailed due to more MPix. But it is a bit hard to judge since you cannon be sure if this is due to the distance from the focus plane or the performance of camera-lens. I understand that it is difficult to repear the comparison but if you will have the change could I suggest to take photos of boring distant brick walls? Maybe include the GFX 100 as well? (I know you have another video but test shoots like that are easier to intepreted). Thank you by the way of showing the interior of the corn exchange, it look amazing. I didn't managed to visit it when I was in Leeds.
One thing that may not have been pointed out is that the rode stock lens may actually have a calibration issue (which sort of happens and is a major down side to these lenses) they're supposed to have complete edge to edge sharpness and if they don't it's usually to do with a calibration issue
Generally yes and this was something that was considered. However the Lens came directly from Phase One and was calibrated by them before any demonstrations.
@@Sondercreative it is entirely possible that the lens itself wasn't calibrated properly by them - Since Cambo/rodenstock usually do it in house, or at least they do now - maybe to stop this issue that you had? I have Rodenstock being sent away for repair because of a similar fault this one is having, some spotting of soft edging, according to their standards it's not supposed to be like that.
@@jamessummerfield5102 it is possible I agree. I tried to minimise any issues by completing the test with Phase One twice on two separate occasions. I was assured on both occasions that everything was working to the best of its ability. Then again mistakes can happen, but I think its unlikely considering the due diligence.
@@Sondercreative Not that I think it's important anymore since this video you've posted has covered all tracks regarding, something is definitely off with that rodenstock lens. Once I get mine back from repair I'll do a little test in similar fashion - I have an A7Riii so I can do a direct comparison. Maybe that revision of Rodenstock lenses were just garbage and overpriced haha
One thing this video didn't really look into was....... Composition and framing. The greatest difference between the formats aren't just the technical specs like what this video suggest. There is a much greater difference in how a photographer approaches a scene with a medium format vs full frame. I think there needs to be an analysis on how the framing on each format affects the way in which the images get composed.
For me image quality is way more than perceived sharpness, colour rendition tonal transition and all those nuances of hues is where it's at. This is a very interesting video nevertherless.
Thank you Ferhan, I'm actually a fan of your videos, I've been watching and following your channel for some time. I agree with you about colour rendition and so on. I've done comparisons between the Canon 5DSR and the Hasselblad and when you use some of the top end Zeiss lenses the 5DSR actually matches the Hasselblad in colours and tones. Not only that, the best full frame lenses are much much sharper than any of the Hasselblad lenses. Having said that Phase One might be different.
Sonder Creative I'm surprised! There is a video knocking about on UA-cam where the SR is pitted against the A7R2 and a phase one. The A7 seemed to out perform the lot. I think it's safe to say all of these cameras produce amazing results, and one camera or lens won't allow you to do anything more than any other, unless of course one views photographs at 400% regularly lol. Thanks for the kind words regarding my channel, need to get more content out, too lazy lol
Slightly confused, from the files in the description, (CF007362.tif) and (5DSR0638.tif) from the IQ250 file, I can read the menu on the other side of the room (Little leeds beerhouse, at 300%, whereas it's impossible to read from the 5D.. Also, some of the IQ images look slightly soft on the lefthand quadrant of the image, possibly due to the use of the Linhof - it's known for having sharpness issues (several people I know have got rid, due to it going soft as it's very difficult to keep the lens 100% parallel to the sensor with a bellows, camera - you can get away with the tolerance on film, but not on digital) that said the 24 TSE is probably the sharpest wideangle on DSLR, but I think you have some issues with the Rodenstock/linhof combo, (regardless of who took the images)
Hi Tim thank you for the comment. Could you please clarify what you mean about the files in the description part of your comment. I'm not sure I full understand. I think you could be right about the linhoff Rodenstock combination, however the centre sharpness remains unaffected especially because we didn't shift the lens. Also this is still the best combination for architectural photography when using medium format, so comparing things specific for that. Thank you for the comment you made some very good points.
I downloaded the files (CF007362.tif) and (5DSR0638.tif) and looked at them, in the centre of the frame, the IQ250 file looks significantly sharper, if you look at the menu for Little leeds beerhouse, on the other side of the room, as in I can read almost 100% of the menu, compared to the 5D file - where I can't read any of it - it seems to me that the Rodenstock is significantly sharper in the centre, (as I can either read the menu, or I can't, which makes it pretty definitive that the Rodenstock/250 is sharper)
Tim Gregory There's a video relating to that but it's only for my Patreon subscribers. In that I explain the detail differences due to the sensor sizes. The 50 back is a smaller sensor so the focal length equivalent is different. I should actually move that folder because it's only for my patreon.
What confuses me slightly, is that you're saying the Canon is as good as the Rodenstock IQ3, or slightly better, considering the Rodenstock 40 has a 35mm equivalent of 27mm on the IQ3100 it's pretty close to the 24 TSE in focal length, but it blows the 24 TSE away. For example, if you look at the file from the IQ3100 (CF001185.tif) and the file from the 5DSR (5DSR0638.tif) put both images to 300% and zoom to the bottom right quadrant, to the "Yellow Sunrise" shop, the writing around the outside of the logo is just an unreadable blur on the Canon, whereas on the IQ3100 I can quite clearly read "global inspirations, Traded Fairly"
There is still a difference in resolution and the Rodenstock has the advantage of having a 100mp sensor behind it. Detail from resolution is, of course, different from lens sharpness. This is what I'm seeing right now looking at the images, not in any RAW processor and the Canon 24mm is noticeably sharper on the edges and the centre sharpness the Canon is a tiny bit better. Again the resolution difference is there too. 1drv.ms/f/s!AqMU3QFCgyGIoKEejqtTZfvdfcNMFg
I would suggest for architecture the camera with a lot of megapixels and best OIS/IBIS possibly, so you can stop down the aperture, put the iso as low possibly and shoot with long exposure times, since the subject is still, and you actually want to get rid of the dof. So I think Sony has a big advantage here over Canon and Phase One which comes with OIS only. By the way, noise is determined by dot size on senor (photodiode) per pixel. So since Phase One is 100mp I bet it has considerably small dot size or simply nothing out from the ordinary CMOS sensors produced for many years, even if you consider the sensor size overall. Also, Sony's sensor is backlit illuminated, so the sensitivity to capture the photons is probably much higher over latter two cameras. Not without a reason smartphone manufacturers came back to 12mp, even if older models had already 16 or even 20; it's to make a bigger dot per pixel, capture more light and automatically create image with less noise with less processing power needed for noise reduction algorithms.
Hi John thanks for the comment, Pixel size on the Phase is actually bigger than the 5DSR and the A7RII but even with that, the Sony is cleaner in the shadows. For architecture, OIS wouldn't really be an advantage, because most if not all of the people I know that shoot this professionally almost always shoot on a tripod. Shooting on a tripod is far more effective not just because of being able to long expose but also because you can take a number of shots and layer them together without much of an issue. Also, a tripod can be much more effective to compose accurately.
Well, that might be true indeed, I didn't search up the dot size of the phase one. Well, for the tripod, sure. But not in every situation you can use it, and good image stabilization in the camera is always a bonus. I'm not trying to be a Sony fanboy or anything, so don't get me wrong. I used for some time Olympus OMD, and I think that would be a great camera for architecture too with the extra resolution mode and IBIS. Nikon 810 would be great candidate too with the 36MP. And obviously Canon 5DSR. In my opinion medium formats are much better choice for cinematography or macro shooting, where sharpness on edges doesn't matter much and where is actually that super duper dof what you want! I forgot to mention in previous post that you need wide angle lenses for architecture which are again hard to find even for FF, figures for medium formats, unless you want to spend hours for image stitching..
I personally think that if you are shooting architecture professionally you should be shooting on a tripod. Tilt shift lenses and Rodenstock lenses which are specifically for architectural photography are all manual and I find it difficult to imagine someone using them without a tripod. Also, I'm a big fan of Sony, I really don't mind you mentioning it. As for the Olympus, I haven't had a chance to use but I've heard some really good things about it. Check out my Hasselblad review where I compare macro lenses and a couple others and the Full frame was still better. and yes wide angle is advisable lol.
A bit weird at the beginning, but agree that canon tilt rocks and softness on the conner have nothing to do with a sensor it’s only optic lens issue so kinda weird to hearing that “phaseone” doing it vs canon tilt - comparing sensor and lens.
peter brady haven’t tried the multi shot but it’s too niche and the sensor is much smaller compared to the 100mp back. I’d have to use the 32mm which isn’t as great as the 40mm. The underexposed shots were only to test dynamic range.
The Phase One sensor is Sony. Its pixel pitch is slightly larger than the a7rii's. The sensor is a factor in the end result but the motherboard, it's components more so and firmware even more. Statements in the video or comments below ... meh
Possibly but I doubt it because the lenses are regularly checked and calibrated. Focus was also checked multiple times prior to each individual shot so it’s unlikely.
Hi Sonder, great informative video with proof. I specialise in architectural photography but it's more of a hobby of mine than a business. I was about to sell my canon 5dsr with all my canon lenses and my canon 5d mk 4 to purchase the phase one system. After seeing your video I think I will stick to canon until they improve thee lenses for phase one. Regarding the comments here who disagree with well I think psychologically they can't accept your results as they have invested in medium format system. So it's hard to agree with you once you have invested £60k on a phase one system with lenses. Thank you for your honest, transparent comparisons backed up with proof.
Thank you for watching and for the comment, what you said means allot to me. I was actually looking to sell a bunch of my equipment and make the leap to a Phase One system too. I only made these videos because of my findings from comparing the systems and realising how they're not really the best but offer worse image quality in many areas. I agree I think that had I have spent that much money I'd be less inclined to believe these findings too. I'm very happy to know this video helped you make your decision. I'm certain with the best lenses from full frame you will get much better results.
Man! Thank you so much for this video! I was about to make an investment in Medium Format just because "everyone says they are better". I never saw a review with that much depth and files being compared side by side! I'm sticking with my Canon bodies and their range of incredible lenses.
Your comparison is nonsense. It's known that canon, Sony etc. strongly precook their raw files. Apply some sharpening, contrast etc. to your phase one files and they will look much better than any FF file. Either you are biased or you have too little experience. I've been working with the best FF like Nikon d810 etc. and phase one backs for years, also with canon and the ts-e II 24 mm, you used for your test. I have a lot of experience with both formats. Your conclusion is simply wrong. You're right about ease of use. But if you want the best image quality possible, you have to work harder and use medium format gear. It's been the same in film days, the gear was even bigger than. A 8x10 inch film camera still delivers the best IQ on the planet, if you know how to use it!
burak uestuen you haven't compared them properly then it seems. All sharpening and anything extra was pre removed. Also sharpening does not bring back detail it only gives the illusion. The heavy diffraction in the Rodenstock lens cannot be adjusted by simply adding some sharpening. Full frame is better this was done with the help from Phase One specialists and over a long period of time going back to them to ensure I was getting the absolute most out of the Phase. Excuses for bad performance.
Sonder Creative , you can not remove all sharpening and other precooking of canon, Sony, Nikon... raw files with Capture One or Lightroom. That's a fact. The rodenstock people clearly say at what aperture the lens should be used. If you don't listen to them, it's your own fault your images look like garbage. Did you even listen to the Pros that helped you? I doubt it.
burak uestuen So do Phase one and Hasselblad. Again sharpening and cooking images cannot overcome diffraction or loss of detail. That's a fact. The lenses for medium format are terrible, still old design lenses that can't compete against full frame. Consider the budgets for medium format they don't have the money available to invest in better lenses. Medium format lenses are horrible in performance they have more diffraction at equivalent aperture and also cannot produce shallower depth of field at the wider end either. No advantage in any area.
burak uestuen The rodenstock was shot at various apertures and all of them it couldn't perform better than full frame and was worse in every test. The recommended apertures are between f8 and f11 and at those apertures you can't get that deep DOF and also the lens still performed worse than full frame at those apertures. Look I get it it's difficult to accept, this is why I had to keep going back to phase cause I couldn't believe the results the either.
Sonder Creative , believe whatever you want, but if you are ready to overcome your narrow view, talk to professional photographers about this. Maybe they can change your mind? I'm sure they could show you some printed work, that will blow you away.
I am probably opening myself to attack, but something doesn't make sense about saying the Canon is better. The plane of focus is supposed to be a *plane* , but the scene is clearly not a flat plane. If the Canon is sharper at the edges when opened up, isn't that just a sign of extreme field curvature? The edge of the image of a curved scene *should* be soft, shouldn't it?
eric moss not opening you’re self to attack at all discussions are welcome. Both cameras are relatively softer on one side or the other. So the Phase is sharper on the right and the Canon is sharper on the left. I only noticed this after someone pointed it out. Having said that the canon is still noticeably sharper in the centre at equivalent apertures.
Best guesses: You'd use the Linhof med format camera you happen to have for the more extensive camera movements for perspective correction, and you can get many focal lengths and still have those movements (I was not aware Rodenstock was the best lens brand for med format). The highlight recovery is more important than shadow recover for architecture. And you can demand more money and fend off high end customers better with the med format. No, none of those reasons is a good enough argument, and you never even got into editing for perfection in Photoshop with the FF files. That defense offered, I happen to own that 24TS, and so this vid is a pleasure to see. I just sold my 5D2 on eBay, and so now I suppose I get a new Sony and anticipate a super Canon 5DsR redo next fall.
John Smith the extra movements are only really an advantage on film. On digital the colour shifts really start to impact the image of you shift to much and it essentially degrades the image. Shadows vs highlights is a preference. Highlights don’t I introduce noise but shadows do so I’d say that’s more important. In regards to Rodenstock being the best, thats what I’ve been told by several professionals in this market but I guess it may be subjective. Thank you for the comment and good luck with the new cameras.
I disagree, as I don't associate camera movement with color shift, nor do I associate DOF with focal length (it's an effect related to "circles of confusion" size in any format size for given field of view. And I differ on diffraction as anything other than a relation to the absolute size of an aperture). I made a post because I very much appreciate your presentations, and I will when I can have a bit of dialogue, not to smack back. On that point, I can't catch your name, which I would like to know as background info, as I intended to view your presentations as they are published. My go-to guy online is Tony Northrup, and he sets a very high bar indeed. You are very complimentary to his work. I'm old, so I do know some things, but like all camera science types I can never get enough of solid presentations. On that point I wonder if you have looked at Kasson.com? It's a blog by a gentleman out West USA who really does some heavy hitting data based science with a variety of products and issues. When I read pg 61 of the A&R2 manual and learned that steady shot will not work on a tripod I was bummed, but Kasson did the science that proved to me not only does it not work, it harms sharpness. His data work is very impressive. Sorry for the length here, I type fast. I welcome being shown to be wrong, that's how a person learns. I am very opinionated and passionate about some aspects of art, including photography so don't mind my intensity. I'm in baby, I subscribe.
John Smith focal length does impact DOF when shooting two lenses at any given distance, aperture and pixel size. Full frame (in this comparison) will have greater circle of confusion and both have a very similar angle of view so you’d shoot them from the same distance to get the same framing. Effectively the full frame camera can produce greater dof at wider apertures when shooting from the same distance with “equivalent” focal lengths. The colour shift occurs on digital sensors when you shift the lens this can be demonstrated quite easily with wide angle lenses.
I presume we acknowledge the same facts, but with differing methods of thinking about them and so different ideas about teaching them. I approach the idea of DOF as "caused" by the needs of the format. Small formats have large DOF because the light ray projection travels a shorter path to the format, and so spreads less. Large format has a longer path & spreads more, so less DOF. Small formats have problems with diffraction because an fstop is a ratio of length to opening, and a "normal" (diagonal of the format) lens is shorter on a smaller format, so any given aperture is dimensionally (real measure) smaller so with more tendency to cause diffraction (hence you won't see f64 on a small format.) So, tell you what, I promise to not torture you like this anymore with my quirky pov, and just watch some of your vids and pass the word that you know your shit and express it well! If I had more $$ I'd hit you off, I like you work, but you are going to be big anyway and I have my own money issues.:0 You are very generous in your replies, I'd suggest you step back for your own peace of mind. I've read some of your threads and some people do like to just argue. I'm pleased to discover your tube!
I agree with you I think we are discussing the same facts but with slightly different methods. I would say that although Full frame lenses don't have F64 they won't necessarily need it because you can simply shoot with a wider lens and get a similar result. In the same way, you won't get f1.2 or f0.95 on medium format because you can just shoot with a longer focal length. There is one advantage for full frame and that is exposure is going to be faster due to the wider apertures available. I don't think you're torturing me at all and the discussions I have on UA-cam are always fun for me until someone becomes personal against me, at that point, it's no longer a discussion and I end the conversation. You have only discussed the points and for me, that's very enjoyable.
As to the Rodenstock lens - which one is it? An old 4"x5" lens or a modern "digital"? Resolution in larger format in line pairs per mm on film was always lower than in top small format 35mm. Sensors now are beyond film capabilities and lenses have become the image quality bottleneck. In this light it is extremely important to know exactly what lenses were used.
Sonder Creative - this is a modern "digital" lens. You can read in the specs it is acceptable down to 5 micrometer pixel pitch - that you are near to with your sensors. Your test illustrates you have hit the resolution limit. Looking at the design, my guess is, this lens should do especially better in absence of chromatic aberration.
Re-reading (YT recommenreminders). Nice comparison, b.t.w. The Linhof can maybe handle 4"x5", but the Rodenstock's lens cannot. This lens's image circle is 90 mm and this sensor/back is the largest Rodenstock recommend. 4"×5" diagonal is 163 mm so you'd need am image circle at least 200mm (8") to be able to do a bit of tilt and shift. I think the answer to the surprising observation is in Rodenstock's advice on maximum resolution (5 micron). That would set the limit for the digital back at 80 mp. Corner sharpness issues, meh.
Good review. I've got the Canon 5DSr and use the EF 24mm TS-E II which is exceptionally good for what I use it for; capturing monuments and built heritage subjects. I do occasional interiors as well and its perfect. I've even used it for landscapes which is my bread and butter source of income and frequently make 4ft prints from it which are near flawless. Unless you're printing billboard size stuff, most don't need an over complicated, overpriced elephant like that technical camera.
Graham Harris Graham Thank you so much for the comment and for watching. You know the 5DSR would be more than sufficient even for billboards. Viewing distances mean you don't need high dpi prints. Thank you again.
I hate to say this but I use a Rodenstock 40HR with an IQ260 and I don't recognise the results you are getting here. It literally eats the Canon 24TSEII for breakfast, in terms of sharpness and at the edges. I know because I used that lens extensively for years in architecture. It looked to me that the Rodenstock wasn't focused correctly - which would give you massively skewed results. Also you are comparing both files at 100% when really the Phase files should be at 70% or so to give an equivalent FOV. And I'm not sure why you are looking at files at F5.6...I use F11 an F16 on the technical camera exclusively. In my experience, the depth of field of technical camera lenses seems to be greater - at least, it looks that way to me. I hardly ever need to touch focus to get razor sharp depth of field front to back.
funland108 I compared various apertures from f5.6 all the way to f22. At equivalent apertures the Rodenstock isn’t sharper it’s a worse lens in terms of sharpness and will have more diffraction. Once again at equivalent apertures and I explain why equivalent apertures are a true and fairer comparison. Also 100% if something is the same relative zoom. 25% of something is always a quarter regardless percentages make it a fairer comparison because I didn’t test resolution from the sensor but sharpness from the lens.
Well, I have to add my comments to this thread since I saw this video and also shoot architectural images for a living. I love good cameras and great lenses. Basically, good or reasonable quality equipment is a given. You must have the right stuff for the work you do as a professional. However, none of this means anything in comparison to how you as a photographer treat your subject. Architectural photography is a story telling process. The idea is to convey what an architect had in his mind regarding how best to solve a problem. This can be done with or with out superior equipment. I'll take a unbelievable photograph that has a less quality but really conveys the point over a ultra high quality images that tells nothing about the subject and add nothing to the story. Composition, lighting and context are my top three priorities not the resonation of one lens over another. Buy what you can afford and do the best you can with it. Learn to be a photographer not a technician.
You misunderstood the concepts of angle of view, depth of field on different format size and magnification ratio. Depth of field IS different at different focal lenghts, but since the angle of view is similar on the lenses, the justification for the difference in D.o.F. is caused by the fact that the bigger the sensor the shallower the D.o.F. is, at the same focal lenght and same aperture. On a 24/70mm lens the bokeh is different at 24 and 70mm because of angle of view AND magnification ratio. You should compare difference in bokeh keeping the framing the same in both shots.
Stefano Catalani sensor size does not affect DOF. Circle of confusion aside smaller sensors and bigger sensors have the same DOF if the distance to subject is the same. What you’re describing is the difference in DOF due to changing the distance to subject but distance will change your DOF regardless. If I take a picture of something using a 24mm lens and then crop the image in Lightroom the angle of view has changed but the DOF and focal length is the same. In my experiment I demonstrate how when shooting from the same distance to subject, different focal lengths will give a different DOF even though the angle of view is the same. This is a physics based fact of light.
should have considered the olympus M4/3 EM1 mkii high res-mode 80mp RAW, then you would really put 'the cat among the pigeons' I think the matter of scale should have been addressed, as the 100mp images would have more detail at the same scale, plus the noise would have been less at the same scale, either way thanks for the video, ignore the haters as UA-cam is always subjective.
Jamie R. Thank you for the kind comment I really appreciate it. In regards to scale both were zoomed in at 200% which is the same relative amount. I was aiming to test the lenses instead of resolution. I’ve heard some really good things about the Olympus, unfortunately I don’t think it would be as effective for architecture due to not having any native tilt shift lenses. Do you have any thoughts about the pixel shift feature in the A7RIII?
I downloaded the files, and as several people have pointed out - the Canon is very soft in the corners compared to the Rodenstock - which seems far sharper, I had to check that I was looking at the correct images - because this doesn't match what you're saying in the video - quite the opposite. I also watched another comparison of someone testing the same lenses, but with an IQ140 - and it's still far sharper than the Canon, I think there's something amiss with your testing regime.
Herpa Derpa the canon is significantly sharper on one side vs the other however the same is true for the Rodenstock. The centre sharpness however is much better with the Canon. Also if you look at them at equivalent apertures and NOT the same apertures then the canon is still sharper due to less diffraction.
I don't know why you're transfixed by centre sharpness, any lens stopped down to F8-F11 is going to be sharp in the centre, even a 17-40L stopped down is sharp in the centre..... That said, I can read the menus at the back of the room on the phase, not on the canon - they're just a blur, so even taking focal length and Mpixels into account, I'm not sure why you're banging on about the canon being sharper... It clearly isn't, and other peoples comparisons on more uniform images show this. As far as lens resolving power goes, the battle is won and lost in the outer quadrants, and as per the MTF charts, the contrast and resolution falls away drastically on the Canon, as it does on your image, whereas the Rodenstock is far more uniform edge to edge - especially in the outer quadrants and corners, this is why the Digarons are so expensive, and are some of the only lenses that have enough resolving power to meet the strict standards of aerial photography, where specific levels of contrast and spatial resolution are required - that don't exist on 35mm lenses.
Herpa Derpa I'm not just banging on about something without providing evidence. The sharpness differences on the edges isn't as uniform as you're claiming with the Rodenstock and diffraction is horrible on the lens. Ultimately the Canon tilt shift is much sharper and centre sharpness is obviously more important than edge sharpness regardless of whether you admit it or not. The other claims you're making if you can demonstrate and prove them I'd be very interested but as with most medium format related stuff it's just unfounded and widely believed claims, nothing more.
You don't know what you're talking about really. Even the MTF data for the Canon shows that it can provide less than half the contrast and resolution of the Digarons, and falls apart at anything more than around 20 lp/mm compared to the Digarons which are good to around 80 lp/mm @ MTF 50. And centre sharpness is not more important with wide angles, that's the whole point of MTF - resolution and contrast across the ENTIRE image circle, that's why very high quality wide angle lenses cost so much money - because it's extremely difficult to make a wide angle lens sharp in the corners, that's what you pay for when you spent lots on a wide angle. You should know this seeing as you claim to be an architecture photographer lol As for claims I'm making - other people in the thread have provided samples of the images showing quite significant distance between your claims vs the reality of the files - you've simply cast them aside as being wrong, so I'm not going to do it again because it's obviously wasted effort.
Herpa Derpa the reason why they're wrong and why you're wrong is because you use one small section of the evidence to draw your claims. You're also not taking into account the heavy diffraction the Rodenstock lens suffers from. I'm using the overall aspects of the evidence to present my argument in a fair and balanced manner and taking int account the diffraction the Rodenstock lens has. Again centre sharpness is still more important you simply won't acknowledge how the canon is superior in that area. The edge sharpness was also better on the canon although both lenses are performing slightly off one a particular side. Consider the fact that the subject is curved so this may be adding to some of the anomalies. The Canon is never soft on one side vs the other. There are other example where the canon has been significantly sharper on both sides of the edge vs the Rodenstock however that will be discussed in an upcoming video. Talk to me about the diffraction because all of the people who disagree have completely overlooked that aspect. Talk to me about the lens performance at equivalent apertures. So easy to pick an choose the info and please don't bother trying to use that argument against me because I have discussed at length and agreed where I was wrong.
Who would have though the 24mm from Canon could be so good. Thank you for your continued support, I really appreciate you watching and leaving such awesome comments on my videos.
Love this video! BTW I own a Nikon D810. Some days I think I'd be happy w/ an APS-c sensor (saving money size and weight) other days I wish we could afford a medium format (and that digital medium formats were full frame medium format (like 6x6) instead of being just 35mm plus)...anyway when I catch myself going from one thought to another I figure 36x24 (as in my D810) may be the sweet spot.
Interesting video.. I have to say though, currently owning the Canon 24mm TS-E II, I find this test a little bit pointless when you are talking about architecture photography but there is no mention of any type of movements? That is when you really see a lens of this type either fall apart or shine...
wallpaperviking comparing the lenses with the movements involved gives a distinct advantage to the Canon lens because it's been fully and properly designed for that specific mount. Too many variables involved when shifting the lenses and comparing that to make a proper test out of it.
Fair enough.... I just know that this is another thing that the Medium Format lenses get praised for, ultimate corner sharpness. What adapter was used for the Sony A7RII? Any loss in sharpness?
Interesting comparison and it shows that full frame is up there for now, but ultimately it will loose out to medium format as pixel count rises and medium format sensors improve. But you are right it's not all about the sensor you have to have the best lenses as well and medium format cost an arm and a leg. Personally I'd use whatever system the client wants me to use, as they say they are only tools for the job, right! To working professionals it isn't about pixel peeping, it's about all the other aspects and keeping the client happy:) If they aren't bothered(!) what you use then use whatever suits you best for the job in hand. At the end of the day if the client is paying then the client gets what the client wants! I think we should also remember the high medium format megapixel chips that Sony are producing are being driven by demand and that's a good thing for everyone. If you can write off the cost of a Phase One or Hasselblad H6 etc. then do it. If not do what lots of pro's do and hire the kit when you need it and pocket the savings! Cheers:)
Hi Kenny, thank you so much for the comment. You raise some interesting points but I don't think I can agree with all of them. For instance, in regards to pixel count and sensor size, similar arguments were made in the early days of full frame about how many pixels can be put onto a sensor. There are far bigger budgets and R&D capabilities available for full frame compared to MF. Canon individually is bigger than DJI and the whole MF industry combined. Currently, there is already a 250mp APSH sensor available from Canon for industrial use which works very well. Tech like BSI sensors and so on make FF more effective even with smaller pixels. Technology evolves and improves and FF manufacturers have the advantage in all those areas. Also, pixel peeping is very appropriate when the price difference is so significant. You have to ask yourself why are you paying so much more for worse results. Even if the results are only slightly worse it's still an illogical waste of money. Even hiring the equipment doesn't make sense because you're still getting worse results. People would justify the cost of MF due to the supposed better image quality. This is no longer the case, you now spend more to get worse. The lenses need to be updated for MF but they just don't have the money for it. MF is a small industry with very little capital for future investment compared to full frame. Sonys FF sensors are outperforming their own MF sensors in several key areas because that's where the money is (relatively speaking). We don't have to agree, however, Kenny, I'm very thankful that you took the time to write all of that and for watching my video. It really does mean allot to me. I hope you're well.
Canon's chip was apparently done as an exercise and for the obvious PR, it's not going into DSLR's etc. anytime soon I think, with perhaps clients with deep pockets e.g. the military possibly, perhaps! 250mp's is going to need some serious hardware to keep up with it. Count me out, my old Mac couldn't cope :(. Sorry hiring does make sense if your thinking of parting with a large chunk of cash and after all if the client is paying, who cares! As for medium format's supposed better quality, it's a yes and no answer. It was in the past due to the larger file sizes than APS etc. Now however, there have been great gains in full frame 35mm CMOS sensors as that's where the money is. Yes I did give it a great deal of thought at one time and tried out both Phase and Hasselblad H systems. I did buy Phase but fitted it onto a Hasselblad H camera, and at that time the larger file sizes were better than what was being offered by Nikon and Canon, Sony at that time was… well crap to be honest. How things have changed eh! Medium format is still playing catch up, but then again, the elephant in the room is Fuji! With the prices dropping as is evident with the new 'medium format' GFX 50S will we want 35mm full frame, obviously only you can decide. So bigger Sony sensor; 50mp less noise than 35mm full frame; relatively affordable; slower certainly; modern Fuji lenses; or full frame 35mm….hmmm..let me think about that! Hopefully this will shake up medium format, Phase One, Pentax and Hasselblad had better look out. How long Fuji intends to continue using Sony's Bayer sensor though will also be interesting. Yup I fully agree MF lens are in the main long overdue an update in some cases, but then again it would have been interesting if the comparison had been made on one of Phase's new bodies and a suitable Schneider lens e.g. the 120mm f5.6 tilt shift lens. I do wonder if the Schneider lens used was faulty though! Not sure, but the 100mp Phase back is a CCD sensor; right(?); if so it's not going to be was effective as a modern CMOS sensor in the shadow areas. Although more than capable for most needs. Re-updating the lenses, yes by all means but you need deep pockets for MF lenses, new or second hand! Cheers:)
Oh yea I completely agree the 250mp sensor is solely for industrial use and not for DSLRs, but, it demonstrates how the technology isn't that far off and the budgets are with FF as opposed to MF. The Phase one back that was used in my video is the latest most up to date modern CMOS sensor with 100mp. It's the current flagship which means that is the absolute best that medium format can achieve in the shadows and so on. Also, the Rodenstock lens was in perfect condition and someone sent by Phase was there to monitor the whole process. I think even with companies like Fuji making their medium format if the lenses aren't up to the standard they will still be behind. The GFX only shoots at 14bit so is there any advantage? Fuji are the ones that make the lenses for Hasselblad and their flagship lenses are terrible in performance. I'd love to try the GFX before I make my conclusions on it though. I have also tested the 50mp Phase One back which is the same sensor that's in the GFX and it's performance was noticeably worse than the 100mp back meaning much worse than the Sony FF. I don't, however, want to use that comparison as a way saying what the GFX can do because Fuji may have updated theirs. I've also tried the 120mm tilt shift and it's an ok performer, deceptively light it doesn't weigh very much at all considering how big it is. I do have a part 3 coming soon where I look at some of the latest Schnider lenses the ones with the blue rings.
This was done with the Phase IQ3 and images were processed in Capture One. When you say S/K lenses, are they Schneider? If that is the case then full frame would still win because I could simply use an Otus lens which is much sharper than the Canon tilt shift. Thank you so much for watching and for the comment I really appreciate you taking the time.
Sonder Creative but your video clear stated you were using Lindhof camera with Rodenstock lens, then you knock the lens for being old and out of date. When I wrote s/k I was referring to the Schneider Kreuznach blue lenses. I'm not trying to be difficult, but as soon as you said lindhoff and rodenstock i questioned whether or not I should take the time to watch the video.
Savage Frieze I didn't for a moment think you were being difficult. Sorry tone of voice doesn't carry through in text form. Rodenstock lenses are still in production and some of the best lenses available for medium format. SK lenses are good but not anywhere near as good as the best available for full frame. Rodenstock lenses may quite possibly be better than SK lenses too but I haven't compared those.
Sonder Creative indeed they are wonderful lenses. My wonderment is that you are comparing three different pieces phase one back, lindhof camera body, and rodenstock lens against Sony and canon systems. I would appreciate seeing the Phase One back, xf body and Schindler lens, he whole system as it was meant to be used, compared against the canon and Sony systems. If you have done that comparison please let me know where I can view the results.
Savage Frieze the reason is because this is the best system for architectural photography. This was a very specific test for one kind of photography. Part three will go further into phase one and some of the SK lenses too. Thank you for the discussions please keep in touch and I'm happy to answer any more questions.
You look to be confused... (sorry for saying that). What do you compare? Medium format vs. 35mm or do you compare lenses? You show the result of the lenses, but pass the buck to the cameras and sensor size comparison.
Photone Photography What do you think I should discuss when it comes to the sensor size? I mentioned how you need to shoot with a longer focal length to get a 24mm full frame equivalent and that's actually not an advantage. The shallower DOF due to the longer focal length (because of the bigger sensor) means you have to stop down allot more and then you get diffraction. Bigger sensor doesn't always mean better.
What do you compare in this video: the performance of the sensors Canon vs PhaseOne? Or the optical performance of the lenses: Canon 24mm vs Rodenstock 40mm? For instance: if you mount a Schneider lens on a 100 MP PhaseOne you get drastically another result. If you compare the performance of both cameras (Canon vs Phase: which is a ridiculous comparison, like comparing a family car to a pickup truck) then, you are supposed to keep the same lens as invariable. If you compare the performance of lenses then you are supposed to keep the same camera as invariable. Sorry, but the video sounds pointless. It doesn't tell what is wrong, what is better, which one sucks (PhaseOne 100 MP sucks? You'd better get a Canon? Never shoot with a Rodenstock lens? etc.) what am I supposed to change in order to get a better result? What if I shoot with a PhaseOne IQ250 and a Schneider Kreuznach 35mm LS f/3.5 lens-for instance? Hundred different possibilities... each one gives a different result. Neither scientific, nor informative. I am really not trying to dispraise, with a friendly intention, the video does not make any sense at all.
Photone Photography I compare the lenses and the canon beats the Rodenstock quite comfortably. I touch upon dynamic range a little and demonstrate how shadows are better on full frame and highlights are better with MF. That's all I compare. Also this is part 2 and so far there hasn't been a single lens from medium format that has beaten the best from full frame. Part three will further demonstrate that point about MF lenses.
Hi! Still haven’t seen the third video in the series. Please post soon. If your results are the same or different from critics stand tall and be proud of your work.
Savage Frieze hey thank you for the comment. There has been a delay but I think my timing for the video will actually work out. I’ll be uploading several new reviews and comparisons with a bunch of medium format cameras. They are coming I promise.
Interessting comparison. It seems to me, that looking with the naked eye at a print I most probably could't tell the difference between these pictures.
DeletedDelusion you’re probably right. Once printed and looking with the naked eye they may look very similar. Goes to show spending all of that money doesn’t give you better results. Thank you for watching and for commenting, much appreciated.
Excellent video. I'm not surprised at the hostile comments. Over the years I've seen similar reactions any and every time technology has brought two classes of photographic equipment closer together in performance. Can the 35mm FF bodies completely replace the MF Phase One in every respect and for every purpose (every lens, every subject, every print size)? No. But the capabilities have never been this close between 35mm and 645. For a wide range of lenses, subjects, and print size combinations you could shoot with any of the three cameras and produce excellent results virtually indistinguishable from one another. And as you demonstrate here it's entirely possible to get a better result, in some cases, on the 35mm bodies simply because the best 35mm lens for that use case was better than the best available MF lens. We truly live in a golden age of photographic equipment and image quality.
I too use TS-E 24 mk2 on 5dsr and find that it can easily resolve cleanly down to individual pixels; it is by far sharpest and highest contrast lens I have used. The only Canon lens I have tried that comes close to the TS-E 24 mk2 is the 300mm f2.8 IS. Thank you for showing, what I think, is your unbiased comparison both in dynamic range and resolution of the three cameras, which by the way I totally agree with. Thank great for the video
Thank you very much for watching my video and for the really kind comment. I was very surprised at how good the 24 tilt shift really is, I agree it's the best lens for architecture and on the 5DSR it's incredible. Thank you again :).
I like to see how the images would come out if you shot 4x5 film instead from that linhof technika camera and compare the neg scanned vs the canon...Im sure the linhof would be sharper overall on the edges, granted if it is shot correctly vs having the medium format camera shoot the ground glass. The circle of image would not be the same an perhaps that is why the edges are softer on the phase one. And maybe the Linhoff should be replaced with a better camera and lens setup like a full featured field camera like an Ebony or Chamonix 4x5.
The same could be said about medium format digital vs medium format film. Film captures far more detail. tones and doesn't produce any digital noise. Point is medium format digital is not anywhere near s good as it's being made out to be. Full frame Digital is noticeably better in almost all areas except megapixels. Also if you were to use a non tilt shift lens on full frame vs a non-technical camera for the Phase then full frame would still win because I would just use the Otus wide angle lens. The Otus is much sharper than the Canon tilt shift which is already one of the best lenses you can buy.
what you like on the images is the contrast and colour of the Canon. The Phase One gives you a flatter image for being processed. And remenber that you are not using Phase One Glass. I mean it when i said it. Test a M43 against your Canon. You will like the Olympus more... its your thig. And mine too :)
I will be uploading a video where I compare native glass on the Phase too. Also in regards to contrast and sharpness, do you believe the Rodenstock is not suffering from diffraction more than the Canon? I actually really want to try the Olympus, I've seen some awesome stuff from it.
they are both getting diffraction really early. REsolution on the Phase shows you more in detail the weaknesses of the lens... its hard to say. where can i get the raw files?
I'll see if I can get hold of the RAW files again, in the meantime I do have some high resolution TIFF files available to download via the link in description.
Fantastic video! Thank you very much for making it! It would be quite interesting to see the comparison with Nikon D850. Although I feel like maybe the latest Phase One lenses should have been used for the comparison.
I might be able to do a comparison with the D850 I'll see if I can get hold of it. Phase One lenses are amazing however they're not geared towards architectural photography, therefore, would have been ineffective for this comparison. If I were to do that test I would use an Otus lens on the Canon.
Sonder Creative Thank you very much for your answer, I really appreciate it. It would be amazing if the Otus line were expanded a little. I have had the privilege of toying with them a little bit and they truly are fantastic. Although, stopped down I see absolutely no difference between the Otus and the Milvus line in real life.
The review is nice because it makes you understand that you do not need a 30k back to take good pictures. 10 years ago there was a big performance gap between digital backs and Nikon/canon. Today that performance gap has decreased a lot. Aniway, I was lucky enough to shoot both formats and i can state that today A DIGITAL BACK IS STILL THE HIGHEST IMAGE QUALITY TOOL AVAILABLE ON THE MARKET. 16 bit and superior dynamic range. Bigger is better hands down. In this specific case of the architecture photo the full frame camera performed better, but no 35mm camera can give you the results of a digital back mounted on an optical bench in studio/table top photography. It all depends what you are going to do. I agree anyway that the cost of medium format is the hot button issue. Cheers and have a nice day.
tonyrebel28 I’m not so sure I can co-sign everything you said. I’ve used and continue using medium format regularly and with cameras like the D850 the advantage is with full frame now. The best lenses are with full frame too. In regards to colours I do show how you can achieve medium format colours and even better in my latest video. This comparison was done with the Trichromatic sensor too. Having said that yes the sensors in these cameras are the best but they don’t have the lenses nor the systems in place to take full advantage of those sensors.
to compare the Phase one back PROPERLY with these, you need to use the back on the XF/XT models, with NATIVE lenses, using the back on a cambo actus is NOT the best to show off the back. with Native lens, image info is passed to the back ( lens data is on database in back), and corrected for, ICC / black balance is saved, and applied to the image in C1 once it is opened, giving best output possible, from the raw files in the camera.
XT models didn't exist at the time of recording that video and Native lenses weren't effective for architecture at the time. This is a comparison specifically for architectural photography. I'm working on getting hold of the XT though.
There are movements on a view camera to manipulate the film plane since LF lenses are longer to achieve the the same DOF. You can't learn the movements in an afternoon for a comparison test.. Sorry, but you need to have an expert on LF photography set up the test and I would just imagine the results would be much different.
steve tilt shift lenses have movements too lol. Also you don’t get a longer DOF you have a different plane of focus. I’m well versed in being able to change the plane of focus with tilt shift and tech cameras. Stop assuming my ability without actually knowing.
Take away what you think is personal and you still got one unfocused image .. Use a flat wall with some texture and use actual large format film and then the phase one and canon crap and try again.. Use a cheap flatbed scanner on the LF film and you'll see resolution man.. That phase one BS and that canon crap will be just that man.. Nothin personal just do it right.! To many variables the way you presented that man.. Dig?
steve no one said anything about large format or suggested anything like that what are you on about? The image is in focus you just don’t like the results get over it. The variables have been countered effectively. Once again get over it canon lens is sharper.
You could not have missed the boat further here. Dynamic range: You said the dynamic range was not much different yet, yet it provided much more shadow quality than one of the SLRs and several stops more in highlight than both cameras....Soo what does that tell you about your exposure on a high dynamic image with the POne? Tells me that you left a massive amount of dynamic range on the table. Shoot for the highlights in a high dynamic range image and get back to me. Tonality. No mention of the quintessential difference between MF and DSLR? My dinosaur 22mpix hassenblad (retired) has much better tonality than my D810.... Amateur really. Micros contrast? What? You do understand that sharpness is a direct trade off with other desirable attributes in a lens? Have you not wonder why a MF landscape lens has 4-8 lens elements while the Canon has how many? Huh! Why is that? I doubt you can even see the difference. Educate yourself...and your eyes Saturation? Saturation is lost with every piece of glass you add right? You though saturation was a slider in the edit? No comment on the look of the focus fall off? BS you say? All this tells me is that you have no idea what you are looking at.
Smokin07ram shadow recovery for full frame is better highlight recovery for MF is better. Even the 5D4 beats the Phase for shadow recovery but not highlights. It comes down to which you value more. Please demonstrate your other points. Claims are easy.
Zef Markaj does the alpa have the same amount of movements required for architecture photography? Also I didn’t decide on which tech camera we used this was decided by phase one based on the best possible option for architectural photography. Remember this was specifically for architecture otherwise I would have used an Otus for the comparison which is significantly beyond both these lenses.
Nonsense comparisons...if you get a commercial gig from an advertising agency shooting sony, I'd be really impressed. Bottomline is, regardless of any comparisons, the art directors wanna see that big fat medium format digital back when they pay 80k for a photoshoot for their client.
You've really missed the point on this one Natan. Also, I've worked on several large commercial shoots and I don't use the Sony, ever. In any case do you wanna know who actually cares about the gear, other photographers. The art directors, clients, agencies none of them care about what camera you're using and on many occasions, they don't even know what you're shooting with. The bottom line is always the result because that's what makes money for the client. Don't get me wrong I do use medium format on occasions for shoots depending on the clients printing requirements. The point of this video was to show how much fluff is behind medium format systems and how allot of it is just romanticised nonsense. Full frame has THE best lenses available.
The canon 24mm mkii is not that great at all, had it for years and find it frustrating with higher res sensors. I just road tested the 32mm HR rodenstock on an IQ150, comparing to my A7Riii with 24mm TS mkii. This is a proper comparison because they are nearly the same megapixels. The medium format absolutely blows my current setup away - way more shift, proper edge to edge sharpness and no chromatic aberration. It's awesome.
I compared different lenses and cameras to you but sure I'd love to see your results, please. Are you ok to provide a link with images where you compare both at equivalent apertures? It is important to mention that when adapting wide angle lenses to Sony you will see some softness on the edges this is not something you'll notice as much or at all on native bodies like the 5DSR. Finally, to make things completely fair, can you please perform a lens cast calibration on the Sony/Canon setup too, please.
The 24mm is v soft at the edges on native bodies too. Maximum 12mm shift is virtually unusable at the very edge. I'll do some test shots, with the Rodie stopped down by a stop or two to account for the longer lens equivalence, and get back to you. I'll see if I can LCC on the Sony but I've never done this before. I know that Capture 1 doesn't support it for Canon.
LCC should work fine on full frame cameras too I use it on Canon all the time, I made a whole video about that too :P. Thank you looking forward to seeing the results, much appreciated.
Fair enough. I've never seen the need to do it on dslrs, can't see a cast with any of the Canon TS line up. A few years ago the Capture One team told me it was pointless so I never bothered. I guess you just use it for light fall off. I'll check out your other vid!
I think Phase One is only the best with it's XF system and Schneider lenses. If you want resolution then buy a Vizelex Rhynocam and a sharp Hasselblad lens and then you will be happy with it, and much cheaper than medium format, or extreme resolution full frame.
Bálint Szász I haven't tried the Vizelex system I might have a look into that. Unfortunately all current Hasselblad lenses have very poor performance not a single one can outperform the best full frame lenses.
Sonder Creative The RhinoCam is just a 500$ slider kinda device, it lets you take 6 full frame images together (or 12 Aps-C, it works like that) and then you stich together them in Photoshop. And also, I meant Hasselblad V lenses, they are cheaper and some of them are much sharper then the new Hasselblads ;)
Oh yes I remember now I saw that and it can be used with a Sony A6000 series camera. I think you might be right about some of the older Hasselblad lenses as they were made by Zeiss. Personally, I would use a tilt shift lens and do a vertical panorama with that on a Canon 5DSR but your method sounds very interesting to me. Thank you for the suggestion.
The two crops shown (left Canon and right Phase One) are not from the same part of the image. I find that it somewhat weakens the demonstration. Very interesting angles of analysis, however.
Café Globulot well it doesn’t really. Angle of view is slightly different however the edge of a lens is still the edge. Also the sections where they’re slightly different are still bricks and you can see if something is sharper or softer.
At 12:05 the medium format had far less noise and was far sharper (while still a bigger image). Your comparison is deceptive....showing them both at 100% is not fair.....the medium format image is way bigger. You should have adjusted for equal size. You got a point with depth of field (f10 vs f16), but I have seen lot of results where medium format always won, and at 12:05 you see clearly the difference (even at a bigger image size, medium format won). Key is buy the best glass.
In this comparison the Canon lens is clearly better and even with the resolution advantage, the rodenstock lens can't keep up. Phase One however do have much better lenses available from Schneider so those tests may show different results but then those lense may not be suited for architecture.
Brian Cowan You’re welcome to download the images the link is still active. The comparison wasn’t about the resolution but about the lens. The lens itself isn’t as capable as the canon and that was the main point.
Thanks for the reply. I've heard that larger format camera lenses are not ground to the sharper standards of smaller formats because they depend on the larger format to make up for the difference on magnifications. Seems odd to me and i don't know how true it is.
brianncowan sounds like nonsense because if you shoot with some lenses for the GFX and even some of the new blue ring lenses from Phase One you see how sharp they actually are. The issue is medium format tends to rely on perception and they just don’t update their lenses as much this is especially true for Hasselblad. They’ve stuck with the older film days level of quality and rely on perception or excuses for poor performance. Shooting with a Zeiss Otus or Milvus lens you can clearly see the quality and don’t need excuses you can see how good they are. Why does medium format need to make so many excuses why isn’t the quality difference as easy to see. Sorry for the long reply lol.
you should use an ARCA RM3Di, a combo, or an Alpa... with Linhof, // isn't perfect all the time ! i keep on testing stuff all the time... and still can see much more difference than in your test !
David Giancatarina can you show some of your test results to me please I’d love to see. Also those cameras you mentioned don’t have any movements do they? It would be ineffective for architecture of they don’t.
Michael Horgan working towards getting hold of one but that's going to be less effective for architecture due to not having any native tilt shift type lenses.
yes canon tilt lenses are really good and the medium format does lack new lenses. The one who say old lenses are better are those old folks only based on feels
Side by side images should be same size. (even if different than 1:1 due to different MP size) Your sharpness comparison / analysis was therefore not identifiable.
Link to images at the same size. Canon at 300% and the Phase at 200%. The canon is still sharper. apalmanac.com/architecture/is-medium-format-the-best-for-architectural-photography-2530
There are a lot of truth in your presentation, but large sensors work on 16bits while all full frame 35 mm are limited to 14 bits, in practice means that the tones are more even and rich. It is true that medium format lenses don’t perform optically to the level of some Carl Zeiss or Sigma lenses manufactured for 35 mm. In most commercial assignments lenses are stop down a lot to gain depth of field in turn the resolving power suffers. Another advantage of having more pixels is that even if the lenses are unable to match the resolution of the sensor, it helps to render smother an more continuous tones, an example could be a simple line crossing at any angle that doesn’t align with the sensor square angle. Besides there is a psicológicas reason when you face a customer with a medium format equipment, you are considered a better and more profesional photographer in contrast to a 35 mm user. The moral is bigger is better!
If you Compare them on 100%, you are comparing two images with different sizes. In this comparison a Canon EOS 10D with 4 megapixels would be sharper than both of them. If you print both images at the same size - for example 120x80 centimetres - the result could be quite different.
warumbraucheichfüryoutubekommentareeinescheissgooglepluspagefragezeichen lenses determine sharpness not resolution. I’m not testing resolution. Diffraction is diffraction.
I agree pixel size has an effect on diffraction but to say it has nothing to do with the lens is completely false. If that were true then stopping your lens down would not increase diffraction.
That's why I said that you need to know pixel size AND aperture. Of course stopping down increases diffraction. Each camera has a "diffraction limited aparture" (DLA) that depends on the pixel size. If you stop down further than the DLA, the diffraction becomes visible.
I hope you realise that all you're doing is explaining why my results are what they are not disputing them. You also said it has NOTHING to do with the lens which is completely false.
You should not use Capture one for the Canon and Sony. It makes no sense. Even with that, your results stay. I love that this time you ask for help with the Phase One .now you are comparing at the different sizes, how will they look at the same size?
Jaime Duncan can’t use Canon and Sony images in capture one? Really??? Come on stop talking nonsense. Zoom factor was percentage of an image simple maths there I don’t think you understand relative zoom or relative distances. 200% zoom on one image is the exact same relative amount on any other image regardless of resolution. Also I’m NOT comparing resolution obviously 100mp is greater than 50mp, I compare lens performance which is very different. I don’t think you know what you’re on about.
I am not a native speaker, is a mistake, I will correct it. What I mean is "you should not " not "you can't". That does not make your comparison unfair for the Phase One, it just give it a little advantage. Using native for all, or standard for all will be fair. As you know Phase One has "secondary gain".
Normally you will want to compare the images at the same size: you either print the images (same size) or share the images (same size). One doesn't take images for pixel peep. I understand the aspect ratio makes the comparison a tinny bit more complicated but it's better. I hope this make you understand why the industry normalizes images sizes (from DXO to reviewers)
Jaime Duncan a percentage of something is still the same relative amount for anything else. 25% of anything is always a quarter of it regardless of how big or small it is. Both were viewed at the same percentage zoom. Also, again, this was not comparing resolution but lens performance.
While this may appear to be true when pixel pepping on your computer. When coming to real world prints for example 24x36 and 48x60 prints then you will realize comparisons is only factual at on a monitor and not I reality large scale printing. And I have use the Fuji medium format system and the sharpness of images can blow any full frame into there grave. I have never use or even seen a phase one in the Caribbean so I can't judge it but from my knowledge of Fuji medium format system and there technologies I will have to there image quality the edge over the canon. But great video and I always known phase one to be at the pinnacle of medium format if this video is free from bias that why I have not seen any in the Caribbean the result price is to high.
You may have a point but if an image is that soft and loses that much detail then I find it difficult to believe that it will be any better in print especially if no sharpening is added. Regarding Fujifilm, their lenses are incredible and some of the best I’ve ever used. They’re not quite as good as say Otus lenses or even the Zeiss 135mm f2 but they’re definitely up there. Once Fujifilm make some tilt shift lenses then we can do a proper comparison but as of right not it’s not the best system for architecture overall.
On your statement about slowing down your process not being worthwhile, it has more to do about maintaining quality of work. If you are concerned about being quick on delivery, that will not help your quality of work. I found your comparisons about lenses and medium format vs FF35mm to be flawed as well. Had you borrowed/rented a df+ or XF body with a 35mm Schneider lens, there would have been a significant difference to speak of. Your video should have been titled technical camera vs 35mm. If you are going to claim medium format, then use the system to give a proper comparison.
Robert Hold slowing down to think about the shot and making sure you maintain quality is a very different thing to your equipment slowing you down. They’re not the same thing and equipment slowing you down is a problem not a benefit. I always take my time when I shoot that’s how architecture is normally shot it’s not generally rushed but I don’t want something else dictating my pace I dictate it. Also why is it flawed making a claim is easy, justify it. This video was about architecture it’s in the title. XF bodies do not have any movements to accommodate requirements for architectural photography. If I were comparing to an XF I’d would have used a an Otus. That’s a completely different comparison then isn’t it?
Sonder Creative No. The problem with the comparison is that you used two systems that are not technical systems to a technical system. If you were comparing medium format systems to 35mm format systems, it would have been a better and far more reasoned to compare system to system. 35mm systems do not have a technical comparison and I cannot find your comparison in the video viable as a result.
I compared the two most viable options for FF and MF if you're specifically shooting architecture. You have an issue because I didn't compare the MF techno to a camera that doesn't exist? Come on Robert.
Canon 5DSR - bhpho.to/2fXcaM2
24mm Tilt Shift - bhpho.to/2BAof0c
Sony A7R II - bhpho.to/2CHl2LJ
Even though it's clear the 5DSR did well, you chose to give it a consistent advantage by not zooming further to get an actual 1:1 scale between the two images.
Imhotep397 I zoomed in 200% for both. A percentage value is the same relative amount. Also I wasn't comparing resolution but comparing lens capabilities between the two systems.
Thanks for watching my video and for the comment.
I've shot architecture exclusively for 15 years. I have an Arca Swiss w/100mp phase back w the Rodenstock 32, 40, and a sony a7rii w the 24tse. To claim the canon glass is superior is just plain wrong. Especially to come to this conclusion so decisively on one location test. You need at least an interior and an exterior shot to get a better idea of what these cameras can do.
I will admit 35mm has come along way, and depth of field is better. When your tech camera is properly tuned the Rodenstock will out resolve the canon any day.
But again I will admit with the new cameras the gap isn't what it used to be. I've been digging into this deeper lately. The a7rii is sharper than the 5dsr in my testing.
There is also something with the way medium format renders space that is really nice. But again that isn't so much a technical issue.
Michael Robinson this was not a one time one location test his has been a work in progress for more than a year.
Compare the diffraction for yourself at equivalent apertures and you'll see the same results.
Your video showed one scene and you seemed to make this judgment based on that one image from each camera, so one would assume you took one shot. I've got numerous images shot on both cameras and I can't tell you that the Canon is sharper because it isn't. Equivalent apertures mean nothing in the real world.. I only shoot at f/8 or f/11 whatever is best for the situation and the lens. Yes the Canon will have better depth of field. But the area that is in focus is sharper on the Rodenstock lenses 100% of the time.
If you put the 24mm on a 5DIII and take a shot, then put that same 24mm on the 5DSR and take the same shot the 5DIII will appear to be sharper. Now if you shrink that 5DSR file down to match the 22MP of the 5DIII the 5DSR file will appear sharper. So appearing sharper in Capture One is one thing but to truly judge these you need to view these files at the same resolution.
I do specifically mention how this video and the result wasn't a product of one comparison.
2:49
Arca Swiss w/100mp phase back w the Rodenstock 32, it gives you an invaluable quality superior to all the sensors in the world
Ive watched many of your videos on medium format and having shot canon for years and Nikon and now shooting only phase one you are not doing something right...? Im shooting an older P65+ back and it littlery blows any full frame available today paired with any lens out of the water. I think you are focusing too much on perceived sharpness vs actual sharpness / detail. The term is Acutance. Something no modern full frame can even come close to matching on even a 10 year old medium format back. Have you done any prints? This method also holds true with 120 film. Film can look very soft on screen at 100% but when you print it out its much clearer than any full frame digital cameras. This is the case with Digital Medium Format. I respect the work put into these videos but pixel peeping on a monitor means nothing. Print the work out and then you will see the truth.
Andrew Thompson thank you for the comment and I really appreciate you watching my videos.
I've only ever used and tested medium format digital and I can very confidently say there isn't a single medium format lens that can outperform the best available from full frame. I can prove my claims and I demonstrate much of it in my videos. Consider lenses like Zeiss milvus, Otus, Sigma Art, latest canon, Nikon and Sony primes. There are allot of better lenses available from full frame.
Your points around sharpness and how well your P65 performs, if you can demonstrate and prove your claims then I'd be very interested.
Andrew Thompson also a quick note yes I do print images for my clients and generally print pretty big.
This is one image that was shot on a 5DSR then split into three floor to ceiling panels. This is only one from the series.
You can see from the link below I shot the printed image with an iPhone so allot of the detail is lost and doesn't look quite as good as when you're up close to it but you get the idea.
1drv.ms/f/s!AqMU3QFCgyGIoKF_b2IBwZ_N3e37wg
Well said Andrew!
Andrew Thompson well said. And I'll have that back off you when you're done!
Well, the files I scanned from the 4 x 5 film by the Epson V800 look sharper than your prints. I do agree with Andrew Thomson, I've seen quite a bit of digital prints in the National Gallery and they didn't impress me. My 12 x 16in prints look way sharper and very smooth with no grain whatsoever.
You need to zoom in to the same level on each. The closer zoom makes the images on the right look softer and less sharp. Cant really compare accurately unless they are zoomed in equally. Images always look softer and softer the more you zoom in. Thanks!
The zoom was the same, it was 200% for both. A percentage of something is always the same relative amount of something else. For example, 25% of something is always a quarter of whatever the whole figure is. On that basis the comparison is fair.
Also, we're not comparing resolution, it's very clear that 100mp is more than 50mp. Diffraction and lens performance was being compared and the differences are clear, the Canon is better.
Thank you for the comment, Richard, appreciate it.
Sorry that is just illogical. The only fair comparison is for equal reproduction sizes, whether on screen or in print.
Sonder Creative That's not what he meant. To do a proper comparison you have to normalize the files. Which means either upscale the 5DSR image to match the size of the Phase One or bring the Phase One to the size of the 5D file.
That's the only way to properly compare different sensors.
Hey, that was an interesting video, but for the sharpness part, is it possible to see a version of the PhaseOne photos sharpened and down sampled? For sure it was softer and a lot of those megapixels were lost, but I think it would be more fair to down sample it to 50MP (or 42MP) to see if there's any gain in sharpness overall. When comparing noise performance between my A7RII and an A7S, I down sampled all the A7RII files to 12MP to see the differences between the two.
Why do you zoom more on the 100mp back? This is definitely not the same framing as you had on Canon.
Tomasz Wachla I zoomed in 200% for both. A percentage on one is the relative same distance for another. Also we already know the Phase one has more resolution we're looking at lens performance not resolution differences.
I am curious which Linhof Technical camera was used? The 679CS? The Techno Field Camera? Or the Technikardan? When I first got started in the technical camera systems I was using the Technikardan with a kapture group sliding back adapter and a Leaf Aptus 22, with various lenses, but the main one being the 35mm Schneider Digitar. The tolerances on these lenses is SO SMALL that it's nearly impossible to achieve critical focus with any of these front/rear standard cameras, which is why you don't see any professional architectural photographer using them, at least I never have.
I moved to a Canon system with the TSE lenses for a while, but the quality was nagging at me. I finally researched and found the technical cameras that come with the helical focus mounts for 4x5 digital lenses. Alpa, Arca Swiss, and Cambo. I setteld on the Cambo WRS 1250, a phase IQ260 and the same lens you used in this test, the Rodenstock 40mm. I've had this setup for many years now and it's results still blow my socks off, I shoot it at f5.6 or f8 and occasionally have to do multiple focal zones to stack later, the results from this camera setup are astounding, and Ive heard even better things about the Alpa once you've shimmed them correctly. I also have a 5ds and just got the a7riii, but those still can't hold a candle to the Phase.
I appreciate the thorough review, but fear that that linhof might get you close to focus but not quite there, I had many shoot where I had the same.
seeing now that, a phase one specialist, setup and took the shots. I also had someone setting up my initial camera system on the Technikardan back in the day, which performed terribly, I do not recommend any of the Linhof's under any circumstances, Phase should fire whoever setup the camera. That all being said, I am very intently watching all the new cool APS-C and FF options coming out and hoping one day I can sell my expensive system and get a light and more affordable system.
Architects are more interested in composition than the IQ at 100% pixel peeping, I've found. Especially now when most images are on Instagram and viewed on mobile devices and the occasional magazine, the Canon system is more than adequate for quality.
I do some work as a first assistant for a high profile architectural and commercial photographer in Melbourne, Australia. On our shoots we will often use the Phase One IQ4 150MP with the blue ring Schneider optics (the best of these is probably the 80mm), but we also use a Canon 5DSR or Sony A7Riii for smaller shoots or personal projects.
I can say with absolute confidence that none of the Full Frame cameras that I have used are as good as the Phase One. @SonderCreative I can understand how you may have come to this conclusion using the technical camera that you did, as this can take a lot of experience to correctly calibrate and set up for the sharpest possible images. This is one of the reasons that we use the Phase One XF system, along with the Schneider lenses, as they are all perfectly calibrated and provide by far the best resolved and highest quality image.
When the images go to print, the print technicians at the marketing agencies that we work with will almost without fail be able to notice when we have used the the smaller sensor sizes of the Canon or Sony. Image quality and detail is more than just about the megapixels, it has a lot to do with the size of each pixel on the sensor, and it’s ability to capture light.
In many of your comments you talk about conducting controlled tests. How can you do this with one system that has been calibrated by someone I assume to be an amateur (you) in using medium format technical cameras, and another that has been expertly calibrated in factory for each lens and sensor combination.
I urge you to use a fully integrated system like the Phase One XF, if you really want to compare apples to apples with a fully integrated Canon or Sony full frame system.
Blue ring lenses are fantastic because they've been updated. The video was more about architectural photography and unfortunately, those type of lenses aren't really geared towards that kind of work.
Having said that Phase One have now produced more up to date lenses for architectural photographers so it could be interesting to see how they perform.
There is also the fact that lenses like Otus lenses don't exists for medium format so if we're discussing optics from an overall perspective, full-frame still has the advantage.
The problem is that full-frame lenses are limited by their cameras and for that reason, overall Phase One will produce better images in most comparisons.
This specific comparison I conducted was not about the full Phase One system but more about a particular niche area.
@@Sondercreative Phase One have also released the XT technical camera system that is a fully integrated system like the XF. It's what we have had they joy of using for most of our architectural shoots for the few months that it was out before our city got locked-down sue to coronavirus. Before that we used a Cambo Wide RS system that was also very well calibrated, and I still feel that that produced much better images than our Canon 5DSR with the 24mm TS.
It would be quite an interesting comparison video for you to do if you were to compare the new Phase One XT technical camera system (I think they call it a 'field camera') and the Canon equivalent.
I did think that your videos are very informative and impressive, however I would just caution you to make sure that you're comparing apples to apples before you go and provide the very wide claim that full frame produces better results than medium format.
Hope you're staying safe and well during this whole pandemic.
Charles Beggs yes I meant the XT, I’d love to review and compare that so I’ll see if I can get my hands on a system.
Thank you very much for the comment and the kind words, really appreciate it.
For which format was the 40 mm Rodenstock lens designed?
I am considering switching from canon to Sony, to the ARiii, and using the canon TS lenses still on the Sony, with the adapter, what are your opinions on this? Thanks!
Luke Lozano I don’t personally recommend that switch. Your tilt shift lenses will be softer in the corners and also in terms of sheer usability Sony is a bit behind.
In any case try it first see if you like the system and then decide for yourself. Don’t just go with my recommendation or someone else’s. You need to love or hate it yourself.
yes, you are right about d.o.f., but THIS is why these cameras have an Automated Exposure averaging and Focus stacking features, so that any problems with this can be sorted out, ie the focus stack ensures focus, by automatically merging images in a series of shots, giving a perfectly focused image; the exposure averaging does the same, but for noise, and diffraction, allowing you to shoot at the best aperture, NO sensor noise ( no need for filters, or long exposure times, as the automation takes 50 or so shots at a faster speed, no hot pixels).
I don't recall those features being available in the IQ3 at the time of recording this video. I'm aware they're available now in the IQ4 and they're great.
Having said that, you can focus stack and merge images quite effectively with pretty much any camera. Although having it built in as a feature is useful.
Thank you very much for the comparison. What are your thoughts on using the 28mm Otus as to the 24mm Cannon not taking shift into account.
You're welcome, thank you for watching.
Hmm, so not including shifts, then wide open (unfair comparison) the tilt shift wins. Having said that at f8 which is more useful for allot of landscape and architecture, the Otus is sharper but the difference is not massive.
At f16 the diffraction is a little worse in the Otus so they are about the same in performance. Both are definitely better than the Rodenstock :).
The images are in different scales?!
Kp Mak different scales as in resolution?
The canon and Sony aren't 100mp that's why and I'm comparing each at their respective best.
Sonder Creative as in respective image area on screen, so 1:1.
KidAeon yes but the zoom factor is based on a percentage therefore relatively speaking it's the same.
100% is still 100% of a given subject and so on.
A problem is that while you can count lines per mm from a scale, apparent sharpness and DR are visual impressions, not metrics per so. So for comparison of quality as viewed, two images should be viewed at the same apparent size, not the same % magnification from sensor output. That's my understanding, open to being corrected.
Did I ask if your system displays true resolution at all magnifications? My older version of Photoshop only does at 50 or 100%, the rest are interpolated.
@@Sondercreative I respectfully disagree. It s like comparing a 4x6 print to an 8x10 print. So you enlarge the 4x6 to a 8x12 and the 8x10 to a 16x20? The objects in the image must be the same size. Also your test catered to the Canon's smaller sensor which I'm not complaining because this is my setup and I'm very pleased with the results. I say this because you are comparing the "corners" of a round room. Do you realize the substantial increase in depth of field ratio? I seriously believe it would have been a different story if the wall was flat. The corners would be so much farther from the sensor of the Phase One
The 5DsR stills holds up today. Solid camera and my camera of choice.
Canon have always done well with highlight recovery for FF. I also agree Usman, recovered shadows rarely look good with out a fair bit of work done after.
im gonna need you to zoom in the same in both images cause if you tell me it loks sharper when its x¿zoomed into the full wall vs a single brick, well its not easy to compare
LeoVomend thanks for the comment. I recommend you download the images because unfortunately it won't be possible for me to update the videos. You'll also be able to see more clearly if you download and compare yourself.
Hope that helps.
Might I suggest to compare pictures at the same printsize. Either blow up the 5dsr or reduce the 100mp to the size of the 5dsr. At least for me that would have made more sense for a sharpness comparison of the two cameras and to point out differences between both.
Michael Schwarz one of the key points of my conclusion was to point out the fact that the lenses for medium format are ineffective and regardless of how many pixels or how good they make their sensors the bottleneck is the lens. Had I adjusted the exposures that point would have been less effective.
Thank you for watching and I appreciate you leaving a comment.
Bill Nichols did a similar test but he was using a phase one 40 mp back and his final result showed that the phase one results were much better than the 50mp canon so how do you justify your results even though you were using a phase one 100MP
Andrew Porfyri crop factor and reach. Greater resolution in a smaller section of the frame.
I wonder if the front standard of the Linhof was out of square ? Years ago when I was shooting 5x4 super wide lens like the Schneider 58XL were tricky to use because the back focus distance was so short. Anything out of alignment on the camera standard and track would cause softness.
Also the focus screen depth had to be properly shimmed and checked with a dial caliper or what was in focus on the screen was not the same as on the image recording plane.
Also the working aperture of f11-16 meant you had to watch for camera shake outside and shooting from building balconies etc.
Finally, view camera wide angle lens are not flat plane designs. The point of focus in the middle of the frame is different to that at the edge due to a convex image plane. Wide angles designed for SLRs are complex inverted telephoto designs to clear the mirror and often have field flattener rear optics.
I forgot something important: if the technical camera was not properly adjusted for the phase one back, using a shim kit, your images will look much worse than they could have. Ask anyone who has ever used an Alpa camera or something similar. The difference is like night and day.
It's very likely that the used linhof doesn't provide the extreme mechanical precision needed for optimal results.
burak uestuen as mentioned in the video the camera was set up and used by a Phase one specialist so that argument doesn't work. Yes it is compatible and yes they do perform at the highest level when not shifted in any direction.
For the images in the comparison the camera was not shifted in any direction.
Sonder Creative , there is a good reason why phase one is working together with alpa not linhof. I'm 100% sure, that if this test would have been done with the right gear and the proper knowledge, the results would have been totally different. To me it's obvious, that you are not the right person for this test. No wonder you come to wrong conclusions. When you have more experience in these matters, you will understand it. Don't believe me, ask Ming thein. He's famous for his knowledge of camera technology.
A Phase one specialist did this comparison with me I didn't take the shots. Personal attacks, the last resort for someone who doesn't have a valid point or no evidence for their claims.
Also you're missing one fundamental point of this comparison and that it's dedicated to architectural photography. The point remains that regardless of what excuse you want to make the full frame system is better with the lens performing better. That's the bottom line and that's the practical side of things. Stop making excuses for the poor performance
Burak, Sonder Creative has mentioned multiple times that he did NOT take the P1 photos; a Phase One specialist took them. Did you not hear or read that? Look up cognitive dissonance. I am not a psychologist, but it might just apply in your case.
I had the same issue when I first started converting my 4x5 architectural workflow over to digital and a salesman at digital transitions in NYC setup this linhof with sliding back system that was just useless, I didn't bite then and finally through my own research wound up with a Cambo system that is calibrated so perfectly, it's as easy as shooting with a point and shoot, and I am getting phenomenal results. I hope one day that everything catches up to the quality needed for big prints, but not seeing it yet, unfortunately. I was knee deep in Cambo when I went and demo'd an alpa system they had shimmed up to perfection and it was mindblowing, was too committed to Cambo to jump ship though.
It must be some fundamental wrong with the comparison. I have a Sony A7R (36Mpix) and a Phase One XF with IQ180 (80Mpix) and it is a huge deferens in resolution and sharpness, if you will call it so, between Sony-files and PhaseOne-files. Phase One have the double resolution and it is impossible not to see that if you compare pictures.
But with 80 or 100 Mpix you must focus exactly, hold the camera still - Phase One have a built in seismometer to control that there is no vibrations - and you must have a lens that can perform resolution for 100 Mpix.
It would be interesting if you do this test again with a Schneider Kreuznach 35mm LS f/3.5 lens or 45mm LS f/3.5 lens on an XF boy. Or, use an Phase One A-series IQ3 100MP System (made for technical cameras) with Alpa body and a ALPAR f/4.0 35mm - made for 100 Mpix.
Joa Silver there's a significant difference between the original A7R and the current one and the 5DSR when it comes to image quality. The original A7R had some major flaws with the shutter slap and caused allot of issue with image quality.
Try it for yourself with a 5DSR and more importantly, the latest 24mm tilt shift from canon.
Also part 3 will go further into Phase One lenses that are rated up to and even beyond 100mp.
Joe is correct. I have been doing a ton of research on technical cameras with digital backs. It comes down to the fact that the digital sensor is very sensitive to the angle that light hits it. All camera manufacturers have addressed this in various ways, but largely through retrofocus design, which moves the lens further from the sensor and allows light to come in at a straighter angle. The digital mf backs require a camera like the XF to truly take advantage of the sensor, for this very reason. This also applies to full frame digital sensors and angle of light coming in from the lens. This is why super wide rangefinder lenses perform so poorly on the Sony A7 series.
So yes, your comparison exposed the flaw of the technical camera + medium format back, however you gloss over the specific setup you use and dismiss medium format for architecture in general. the Phase One backs would substantially outperform the full frame Canon or Sony in a more apples to apples comparison. What I mean by that is the tilt shift lens was essentially at a proper distance from the sensor, while the mf lens was not. Also, as Joe pointed out, the Phase One A series would have given you better results. It doesn't have the tilt motions you desire, but the body is shimmed to have sub millimeter control of the distance between the lens and the back, to get as much out of these lenses as possible. It also, apparently, helps to use lenses with wider image circles, beyond what the sensor actually requires.
If you can't live without tilt/shift, then keep doing what you're doing. But don't lead everyone on the internet who hasn't done the research to think medium format lenses are crap, they are certainly not.
Chris Duesing this whole review was specific for architectural photography. Tilt shift lenses are almost a requirement for that. Consider the context before making your statements.
In general the sensors on medium format are better but the lenses from full frame are better.
If you’re not going to be using a camera with movements for this comparison for medium format then I would use the Otus lens instead which is significantly beyond both of these lenses.
You have learned to make very authoritative and educated sounding videos, its too bad they aren't based in facts. When that is pointed out in comments you ignore basically every detail of the counter argument.
I suspect that is why your video is creating so much controversy. You have managed to make a very interesting and correct point and yet at the same time draw some very wrong general conclusions. Nothing about your analysis reflects on the quality of medium format lenses. It is the shortcoming of digital sensors that leads to soft corners. You ignored that counter argument and spoke down to me in generalities.
Chris Duesing Tone of voice doesn’t get passed through text form and I only talk about the points and the evidence I present. I’m being specific and to the point, please don’t take that as condescension or me trying to be rude. The individual must always be respected but I don’t need to respect your opinions or ideas if you don’t present evidence.
You’ve only made claims, claims are easy, anyone can do that.
Your point about digital sensors having limitations and soft corners is strange. You do realise both are digital sensors that I’m comparing right?
Perhaps the Rodenstock is not that well suited for the Phase One back. And as some people point out, there might be some lens correction in place with the full-frame examples (I don't know how Capture One is handling this). Sharpness is not always the main signifier, and sometimes you can get more data to work on out of a less "sharp" lens - applying algorithms to a seemingly "softer" image might produce a superior result in the end.
But under practical considerations I would also go with a full-frame digital camera (the Sony A7RII in this case) and be fine …
Christian Augustin you make some very good points and I agree to some extent went.
Two things, for architectural photography this is the system to use so if it's not compatible then canon is still better.
If we're comparing just lenses there is also the Zeiss 28mm Otus which is much sharper than the canon. I'd say the Zeiss is significantly better than the Rodenstock lens. Ultimately full frame still has better lenses available.
Thank you for the comment.
I think the Rodenstock was originally made for large format photography on film (it looks like such a beast), and there it might’ve been actually a great lens. But when it comes to specialized lenses with extrem high resolution - current full-frame lenses have the advantage of a broader audience (and therefor get more love than large-format lenses).
Christian Augustin I agree I think it was for large format and yes more love for full frame which also has meant more research and investment.
Hi !
could you make video phase one fx vs canon 5dsr?
I shoot architecture on the A7r with Canon glass. I have have a Canon 5dII and a Phase P20. I can tell you that the 2004 Phase beats the other cameras hands down. Hands down. For me it's the colour and the graceful transition between tones. If I had the MF lens options, I'd use it in my work without hesitation.
Also, why do you have only one LCC capture on screen? It needs to be done for every body/lens/aperture/ISO/shift combo.
Joseph Asghar you're comparing it to the original A7R which was known to be notoriously bad. I'm comparing two very different cameras, you can't use your experience with another couple of cameras as a way of determining my findings, that's illogical.
The shutter shake on the A7R was really bad and even on DXO mark the same lenses that perform amazingly well on the mark II perform really bad on the original A7R.
The same LCC was sufficient in clearing up all of the images however I do agree with you about that point.
Sonder Creative The difference between the two Sony cameras is 5% of linear resolution. Shutter shake is irrelevant to my work, on a gitzo with a manfrotto 410.
Your misunderstanding of the purpose and workings of LCC calibration negates your review - it fundamentally affects the image if not specific to the image.
Joseph Asghar the LCC was sufficient, is has absolutely no impact on resolution and detail.
Also the difference between both cameras is not 5% look it up.
Try comparing the cameras I did and then we can talk until then you don't have a valid point.
And you would be very wise to do so Joseph. Here's an article I wrote about the journey from 35mm digital to medium format a few years back: blog.cambo.com/2012/06/28/a-new-camera-the-odyssey-from-35mm-to-medium-format-digital-photography-by-richard-osbourne/
Thank you Usman for another great video. I actually watched for a 3rd time, but this time I desided to actually download the images you generously provided. I must say that I notice that the focal plane of the two cameras-lenses doesn't match. Suprisingly, is the Canon which seems to have more areas out of focus (most of the right part of the photo, look at "Yellow Sunrise" sign, and the upper part of the left part as well. Maybe you had a slight tilt on the lesn?).
I did what the other comments suggested and match the resolution of the photos, and I do mostly agree with you fundings, in areas where both cameras are in focus the canon seems more contrast with almost the same resolution but the phase one seems more detailed due to more MPix. But it is a bit hard to judge since you cannon be sure if this is due to the distance from the focus plane or the performance of camera-lens.
I understand that it is difficult to repear the comparison but if you will have the change could I suggest to take photos of boring distant brick walls? Maybe include the GFX 100 as well? (I know you have another video but test shoots like that are easier to intepreted).
Thank you by the way of showing the interior of the corn exchange, it look amazing. I didn't managed to visit it when I was in Leeds.
One thing that may not have been pointed out is that the rode stock lens may actually have a calibration issue (which sort of happens and is a major down side to these lenses) they're supposed to have complete edge to edge sharpness and if they don't it's usually to do with a calibration issue
Generally yes and this was something that was considered. However the Lens came directly from Phase One and was calibrated by them before any demonstrations.
@@Sondercreative it is entirely possible that the lens itself wasn't calibrated properly by them - Since Cambo/rodenstock usually do it in house, or at least they do now - maybe to stop this issue that you had?
I have Rodenstock being sent away for repair because of a similar fault this one is having, some spotting of soft edging, according to their standards it's not supposed to be like that.
@@jamessummerfield5102 it is possible I agree.
I tried to minimise any issues by completing the test with Phase One twice on two separate occasions. I was assured on both occasions that everything was working to the best of its ability.
Then again mistakes can happen, but I think its unlikely considering the due diligence.
@@Sondercreative Not that I think it's important anymore since this video you've posted has covered all tracks regarding, something is definitely off with that rodenstock lens.
Once I get mine back from repair I'll do a little test in similar fashion - I have an A7Riii so I can do a direct comparison.
Maybe that revision of Rodenstock lenses were just garbage and overpriced haha
One thing this video didn't really look into was....... Composition and framing.
The greatest difference between the formats aren't just the technical specs like what this video suggest. There is a much greater difference in how a photographer approaches a scene with a medium format vs full frame.
I think there needs to be an analysis on how the framing on each format affects the way in which the images get composed.
For me image quality is way more than perceived sharpness, colour rendition tonal transition and all those nuances of hues is where it's at. This is a very interesting video nevertherless.
Thank you Ferhan, I'm actually a fan of your videos, I've been watching and following your channel for some time.
I agree with you about colour rendition and so on. I've done comparisons between the Canon 5DSR and the Hasselblad and when you use some of the top end Zeiss lenses the 5DSR actually matches the Hasselblad in colours and tones. Not only that, the best full frame lenses are much much sharper than any of the Hasselblad lenses.
Having said that Phase One might be different.
Sonder Creative I'm surprised! There is a video knocking about on UA-cam where the SR is pitted against the A7R2 and a phase one. The A7 seemed to out perform the lot. I think it's safe to say all of these cameras produce amazing results, and one camera or lens won't allow you to do anything more than any other, unless of course one views photographs at 400% regularly lol. Thanks for the kind words regarding my channel, need to get more content out, too lazy lol
Ferhan Khan Photography it's amazing how well frame cameras perform these days. Medium format is becoming less and less viable.
Slightly confused, from the files in the description, (CF007362.tif) and (5DSR0638.tif) from the IQ250 file, I can read the menu on the other side of the room (Little leeds beerhouse, at 300%, whereas it's impossible to read from the 5D.. Also, some of the IQ images look slightly soft on the lefthand quadrant of the image, possibly due to the use of the Linhof - it's known for having sharpness issues (several people I know have got rid, due to it going soft as it's very difficult to keep the lens 100% parallel to the sensor with a bellows, camera - you can get away with the tolerance on film, but not on digital) that said the 24 TSE is probably the sharpest wideangle on DSLR, but I think you have some issues with the Rodenstock/linhof combo, (regardless of who took the images)
Hi Tim thank you for the comment. Could you please clarify what you mean about the files in the description part of your comment. I'm not sure I full understand.
I think you could be right about the linhoff Rodenstock combination, however the centre sharpness remains unaffected especially because we didn't shift the lens. Also this is still the best combination for architectural photography when using medium format, so comparing things specific for that.
Thank you for the comment you made some very good points.
I downloaded the files (CF007362.tif) and (5DSR0638.tif) and looked at them, in the centre of the frame, the IQ250 file looks significantly sharper, if you look at the menu for Little leeds beerhouse, on the other side of the room, as in I can read almost 100% of the menu, compared to the 5D file - where I can't read any of it - it seems to me that the Rodenstock is significantly sharper in the centre, (as I can either read the menu, or I can't, which makes it pretty definitive that the Rodenstock/250 is sharper)
Tim Gregory There's a video relating to that but it's only for my Patreon subscribers. In that I explain the detail differences due to the sensor sizes. The 50 back is a smaller sensor so the focal length equivalent is different.
I should actually move that folder because it's only for my patreon.
What confuses me slightly, is that you're saying the Canon is as good as the Rodenstock IQ3, or slightly better, considering the Rodenstock 40 has a 35mm equivalent of 27mm on the IQ3100 it's pretty close to the 24 TSE in focal length, but it blows the 24 TSE away. For example, if you look at the file from the IQ3100 (CF001185.tif) and the file from the 5DSR (5DSR0638.tif) put both images to 300% and zoom to the bottom right quadrant, to the "Yellow Sunrise" shop, the writing around the outside of the logo is just an unreadable blur on the Canon, whereas on the IQ3100 I can quite clearly read "global inspirations, Traded Fairly"
There is still a difference in resolution and the Rodenstock has the advantage of having a 100mp sensor behind it. Detail from resolution is, of course, different from lens sharpness.
This is what I'm seeing right now looking at the images, not in any RAW processor and the Canon 24mm is noticeably sharper on the edges and the centre sharpness the Canon is a tiny bit better. Again the resolution difference is there too.
1drv.ms/f/s!AqMU3QFCgyGIoKEejqtTZfvdfcNMFg
I would suggest for architecture the camera with a lot of megapixels and best OIS/IBIS possibly, so you can stop down the aperture, put the iso as low possibly and shoot with long exposure times, since the subject is still, and you actually want to get rid of the dof. So I think Sony has a big advantage here over Canon and Phase One which comes with OIS only.
By the way, noise is determined by dot size on senor (photodiode) per pixel. So since Phase One is 100mp I bet it has considerably small dot size or simply nothing out from the ordinary CMOS sensors produced for many years, even if you consider the sensor size overall. Also, Sony's sensor is backlit illuminated, so the sensitivity to capture the photons is probably much higher over latter two cameras.
Not without a reason smartphone manufacturers came back to 12mp, even if older models had already 16 or even 20; it's to make a bigger dot per pixel, capture more light and automatically create image with less noise with less processing power needed for noise reduction algorithms.
Hi John thanks for the comment,
Pixel size on the Phase is actually bigger than the 5DSR and the A7RII but even with that, the Sony is cleaner in the shadows.
For architecture, OIS wouldn't really be an advantage, because most if not all of the people I know that shoot this professionally almost always shoot on a tripod. Shooting on a tripod is far more effective not just because of being able to long expose but also because you can take a number of shots and layer them together without much of an issue. Also, a tripod can be much more effective to compose accurately.
Well, that might be true indeed, I didn't search up the dot size of the phase one.
Well, for the tripod, sure. But not in every situation you can use it, and good image stabilization in the camera is always a bonus. I'm not trying to be a Sony fanboy or anything, so don't get me wrong. I used for some time Olympus OMD, and I think that would be a great camera for architecture too with the extra resolution mode and IBIS. Nikon 810 would be great candidate too with the 36MP. And obviously Canon 5DSR.
In my opinion medium formats are much better choice for cinematography or macro shooting, where sharpness on edges doesn't matter much and where is actually that super duper dof what you want!
I forgot to mention in previous post that you need wide angle lenses for architecture which are again hard to find even for FF, figures for medium formats, unless you want to spend hours for image stitching..
I personally think that if you are shooting architecture professionally you should be shooting on a tripod. Tilt shift lenses and Rodenstock lenses which are specifically for architectural photography are all manual and I find it difficult to imagine someone using them without a tripod.
Also, I'm a big fan of Sony, I really don't mind you mentioning it. As for the Olympus, I haven't had a chance to use but I've heard some really good things about it. Check out my Hasselblad review where I compare macro lenses and a couple others and the Full frame was still better.
and yes wide angle is advisable lol.
A bit weird at the beginning, but agree that canon tilt rocks and softness on the conner have nothing to do with a sensor it’s only optic lens issue so kinda weird to hearing that “phaseone” doing it vs canon tilt - comparing sensor and lens.
Sorry I don’t understand what you mean.
With what are you shooting the vlog ? is damn sharp the footage 😃
If I remember correctly I think that was the a7R II
@@SondercreativeNice quality 👍😃
Have you tried a Hasselblad camera in multi-shot form and who underexposes by 3 stops?
peter brady haven’t tried the multi shot but it’s too niche and the sensor is much smaller compared to the 100mp back. I’d have to use the 32mm which isn’t as great as the 40mm.
The underexposed shots were only to test dynamic range.
The Phase One sensor is Sony. Its pixel pitch is slightly larger than the a7rii's. The sensor is a factor in the end result but the motherboard, it's components more so and firmware even more. Statements in the video or comments below ... meh
There is also the possibility of out of focus or a bad copy of the lens in this case the one used on the large format sensor.
Possibly but I doubt it because the lenses are regularly checked and calibrated. Focus was also checked multiple times prior to each individual shot so it’s unlikely.
Sonder Creative
Who knows, but it is not normal. Thanks for your reply.
Hi Sonder, great informative video with proof. I specialise in architectural photography but it's more of a hobby of mine than a business. I was about to sell my canon 5dsr with all my canon lenses and my canon 5d mk 4 to purchase the phase one system. After seeing your video I think I will stick to canon until they improve thee lenses for phase one. Regarding the comments here who disagree with well I think psychologically they can't accept your results as they have invested in medium format system. So it's hard to agree with you once you have invested £60k on a phase one system with lenses. Thank you for your honest, transparent comparisons backed up with proof.
Thank you for watching and for the comment, what you said means allot to me. I was actually looking to sell a bunch of my equipment and make the leap to a Phase One system too. I only made these videos because of my findings from comparing the systems and realising how they're not really the best but offer worse image quality in many areas. I agree I think that had I have spent that much money I'd be less inclined to believe these findings too.
I'm very happy to know this video helped you make your decision. I'm certain with the best lenses from full frame you will get much better results.
Man! Thank you so much for this video! I was about to make an investment in Medium Format just because "everyone says they are better". I never saw a review with that much depth and files being compared side by side! I'm sticking with my Canon bodies and their range of incredible lenses.
Thanks for the comparison. Answered a lot of questions I had about both formats. Time to invest in a TS Canon. :-)
Having your comparison and discussion reveals valuable questions. Thank you!
Thank you for watching ReNaldo, I hope my video helped.
Your comparison is nonsense. It's known that canon, Sony etc. strongly precook their raw files. Apply some sharpening, contrast etc. to your phase one files and they will look much better than any FF file. Either you are biased or you have too little experience. I've been working with the best FF like Nikon d810 etc. and phase one backs for years, also with canon and the ts-e II 24 mm, you used for your test.
I have a lot of experience with both formats. Your conclusion is simply wrong. You're right about ease of use. But if you want the best image quality possible, you have to work harder and use medium format gear. It's been the same in film days, the gear was even bigger than. A 8x10 inch film camera still delivers the best IQ on the planet, if you know how to use it!
burak uestuen you haven't compared them properly then it seems.
All sharpening and anything extra was pre removed. Also sharpening does not bring back detail it only gives the illusion. The heavy diffraction in the Rodenstock lens cannot be adjusted by simply adding some sharpening.
Full frame is better this was done with the help from Phase One specialists and over a long period of time going back to them to ensure I was getting the absolute most out of the Phase.
Excuses for bad performance.
Sonder Creative , you can not remove all sharpening and other precooking of canon, Sony, Nikon... raw files with Capture One or Lightroom.
That's a fact. The rodenstock people clearly say at what aperture the lens should be used. If you don't listen to them, it's your own fault your images look like garbage. Did you even listen to the Pros that helped you? I doubt it.
burak uestuen So do Phase one and Hasselblad.
Again sharpening and cooking images cannot overcome diffraction or loss of detail. That's a fact. The lenses for medium format are terrible, still old design lenses that can't compete against full frame. Consider the budgets for medium format they don't have the money available to invest in better lenses.
Medium format lenses are horrible in performance they have more diffraction at equivalent aperture and also cannot produce shallower depth of field at the wider end either. No advantage in any area.
burak uestuen The rodenstock was shot at various apertures and all of them it couldn't perform better than full frame and was worse in every test. The recommended apertures are between f8 and f11 and at those apertures you can't get that deep DOF and also the lens still performed worse than full frame at those apertures.
Look I get it it's difficult to accept, this is why I had to keep going back to phase cause I couldn't believe the results the either.
Sonder Creative , believe whatever you want, but if you are ready to overcome your narrow view, talk to professional photographers about this.
Maybe they can change your mind? I'm sure they could show you some printed work, that will blow you away.
I am probably opening myself to attack, but something doesn't make sense about saying the Canon is better. The plane of focus is supposed to be a *plane* , but the scene is clearly not a flat plane. If the Canon is sharper at the edges when opened up, isn't that just a sign of extreme field curvature? The edge of the image of a curved scene *should* be soft, shouldn't it?
eric moss not opening you’re self to attack at all discussions are welcome.
Both cameras are relatively softer on one side or the other. So the Phase is sharper on the right and the Canon is sharper on the left. I only noticed this after someone pointed it out.
Having said that the canon is still noticeably sharper in the centre at equivalent apertures.
Best guesses: You'd use the Linhof med format camera you happen to have for the more extensive camera movements for perspective correction, and you can get many focal lengths and still have those movements (I was not aware Rodenstock was the best lens brand for med format). The highlight recovery is more important than shadow recover for architecture. And you can demand more money and fend off high end customers better with the med format.
No, none of those reasons is a good enough argument, and you never even got into editing for perfection in Photoshop with the FF files. That defense offered, I happen to own that 24TS, and so this vid is a pleasure to see. I just sold my 5D2 on eBay, and so now I suppose I get a new Sony and anticipate a super Canon 5DsR redo next fall.
John Smith the extra movements are only really an advantage on film. On digital the colour shifts really start to impact the image of you shift to much and it essentially degrades the image.
Shadows vs highlights is a preference. Highlights don’t I introduce noise but shadows do so I’d say that’s more important.
In regards to Rodenstock being the best, thats what I’ve been told by several professionals in this market but I guess it may be subjective.
Thank you for the comment and good luck with the new cameras.
I disagree, as I don't associate camera movement with color shift, nor do I associate DOF with focal length (it's an effect related to "circles of confusion" size in any format size for given field of view. And I differ on diffraction as anything other than a relation to the absolute size of an aperture).
I made a post because I very much appreciate your presentations, and I will when I can have a bit of dialogue, not to smack back. On that point, I can't catch your name, which I would like to know as background info, as I intended to view your presentations as they are published. My go-to guy online is Tony Northrup, and he sets a very high bar indeed. You are very complimentary to his work.
I'm old, so I do know some things, but like all camera science types I can never get enough of solid presentations. On that point I wonder if you have looked at Kasson.com? It's a blog by a gentleman out West USA who really does some heavy hitting data based science with a variety of products and issues. When I read pg 61 of the A&R2 manual and learned that steady shot will not work on a tripod I was bummed, but Kasson did the science that proved to me not only does it not work, it harms sharpness. His data work is very impressive.
Sorry for the length here, I type fast. I welcome being shown to be wrong, that's how a person learns. I am very opinionated and passionate about some aspects of art, including photography so don't mind my intensity. I'm in baby, I subscribe.
John Smith focal length does impact DOF when shooting two lenses at any given distance, aperture and pixel size. Full frame (in this comparison) will have greater circle of confusion and both have a very similar angle of view so you’d shoot them from the same distance to get the same framing.
Effectively the full frame camera can produce greater dof at wider apertures when shooting from the same distance with “equivalent” focal lengths.
The colour shift occurs on digital sensors when you shift the lens this can be demonstrated quite easily with wide angle lenses.
I presume we acknowledge the same facts, but with differing methods of thinking about them and so different ideas about teaching them. I approach the idea of DOF as "caused" by the needs of the format. Small formats have large DOF because the light ray projection travels a shorter path to the format, and so spreads less. Large format has a longer path & spreads more, so less DOF. Small formats have problems with diffraction because an fstop is a ratio of length to opening, and a "normal" (diagonal of the format) lens is shorter on a smaller format, so any given aperture is dimensionally (real measure) smaller so with more tendency to cause diffraction (hence you won't see f64 on a small format.)
So, tell you what, I promise to not torture you like this anymore with my quirky pov, and just watch some of your vids and pass the word that you know your shit and express it well! If I had more $$ I'd hit you off, I like you work, but you are going to be big anyway and I have my own money issues.:0
You are very generous in your replies, I'd suggest you step back for your own peace of mind. I've read some of your threads and some people do like to just argue. I'm pleased to discover your tube!
I agree with you I think we are discussing the same facts but with slightly different methods. I would say that although Full frame lenses don't have F64 they won't necessarily need it because you can simply shoot with a wider lens and get a similar result. In the same way, you won't get f1.2 or f0.95 on medium format because you can just shoot with a longer focal length. There is one advantage for full frame and that is exposure is going to be faster due to the wider apertures available.
I don't think you're torturing me at all and the discussions I have on UA-cam are always fun for me until someone becomes personal against me, at that point, it's no longer a discussion and I end the conversation. You have only discussed the points and for me, that's very enjoyable.
As to the Rodenstock lens - which one is it? An old 4"x5" lens or a modern "digital"? Resolution in larger format in line pairs per mm on film was always lower than in top small format 35mm. Sensors now are beyond film capabilities and lenses have become the image quality bottleneck. In this light it is extremely important to know exactly what lenses were used.
Not sure if it was the film or digital version but I've placed a link below to the lens we used.
bhpho.to/2C15Iug
Sonder Creative - this is a modern "digital" lens. You can read in the specs it is acceptable down to 5 micrometer pixel pitch - that you are near to with your sensors. Your test illustrates you have hit the resolution limit. Looking at the design, my guess is, this lens should do especially better in absence of chromatic aberration.
Re-reading (YT recommenreminders). Nice comparison, b.t.w. The Linhof can maybe handle 4"x5", but the Rodenstock's lens cannot. This lens's image circle is 90 mm and this sensor/back is the largest Rodenstock recommend. 4"×5" diagonal is 163 mm so you'd need am image circle at least 200mm (8") to be able to do a bit of tilt and shift.
I think the answer to the surprising observation is in Rodenstock's advice on maximum resolution (5 micron). That would set the limit for the digital back at 80 mp. Corner sharpness issues, meh.
Can you please make a comparison Canon and Nikon 24 tilt shift?
Unfortunately, it's unlikely that I'll be able to do that. I don't own a Nikon system but I'll ask around and see what I can come up with.
Good review. I've got the Canon 5DSr and use the EF 24mm TS-E II which is exceptionally good for what I use it for; capturing monuments and built heritage subjects. I do occasional interiors as well and its perfect. I've even used it for landscapes which is my bread and butter source of income and frequently make 4ft prints from it which are near flawless.
Unless you're printing billboard size stuff, most don't need an over complicated, overpriced elephant like that technical camera.
Graham Harris Graham Thank you so much for the comment and for watching.
You know the 5DSR would be more than sufficient even for billboards. Viewing distances mean you don't need high dpi prints.
Thank you again.
I hate to say this but I use a Rodenstock 40HR with an IQ260 and I don't recognise the results you are getting here. It literally eats the Canon 24TSEII for breakfast, in terms of sharpness and at the edges. I know because I used that lens extensively for years in architecture. It looked to me that the Rodenstock wasn't focused correctly - which would give you massively skewed results. Also you are comparing both files at 100% when really the Phase files should be at 70% or so to give an equivalent FOV. And I'm not sure why you are looking at files at F5.6...I use F11 an F16 on the technical camera exclusively. In my experience, the depth of field of technical camera lenses seems to be greater - at least, it looks that way to me. I hardly ever need to touch focus to get razor sharp depth of field front to back.
funland108 I compared various apertures from f5.6 all the way to f22.
At equivalent apertures the Rodenstock isn’t sharper it’s a worse lens in terms of sharpness and will have more diffraction. Once again at equivalent apertures and I explain why equivalent apertures are a true and fairer comparison.
Also 100% if something is the same relative zoom. 25% of something is always a quarter regardless percentages make it a fairer comparison because I didn’t test resolution from the sensor but sharpness from the lens.
Comparison with the new Fuji GFX100s & their superb GF lenses would be great. Price:performance best yet :)
Well, I have to add my comments to this thread since I saw this video and also shoot architectural images for a living. I love good cameras and great lenses. Basically, good or reasonable quality equipment is a given. You must have the right stuff for the work you do as a professional. However, none of this means anything in comparison to how you as a photographer treat your subject. Architectural photography is a story telling process. The idea is to convey what an architect had in his mind regarding how best to solve a problem. This can be done with or with out superior equipment.
I'll take a unbelievable photograph that has a less quality but really conveys the point over a ultra high quality images that tells nothing about the subject and add nothing to the story. Composition, lighting and context are my top three priorities not the resonation of one lens over another. Buy what you can afford and do the best you can with it. Learn to be a photographer not a technician.
Scott Pease you’ve completely missed the point of the video. This is not a video about creativity.
You misunderstood the concepts of angle of view, depth of field on different format size and magnification ratio.
Depth of field IS different at different focal lenghts, but since the angle of view is similar on the lenses, the justification for the difference in D.o.F. is caused by the fact that the bigger the sensor the shallower the D.o.F. is, at the same focal lenght and same aperture.
On a 24/70mm lens the bokeh is different at 24 and 70mm because of angle of view AND magnification ratio. You should compare difference in bokeh keeping the framing the same in both shots.
Stefano Catalani sensor size does not affect DOF. Circle of confusion aside smaller sensors and bigger sensors have the same DOF if the distance to subject is the same. What you’re describing is the difference in DOF due to changing the distance to subject but distance will change your DOF regardless. If I take a picture of something using a 24mm lens and then crop the image in Lightroom the angle of view has changed but the DOF and focal length is the same.
In my experiment I demonstrate how when shooting from the same distance to subject, different focal lengths will give a different DOF even though the angle of view is the same. This is a physics based fact of light.
should have considered the olympus M4/3 EM1 mkii high res-mode 80mp RAW, then you would really put 'the cat among the pigeons' I think the matter of scale should have been addressed, as the 100mp images would have more detail at the same scale, plus the noise would have been less at the same scale, either way thanks for the video, ignore the haters as UA-cam is always subjective.
Jamie R. Thank you for the kind comment I really appreciate it.
In regards to scale both were zoomed in at 200% which is the same relative amount. I was aiming to test the lenses instead of resolution.
I’ve heard some really good things about the Olympus, unfortunately I don’t think it would be as effective for architecture due to not having any native tilt shift lenses.
Do you have any thoughts about the pixel shift feature in the A7RIII?
I downloaded the files, and as several people have pointed out - the Canon is very soft in the corners compared to the Rodenstock - which seems far sharper, I had to check that I was looking at the correct images - because this doesn't match what you're saying in the video - quite the opposite.
I also watched another comparison of someone testing the same lenses, but with an IQ140 - and it's still far sharper than the Canon, I think there's something amiss with your testing regime.
Herpa Derpa the canon is significantly sharper on one side vs the other however the same is true for the Rodenstock. The centre sharpness however is much better with the Canon.
Also if you look at them at equivalent apertures and NOT the same apertures then the canon is still sharper due to less diffraction.
I don't know why you're transfixed by centre sharpness, any lens stopped down to F8-F11 is going to be sharp in the centre, even a 17-40L stopped down is sharp in the centre..... That said, I can read the menus at the back of the room on the phase, not on the canon - they're just a blur, so even taking focal length and Mpixels into account, I'm not sure why you're banging on about the canon being sharper... It clearly isn't, and other peoples comparisons on more uniform images show this.
As far as lens resolving power goes, the battle is won and lost in the outer quadrants, and as per the MTF charts, the contrast and resolution falls away drastically on the Canon, as it does on your image, whereas the Rodenstock is far more uniform edge to edge - especially in the outer quadrants and corners, this is why the Digarons are so expensive, and are some of the only lenses that have enough resolving power to meet the strict standards of aerial photography, where specific levels of contrast and spatial resolution are required - that don't exist on 35mm lenses.
Herpa Derpa I'm not just banging on about something without providing evidence. The sharpness differences on the edges isn't as uniform as you're claiming with the Rodenstock and diffraction is horrible on the lens.
Ultimately the Canon tilt shift is much sharper and centre sharpness is obviously more important than edge sharpness regardless of whether you admit it or not.
The other claims you're making if you can demonstrate and prove them I'd be very interested but as with most medium format related stuff it's just unfounded and widely believed claims, nothing more.
You don't know what you're talking about really.
Even the MTF data for the Canon shows that it can provide less than half the contrast and resolution of the Digarons, and falls apart at anything more than around 20 lp/mm compared to the Digarons which are good to around 80 lp/mm @ MTF 50.
And centre sharpness is not more important with wide angles, that's the whole point of MTF - resolution and contrast across the ENTIRE image circle, that's why very high quality wide angle lenses cost so much money - because it's extremely difficult to make a wide angle lens sharp in the corners, that's what you pay for when you spent lots on a wide angle. You should know this seeing as you claim to be an architecture photographer lol
As for claims I'm making - other people in the thread have provided samples of the images showing quite significant distance between your claims vs the reality of the files - you've simply cast them aside as being wrong, so I'm not going to do it again because it's obviously wasted effort.
Herpa Derpa the reason why they're wrong and why you're wrong is because you use one small section of the evidence to draw your claims.
You're also not taking into account the heavy diffraction the Rodenstock lens suffers from.
I'm using the overall aspects of the evidence to present my argument in a fair and balanced manner and taking int account the diffraction the Rodenstock lens has.
Again centre sharpness is still more important you simply won't acknowledge how the canon is superior in that area. The edge sharpness was also better on the canon although both lenses are performing slightly off one a particular side.
Consider the fact that the subject is curved so this may be adding to some of the anomalies. The Canon is never soft on one side vs the other. There are other example where the canon has been significantly sharper on both sides of the edge vs the Rodenstock however that will be discussed in an upcoming video.
Talk to me about the diffraction because all of the people who disagree have completely overlooked that aspect. Talk to me about the lens performance at equivalent apertures.
So easy to pick an choose the info and please don't bother trying to use that argument against me because I have discussed at length and agreed where I was wrong.
Another great video with a twist. Keep up the good work. I m surprised by the Canon glass quality.
Who would have though the 24mm from Canon could be so good. Thank you for your continued support, I really appreciate you watching and leaving such awesome comments on my videos.
Love this video!
BTW I own a Nikon D810. Some days I think I'd be happy w/ an APS-c sensor (saving money size and weight) other days I wish we could afford a medium format (and that digital medium formats were full frame medium format (like 6x6) instead of being just 35mm plus)...anyway when I catch myself going from one thought to another I figure 36x24 (as in my D810) may be the sweet spot.
Interesting video..
I have to say though, currently owning the Canon 24mm TS-E II, I find this test a little bit pointless when you are talking about architecture photography but there is no mention of any type of movements? That is when you really see a lens of this type either fall apart or shine...
wallpaperviking comparing the lenses with the movements involved gives a distinct advantage to the Canon lens because it's been fully and properly designed for that specific mount.
Too many variables involved when shifting the lenses and comparing that to make a proper test out of it.
Fair enough....
I just know that this is another thing that the Medium Format lenses get praised for, ultimate corner sharpness.
What adapter was used for the Sony A7RII? Any loss in sharpness?
Interesting comparison and it shows that full frame is up there for now, but ultimately it will loose out to medium format as pixel count rises and medium format sensors improve. But you are right it's not all about the sensor you have to have the best lenses as well and medium format cost an arm and a leg. Personally I'd use whatever system the client wants me to use, as they say they are only tools for the job, right! To working professionals it isn't about pixel peeping, it's about all the other aspects and keeping the client happy:) If they aren't bothered(!) what you use then use whatever suits you best for the job in hand. At the end of the day if the client is paying then the client gets what the client wants!
I think we should also remember the high medium format megapixel chips that Sony are producing are being driven by demand and that's a good thing for everyone. If you can write off the cost of a Phase One or Hasselblad H6 etc. then do it. If not do what lots of pro's do and hire the kit when you need it and pocket the savings! Cheers:)
Hi Kenny, thank you so much for the comment.
You raise some interesting points but I don't think I can agree with all of them. For instance, in regards to pixel count and sensor size, similar arguments were made in the early days of full frame about how many pixels can be put onto a sensor. There are far bigger budgets and R&D capabilities available for full frame compared to MF. Canon individually is bigger than DJI and the whole MF industry combined. Currently, there is already a 250mp APSH sensor available from Canon for industrial use which works very well. Tech like BSI sensors and so on make FF more effective even with smaller pixels. Technology evolves and improves and FF manufacturers have the advantage in all those areas.
Also, pixel peeping is very appropriate when the price difference is so significant. You have to ask yourself why are you paying so much more for worse results. Even if the results are only slightly worse it's still an illogical waste of money. Even hiring the equipment doesn't make sense because you're still getting worse results. People would justify the cost of MF due to the supposed better image quality. This is no longer the case, you now spend more to get worse.
The lenses need to be updated for MF but they just don't have the money for it. MF is a small industry with very little capital for future investment compared to full frame. Sonys FF sensors are outperforming their own MF sensors in several key areas because that's where the money is (relatively speaking).
We don't have to agree, however, Kenny, I'm very thankful that you took the time to write all of that and for watching my video. It really does mean allot to me. I hope you're well.
Canon's chip was apparently done as an exercise and for the obvious PR, it's not going into DSLR's etc. anytime soon I think, with perhaps clients with deep pockets e.g. the military possibly, perhaps! 250mp's is going to need some serious hardware to keep up with it. Count me out, my old Mac couldn't cope :(.
Sorry hiring does make sense if your thinking of parting with a large chunk of cash and after all if the client is paying, who cares! As for medium format's supposed better quality, it's a yes and no answer. It was in the past due to the larger file sizes than APS etc. Now however, there have been great gains in full frame 35mm CMOS sensors as that's where the money is. Yes I did give it a great deal of thought at one time and tried out both Phase and Hasselblad H systems. I did buy Phase but fitted it onto a Hasselblad H camera, and at that time the larger file sizes were better than what was being offered by Nikon and Canon, Sony at that time was… well crap to be honest. How things have changed eh!
Medium format is still playing catch up, but then again, the elephant in the room is Fuji! With the prices dropping as is evident with the new 'medium format' GFX 50S will we want 35mm full frame, obviously only you can decide. So bigger Sony sensor; 50mp less noise than 35mm full frame; relatively affordable; slower certainly; modern Fuji lenses; or full frame 35mm….hmmm..let me think about that! Hopefully this will shake up medium format, Phase One, Pentax and Hasselblad had better look out. How long Fuji intends to continue using Sony's Bayer sensor though will also be interesting.
Yup I fully agree MF lens are in the main long overdue an update in some cases, but then again it would have been interesting if the comparison had been made on one of Phase's new bodies and a suitable Schneider lens e.g. the 120mm f5.6 tilt shift lens. I do wonder if the Schneider lens used was faulty though! Not sure, but the 100mp Phase back is a CCD sensor; right(?); if so it's not going to be was effective as a modern CMOS sensor in the shadow areas. Although more than capable for most needs. Re-updating the lenses, yes by all means but you need deep pockets for MF lenses, new or second hand!
Cheers:)
Oh yea I completely agree the 250mp sensor is solely for industrial use and not for DSLRs, but, it demonstrates how the technology isn't that far off and the budgets are with FF as opposed to MF.
The Phase one back that was used in my video is the latest most up to date modern CMOS sensor with 100mp. It's the current flagship which means that is the absolute best that medium format can achieve in the shadows and so on. Also, the Rodenstock lens was in perfect condition and someone sent by Phase was there to monitor the whole process.
I think even with companies like Fuji making their medium format if the lenses aren't up to the standard they will still be behind. The GFX only shoots at 14bit so is there any advantage? Fuji are the ones that make the lenses for Hasselblad and their flagship lenses are terrible in performance. I'd love to try the GFX before I make my conclusions on it though. I have also tested the 50mp Phase One back which is the same sensor that's in the GFX and it's performance was noticeably worse than the 100mp back meaning much worse than the Sony FF. I don't, however, want to use that comparison as a way saying what the GFX can do because Fuji may have updated theirs.
I've also tried the 120mm tilt shift and it's an ok performer, deceptively light it doesn't weigh very much at all considering how big it is.
I do have a part 3 coming soon where I look at some of the latest Schnider lenses the ones with the blue rings.
I look forward to the next instalment then:)
what's that thing that moves in the background at minute 2,49???
andrea da gasso lol you mean my Batman mug?
Oh Yess!!! I thought it was something far on the floor (a tower), as it's blur... perfect, it makes sense, thank you!
Great job on this! That 5DsR IQ is beastly
Thank you so much for the comment. I agree the 5DSR is a beast of a camera.
I'd very much like to see this test repeated with the Phase One iq3 camera and s/k lenses and images processed with capture one.
This was done with the Phase IQ3 and images were processed in Capture One. When you say S/K lenses, are they Schneider? If that is the case then full frame would still win because I could simply use an Otus lens which is much sharper than the Canon tilt shift.
Thank you so much for watching and for the comment I really appreciate you taking the time.
Sonder Creative but your video clear stated you were using Lindhof camera with Rodenstock lens, then you knock the lens for being old and out of date. When I wrote s/k I was referring to the Schneider Kreuznach blue lenses. I'm not trying to be difficult, but as soon as you said lindhoff and rodenstock i questioned whether or not I should take the time to watch the video.
Savage Frieze I didn't for a moment think you were being difficult. Sorry tone of voice doesn't carry through in text form.
Rodenstock lenses are still in production and some of the best lenses available for medium format. SK lenses are good but not anywhere near as good as the best available for full frame.
Rodenstock lenses may quite possibly be better than SK lenses too but I haven't compared those.
Sonder Creative indeed they are wonderful lenses. My wonderment is that you are comparing three different pieces phase one back, lindhof camera body, and rodenstock lens against Sony and canon systems. I would appreciate seeing the Phase One back, xf body and Schindler lens, he whole system as it was meant to be used, compared against the canon and Sony systems. If you have done that comparison please let me know where I can view the results.
Savage Frieze the reason is because this is the best system for architectural photography. This was a very specific test for one kind of photography.
Part three will go further into phase one and some of the SK lenses too.
Thank you for the discussions please keep in touch and I'm happy to answer any more questions.
You look to be confused... (sorry for saying that). What do you compare? Medium format vs. 35mm or do you compare lenses? You show the result of the lenses, but pass the buck to the cameras and sensor size comparison.
Photone Photography What do you think I should discuss when it comes to the sensor size?
I mentioned how you need to shoot with a longer focal length to get a 24mm full frame equivalent and that's actually not an advantage. The shallower DOF due to the longer focal length (because of the bigger sensor) means you have to stop down allot more and then you get diffraction.
Bigger sensor doesn't always mean better.
What do you compare in this video: the performance of the sensors Canon vs PhaseOne? Or the optical performance of the lenses: Canon 24mm vs Rodenstock 40mm? For instance: if you mount a Schneider lens on a 100 MP PhaseOne you get drastically another result. If you compare the performance of both cameras (Canon vs Phase: which is a ridiculous comparison, like comparing a family car to a pickup truck) then, you are supposed to keep the same lens as invariable. If you compare the performance of lenses then you are supposed to keep the same camera as invariable.
Sorry, but the video sounds pointless. It doesn't tell what is wrong, what is better, which one sucks (PhaseOne 100 MP sucks? You'd better get a Canon? Never shoot with a Rodenstock lens? etc.) what am I supposed to change in order to get a better result? What if I shoot with a PhaseOne IQ250 and a Schneider Kreuznach 35mm LS f/3.5 lens-for instance? Hundred different possibilities... each one gives a different result.
Neither scientific, nor informative. I am really not trying to dispraise, with a friendly intention, the video does not make any sense at all.
Photone Photography I compare the lenses and the canon beats the Rodenstock quite comfortably.
I touch upon dynamic range a little and demonstrate how shadows are better on full frame and highlights are better with MF.
That's all I compare. Also this is part 2 and so far there hasn't been a single lens from medium format that has beaten the best from full frame.
Part three will further demonstrate that point about MF lenses.
Hi! Still haven’t seen the third video in the series. Please post soon. If your results are the same or different from critics stand tall and be proud of your work.
Savage Frieze hey thank you for the comment. There has been a delay but I think my timing for the video will actually work out. I’ll be uploading several new reviews and comparisons with a bunch of medium format cameras. They are coming I promise.
how about Nikon & the Tilt shift nikon lenses ?
Gokul Rao Kadam I haven't actually tried the nikon version yet but I am planning on getting hold of one soon.
Thanks for watching.
Interessting comparison.
It seems to me, that looking with the naked eye at a print I most probably could't tell the difference between these pictures.
DeletedDelusion you’re probably right. Once printed and looking with the naked eye they may look very similar. Goes to show spending all of that money doesn’t give you better results.
Thank you for watching and for commenting, much appreciated.
Excellent video. I'm not surprised at the hostile comments. Over the years I've seen similar reactions any and every time technology has brought two classes of photographic equipment closer together in performance.
Can the 35mm FF bodies completely replace the MF Phase One in every respect and for every purpose (every lens, every subject, every print size)? No. But the capabilities have never been this close between 35mm and 645. For a wide range of lenses, subjects, and print size combinations you could shoot with any of the three cameras and produce excellent results virtually indistinguishable from one another. And as you demonstrate here it's entirely possible to get a better result, in some cases, on the 35mm bodies simply because the best 35mm lens for that use case was better than the best available MF lens.
We truly live in a golden age of photographic equipment and image quality.
I too use TS-E 24 mk2 on 5dsr and find that it can easily resolve
cleanly down to individual pixels; it is by far sharpest and highest contrast lens
I have used. The only Canon lens I have tried that comes close to the TS-E 24
mk2 is the 300mm f2.8 IS. Thank you for showing, what I think, is your unbiased
comparison both in dynamic range and resolution of the three cameras, which by
the way I totally agree with. Thank great for the video
Thank you very much for watching my video and for the really kind comment. I was very surprised at how good the 24 tilt shift really is, I agree it's the best lens for architecture and on the 5DSR it's incredible.
Thank you again :).
I like to see how the images would come out if you shot 4x5 film instead from that linhof technika camera and compare the neg scanned vs the canon...Im sure the linhof would be sharper overall on the edges, granted if it is shot correctly vs having the medium format camera shoot the ground glass. The circle of image would not be the same an perhaps that is why the edges are softer on the phase one. And maybe the Linhoff should be replaced with a better camera and lens setup like a full featured field camera like an Ebony or Chamonix 4x5.
The same could be said about medium format digital vs medium format film. Film captures far more detail. tones and doesn't produce any digital noise.
Point is medium format digital is not anywhere near s good as it's being made out to be. Full frame Digital is noticeably better in almost all areas except megapixels.
Also if you were to use a non tilt shift lens on full frame vs a non-technical camera for the Phase then full frame would still win because I would just use the Otus wide angle lens. The Otus is much sharper than the Canon tilt shift which is already one of the best lenses you can buy.
You are confusing sharpness with contrast...
Contrast does not improve diffraction. This is sharpness being displayed not contrast, you're confusing it.
what you like on the images is the contrast and colour of the Canon. The Phase One gives you a flatter image for being processed. And remenber that you are not using Phase One Glass. I mean it when i said it. Test a M43 against your Canon. You will like the Olympus more... its your thig. And mine too :)
I will be uploading a video where I compare native glass on the Phase too.
Also in regards to contrast and sharpness, do you believe the Rodenstock is not suffering from diffraction more than the Canon?
I actually really want to try the Olympus, I've seen some awesome stuff from it.
they are both getting diffraction really early. REsolution on the Phase shows you more in detail the weaknesses of the lens... its hard to say. where can i get the raw files?
I'll see if I can get hold of the RAW files again, in the meantime I do have some high resolution TIFF files available to download via the link in description.
Fantastic video! Thank you very much for making it! It would be quite interesting to see the comparison with Nikon D850. Although I feel like maybe the latest Phase One lenses should have been used for the comparison.
I might be able to do a comparison with the D850 I'll see if I can get hold of it. Phase One lenses are amazing however they're not geared towards architectural photography, therefore, would have been ineffective for this comparison. If I were to do that test I would use an Otus lens on the Canon.
Sonder Creative Thank you very much for your answer, I really appreciate it. It would be amazing if the Otus line were expanded a little. I have had the privilege of toying with them a little bit and they truly are fantastic. Although, stopped down I see absolutely no difference between the Otus and the Milvus line in real life.
@@danieldelorme4021 The Otus is more expensive; that's very real! ;-)
The review is nice because it makes you understand that you do not need a 30k back to take good pictures. 10 years ago there was a big performance gap between digital backs and Nikon/canon. Today that performance gap has decreased a lot. Aniway, I was lucky enough to shoot both formats and i can state that today A DIGITAL BACK IS STILL THE HIGHEST IMAGE QUALITY TOOL AVAILABLE ON THE MARKET. 16 bit and superior dynamic range. Bigger is better hands down. In this specific case of the architecture photo the full frame camera performed better, but no 35mm camera can give you the results of a digital back mounted on an optical bench in studio/table top photography. It all depends what you are going to do. I agree anyway that the cost of medium format is the hot button issue.
Cheers and have a nice day.
tonyrebel28 I’m not so sure I can co-sign everything you said. I’ve used and continue using medium format regularly and with cameras like the D850 the advantage is with full frame now. The best lenses are with full frame too.
In regards to colours I do show how you can achieve medium format colours and even better in my latest video. This comparison was done with the Trichromatic sensor too.
Having said that yes the sensors in these cameras are the best but they don’t have the lenses nor the systems in place to take full advantage of those sensors.
Sonder Creative, Excellent video of comparison.
you should do another informative test on the iphone VS the phase one, like in one of your previous videos.
Sine Nomine that was a one off comparison done purely for fun and out of curiosity.
to compare the Phase one back PROPERLY with these, you need to use the back on the XF/XT models, with NATIVE lenses, using the back on a cambo actus is NOT the best to show off the back. with Native lens, image info is passed to the back ( lens data is on database in back), and corrected for, ICC / black balance is saved, and applied to the image in C1 once it is opened, giving best output possible, from the raw files in the camera.
XT models didn't exist at the time of recording that video and Native lenses weren't effective for architecture at the time.
This is a comparison specifically for architectural photography.
I'm working on getting hold of the XT though.
You have to take quick a rendez vous with an Ophtalmologyst
Paul Dahan LOL,
I did, she agreed with my results from this video too :p
There are movements on a view camera to manipulate the film plane since LF lenses are longer to achieve the the same DOF. You can't learn the movements in an afternoon for a comparison test.. Sorry, but you need to have an expert on LF photography set up the test and I would just imagine the results would be much different.
steve tilt shift lenses have movements too lol.
Also you don’t get a longer DOF you have a different plane of focus. I’m well versed in being able to change the plane of focus with tilt shift and tech cameras. Stop assuming my ability without actually knowing.
Take away what you think is personal and you still got one unfocused image .. Use a flat wall with some texture and use actual large format film and then the phase one and canon crap and try again.. Use a cheap flatbed scanner on the LF film and you'll see resolution man.. That phase one BS and that canon crap will be just that man.. Nothin personal just do it right.! To many variables the way you presented that man.. Dig?
You see, just cause you put a phase one back on a large format camera don't make that image it produces a large format image. Hell no.
steve no one said anything about large format or suggested anything like that what are you on about?
The image is in focus you just don’t like the results get over it. The variables have been countered effectively.
Once again get over it canon lens is sharper.
You could not have missed the boat further here.
Dynamic range: You said the dynamic range was not much different yet, yet it provided much more shadow quality than one of the SLRs and several stops more in highlight than both cameras....Soo what does that tell you about your exposure on a high dynamic image with the POne? Tells me that you left a massive amount of dynamic range on the table. Shoot for the highlights in a high dynamic range image and get back to me.
Tonality. No mention of the quintessential difference between MF and DSLR?
My dinosaur 22mpix hassenblad (retired) has much better tonality than my D810....
Amateur really.
Micros contrast? What? You do understand that sharpness is a direct trade off with other desirable attributes in a lens?
Have you not wonder why a MF landscape lens has 4-8 lens elements while the Canon has how many? Huh!
Why is that? I doubt you can even see the difference.
Educate yourself...and your eyes
Saturation?
Saturation is lost with every piece of glass you add right?
You though saturation was a slider in the edit?
No comment on the look of the focus fall off?
BS you say?
All this tells me is that you have no idea what you are looking at.
Smokin07ram shadow recovery for full frame is better highlight recovery for MF is better.
Even the 5D4 beats the Phase for shadow recovery but not highlights. It comes down to which you value more.
Please demonstrate your other points. Claims are easy.
Why didnt you use Alpa incase you need a technical camera or why didnt you use the XF camera? This comparison is nonsense to me.
Zef Markaj does the alpa have the same amount of movements required for architecture photography?
Also I didn’t decide on which tech camera we used this was decided by phase one based on the best possible option for architectural photography.
Remember this was specifically for architecture otherwise I would have used an Otus for the comparison which is significantly beyond both these lenses.
A very knowledgeable gentleman
Whoops. Meant to say here that I’m holding my breath and waiting anxiously for the third video!
Also wondering how the Phase S/K tilt shift lens works for your application.
Thank you for the comment and the interest I’m going to work and aim to have it ready for either this week or early next.
Nonsense comparisons...if you get a commercial gig from an advertising agency shooting sony, I'd be really impressed. Bottomline is, regardless of any comparisons, the art directors wanna see that big fat medium format digital back when they pay 80k for a photoshoot for their client.
You've really missed the point on this one Natan.
Also, I've worked on several large commercial shoots and I don't use the Sony, ever. In any case do you wanna know who actually cares about the gear, other photographers. The art directors, clients, agencies none of them care about what camera you're using and on many occasions, they don't even know what you're shooting with. The bottom line is always the result because that's what makes money for the client.
Don't get me wrong I do use medium format on occasions for shoots depending on the clients printing requirements. The point of this video was to show how much fluff is behind medium format systems and how allot of it is just romanticised nonsense.
Full frame has THE best lenses available.
The canon 24mm mkii is not that great at all, had it for years and find it frustrating with higher res sensors. I just road tested the 32mm HR rodenstock on an IQ150, comparing to my A7Riii with 24mm TS mkii. This is a proper comparison because they are nearly the same megapixels. The medium format absolutely blows my current setup away - way more shift, proper edge to edge sharpness and no chromatic aberration. It's awesome.
I compared different lenses and cameras to you but sure I'd love to see your results, please. Are you ok to provide a link with images where you compare both at equivalent apertures? It is important to mention that when adapting wide angle lenses to Sony you will see some softness on the edges this is not something you'll notice as much or at all on native bodies like the 5DSR.
Finally, to make things completely fair, can you please perform a lens cast calibration on the Sony/Canon setup too, please.
The 24mm is v soft at the edges on native bodies too. Maximum 12mm shift is virtually unusable at the very edge. I'll do some test shots, with the Rodie stopped down by a stop or two to account for the longer lens equivalence, and get back to you. I'll see if I can LCC on the Sony but I've never done this before. I know that Capture 1 doesn't support it for Canon.
LCC should work fine on full frame cameras too I use it on Canon all the time, I made a whole video about that too :P.
Thank you looking forward to seeing the results, much appreciated.
Fair enough. I've never seen the need to do it on dslrs, can't see a cast with any of the Canon TS line up. A few years ago the Capture One team told me it was pointless so I never bothered. I guess you just use it for light fall off. I'll check out your other vid!
Would love to see a comparison including a foveon camera
I think Phase One is only the best with it's XF system and Schneider lenses. If you want resolution then buy a Vizelex Rhynocam and a sharp Hasselblad lens and then you will be happy with it, and much cheaper than medium format, or extreme resolution full frame.
Bálint Szász I haven't tried the Vizelex system I might have a look into that. Unfortunately all current Hasselblad lenses have very poor performance not a single one can outperform the best full frame lenses.
Sonder Creative The RhinoCam is just a 500$ slider kinda device, it lets you take 6 full frame images together (or 12 Aps-C, it works like that) and then you stich together them in Photoshop. And also, I meant Hasselblad V lenses, they are cheaper and some of them are much sharper then the new Hasselblads ;)
Oh yes I remember now I saw that and it can be used with a Sony A6000 series camera. I think you might be right about some of the older Hasselblad lenses as they were made by Zeiss.
Personally, I would use a tilt shift lens and do a vertical panorama with that on a Canon 5DSR but your method sounds very interesting to me. Thank you for the suggestion.
They have various mounts for both the camera and the lens, but for DSLR you can only get the Hasselblad V mount.
The two crops shown (left Canon and right Phase One) are not from the same part of the image. I find that it somewhat weakens the demonstration. Very interesting angles of analysis, however.
Café Globulot well it doesn’t really. Angle of view is slightly different however the edge of a lens is still the edge. Also the sections where they’re slightly different are still bricks and you can see if something is sharper or softer.
By the way… an interesting article about autofocus lens calibration : scenictraverse.com/blog/2017/11/10/calibration
Yea but these are manual focus lenses what use is AF calibration for MF lenses?
At 12:05 the medium format had far less noise and was far sharper (while still a bigger image). Your comparison is deceptive....showing them both at 100% is not fair.....the medium format image is way bigger. You should have adjusted for equal size. You got a point with depth of field (f10 vs f16), but I have seen lot of results where medium format always won, and at 12:05 you see clearly the difference (even at a bigger image size, medium format won). Key is buy the best glass.
In this comparison the Canon lens is clearly better and even with the resolution advantage, the rodenstock lens can't keep up. Phase One however do have much better lenses available from Schneider so those tests may show different results but then those lense may not be suited for architecture.
I like watching videos on cameras I will never be able to afford.
Kuhwristchin P if you think you can/if you think you can’t.
For a real comparison, I'd want to see the images at identical magnifications. Also, show them photographing the exact same areas in each test.
Brian Cowan You’re welcome to download the images the link is still active.
The comparison wasn’t about the resolution but about the lens. The lens itself isn’t as capable as the canon and that was the main point.
Thanks for the reply. I've heard that larger format camera lenses are not ground to the sharper standards of smaller formats because they depend on the larger format to make up for the difference on magnifications. Seems odd to me and i don't know how true it is.
brianncowan sounds like nonsense because if you shoot with some lenses for the GFX and even some of the new blue ring lenses from Phase One you see how sharp they actually are.
The issue is medium format tends to rely on perception and they just don’t update their lenses as much this is especially true for Hasselblad. They’ve stuck with the older film days level of quality and rely on perception or excuses for poor performance.
Shooting with a Zeiss Otus or Milvus lens you can clearly see the quality and don’t need excuses you can see how good they are. Why does medium format need to make so many excuses why isn’t the quality difference as easy to see.
Sorry for the long reply lol.
you should use an ARCA RM3Di, a combo, or an Alpa... with Linhof, // isn't perfect all the time !
i keep on testing stuff all the time... and still can see much more difference than in your test !
David Giancatarina can you show some of your test results to me please I’d love to see.
Also those cameras you mentioned don’t have any movements do they? It would be ineffective for architecture of they don’t.
they are build for movements !
vimeo.com/24366528
most of times, i don't keep tests... they are just test...
Interesting.
If you are doing any tests between the medium format system and Canon system like I did I'd love to see those results.
ok, i will keep you inform !
Great video. Really enjoyed it.
What about the Fuji GFX!
Michael Horgan working towards getting hold of one but that's going to be less effective for architecture due to not having any native tilt shift type lenses.
Thank you.
yes canon tilt lenses are really good and the medium format does lack new lenses. The one who say old lenses are better are those old folks only based on feels
Side by side images should be same size. (even if different than 1:1 due to different MP size) Your sharpness comparison / analysis was therefore not identifiable.
Link to images at the same size. Canon at 300% and the Phase at 200%. The canon is still sharper.
apalmanac.com/architecture/is-medium-format-the-best-for-architectural-photography-2530
There are a lot of truth in your presentation, but large sensors work on 16bits while all full frame 35 mm are limited to 14 bits, in practice means that the tones are more even and rich. It is true that medium format lenses don’t perform optically to the level of some Carl Zeiss or Sigma lenses manufactured for 35 mm. In most commercial assignments lenses are stop down a lot to gain depth of field in turn the resolving power suffers. Another advantage of having more pixels is that even if the lenses are unable to match the resolution of the sensor, it helps to render smother an more continuous tones, an example could be a simple line crossing at any angle that doesn’t align with the sensor square angle. Besides there is a psicológicas reason when you face a customer with a medium format equipment, you are considered a better and more profesional photographer in contrast to a 35 mm user. The moral is bigger is better!
I can only agree with you :-).
This is seriously laughable.
J.D. Closser why? And don’t just make claims cause that’s easy, anyone with a keyboard can. Address my points, provide evidence.
If you Compare them on 100%, you are comparing two images with different sizes. In this comparison a Canon EOS 10D with 4 megapixels would be sharper than both of them.
If you print both images at the same size - for example 120x80 centimetres - the result could be quite different.
warumbraucheichfüryoutubekommentareeinescheissgooglepluspagefragezeichen lenses determine sharpness not resolution.
I’m not testing resolution. Diffraction is diffraction.
Actually diffraction has nothing to do with the lens. If you know the pixel size and the aperture, you can already calculate the diffraction.
I agree pixel size has an effect on diffraction but to say it has nothing to do with the lens is completely false. If that were true then stopping your lens down would not increase diffraction.
That's why I said that you need to know pixel size AND aperture. Of course stopping down increases diffraction. Each camera has a "diffraction limited aparture" (DLA) that depends on the pixel size. If you stop down further than the DLA, the diffraction becomes visible.
I hope you realise that all you're doing is explaining why my results are what they are not disputing them.
You also said it has NOTHING to do with the lens which is completely false.
You should not use Capture one for the Canon and Sony. It makes no sense. Even with that, your results stay. I love that this time you ask for help with the Phase One .now you are comparing at the different sizes, how will they look at the same size?
Jaime Duncan can’t use Canon and Sony images in capture one?
Really??? Come on stop talking nonsense.
Zoom factor was percentage of an image simple maths there I don’t think you understand relative zoom or relative distances. 200% zoom on one image is the exact same relative amount on any other image regardless of resolution. Also I’m NOT comparing resolution obviously 100mp is greater than 50mp, I compare lens performance which is very different.
I don’t think you know what you’re on about.
I am not a native speaker, is a mistake, I will correct it. What I mean is "you should not " not "you can't". That does not make your comparison unfair for the Phase One, it just give it a little advantage. Using native for all, or standard for all will be fair. As you know Phase One has "secondary gain".
Jaime Duncan ah I see.
I wanted to give as many advantages to the Phase One as possible because that way there are no excuses.
Normally you will want to compare the images at the same size: you either print the images (same size) or share the images (same size). One doesn't take images for pixel peep. I understand the aspect ratio makes the comparison a tinny bit more complicated but it's better. I hope this make you understand why the industry normalizes images sizes (from DXO to reviewers)
Jaime Duncan a percentage of something is still the same relative amount for anything else.
25% of anything is always a quarter of it regardless of how big or small it is.
Both were viewed at the same percentage zoom. Also, again, this was not comparing resolution but lens performance.
Thanks mate. Great video!
Thank you for watching :)
Thanks .
You're welcome :)
While this may appear to be true when pixel pepping on your computer. When coming to real world prints for example 24x36 and 48x60 prints then you will realize comparisons is only factual at on a monitor and not I reality large scale printing. And I have use the Fuji medium format system and the sharpness of images can blow any full frame into there grave. I have never use or even seen a phase one in the Caribbean so I can't judge it but from my knowledge of Fuji medium format system and there technologies I will have to there image quality the edge over the canon. But great video and I always known phase one to be at the pinnacle of medium format if this video is free from bias that why I have not seen any in the Caribbean the result price is to high.
You may have a point but if an image is that soft and loses that much detail then I find it difficult to believe that it will be any better in print especially if no sharpening is added.
Regarding Fujifilm, their lenses are incredible and some of the best I’ve ever used. They’re not quite as good as say Otus lenses or even the Zeiss 135mm f2 but they’re definitely up there. Once Fujifilm make some tilt shift lenses then we can do a proper comparison but as of right not it’s not the best system for architecture overall.
Well yeah , smaller sensor same f stop , more depth of field
Not exactly that, actually quite a bit different.
On your statement about slowing down your process not being worthwhile, it has more to do about maintaining quality of work. If you are concerned about being quick on delivery, that will not help your quality of work.
I found your comparisons about lenses and medium format vs FF35mm to be flawed as well. Had you borrowed/rented a df+ or XF body with a 35mm Schneider lens, there would have been a significant difference to speak of. Your video should have been titled technical camera vs 35mm. If you are going to claim medium format, then use the system to give a proper comparison.
Robert Hold slowing down to think about the shot and making sure you maintain quality is a very different thing to your equipment slowing you down. They’re not the same thing and equipment slowing you down is a problem not a benefit.
I always take my time when I shoot that’s how architecture is normally shot it’s not generally rushed but I don’t want something else dictating my pace I dictate it.
Also why is it flawed making a claim is easy, justify it.
This video was about architecture it’s in the title. XF bodies do not have any movements to accommodate requirements for architectural photography. If I were comparing to an XF I’d would have used a an Otus. That’s a completely different comparison then isn’t it?
Sonder Creative No. The problem with the comparison is that you used two systems that are not technical systems to a technical system. If you were comparing medium format systems to 35mm format systems, it would have been a better and far more reasoned to compare system to system. 35mm systems do not have a technical comparison and I cannot find your comparison in the video viable as a result.
I can't even begin to tell you how little sense that makes.
Anyway I doubt I'll change your mind but thank you for watching.
Sonder Creative I am not the only photographer who has pointed this issue in your comparison out.
Thanks anyways.
I compared the two most viable options for FF and MF if you're specifically shooting architecture. You have an issue because I didn't compare the MF techno to a camera that doesn't exist?
Come on Robert.