People need to understand that numbers on a spec sheet don’t actually matter. I DP’d a comedy special recently that we shot on a fleet of C300mk2’s in Clog2, our editor used the “Utopia” LUTS (Arri conversion LUTS) and we’ve had several people ask us if we shot the comedy special on Alexa 35’s. Keep in mind the C300mk2 is an 8 year old camera, and it is Netflix approved. Sure, newer cameras are sharper, and have 1 or 2 stops more dynamic range. At the end of the day it’s not the equipment, it’s how you use it.
Unless you're making thousands of dollars with high-end clients, don't waste your money chasing gear. This is completely asinine. A few months later, they will be posting another piece of gear that everyone MUST HAVE.
Cameras aren’t about “retaining value” they are about delivering value. This camera does deliver value. Gear by itself doesn’t make the image better but gear and skill do.
@@keylanph right about cameras are just a gear, your work is delivering the value. speaking to that, this camera is delivering value exactly like a lot of 8 - 10 year old cameras you can buy used for 1000 - 3000€, depending on your needs. isnt that a great time we are livin in? just dont worry about the new gear, worry about your quality of work :) - yea, all the influencers will tell you another story, but their main income is to sell the hot shit from the manufacturer.
I wouldn't necessarily relate the AF capabilities of C70/C300mk3 as bad. C300mk3 was released about the same time as R5/R6 were, and C300 had the best AF out of all C-line of cams at the time. I highly doubt that AF can't be improved on the DGO sensor, and if it's connected with rolling shutter capabilities - then every single Alexa is a good proof that DGO can have a faster readout. Besides, all recent sensors on the market seem to have a remarkably fast readout in 2024. I suppose they had to pay ARRI licensing fees to be able to use DGO, and it does seem that it didn't work out really well business-wise for them (C300mk3 didn't gain that much traction due to fierce competition on the market).
My C400 out of the box was alarmingly noisy. After I performed an auto black balance the noise seemed to be a fair bit better. Just curious if you ran a ABB by chance for these tests? Great video by the way!
I haven't seen such a good cam review in a while. Awesome job, mate! learned a lot. As an FX30 owner I would have loved to see more test shots from that cam but that quick noise comparison with the C400 really blew me away.
As soon as you started talking about Dynamic Range, I suspected you were talking about Shagral’s video. I already loved your content, but seeing that you’re working on growing & perfecting your craft with top notch info just made me love and respect your channel even more. I subbed to your channel for the immaculate aesthetic that you bring, so glad to see you also run deep into the technical, even more than I had previously imagined. Hope I’m making sense, keep up the fantastic work!
It already exists and it's called C300MKIII - the C400's body is shorter but also wider. I all honesty, I didn't see a great deal of difference save for the LCD brace, which is better articulated although very plastic-y.
@@emilianoranzani6049 totally, it's purely a mount limitation. For doco I really love my fuji zoom, I don't like stills lenses, the fuji 18-55 can't adapt to ef.. other EF cine zooms are a lot bigger which is a bit intense for low footprint verite style filming
i got a used s5iix for about $1500 and it's insane you can get similar image quality (worse frame rates and bad rolling shutter aside) for literally a fraction of the cost i was hoping a c80 would be a good 'A Cam' to the S5iix but at this point im going to hold off for what lumix announces in 2025
Totally agree, I got the test the S5iix earlier this year and was blown away by the image quality and features vs price. On image quality alone, I would much rather grab an S5iix. I have my fingers crossed for an EVA-2 announcement this year
Oh yeah, I watched both videos before making mine for research. On The Film Guy's video the C400 has less noise in the comparison, but that's because they were comparing compressed 10-bit. The FX6 has the ability to turn off noise reduction (which allows more visible noise but the noise looks better) and the C400 has a ton of noise reduction you can't turn off (which takes away some noise, but the noise looks much worse a very digital). In that case, I would much prefer to have more noise, and then apply the perfect amount of noise reduction in post, rather than my camera applying a ton of noise reduction by default. In the same video the FX6 also does better in the latitude test and holds on to colors longer than the C400 (which basically immediately loses all saturation and color detail)
@@jakehayden998 You are far more technical than I. But Northwest Camera Co just dropped a C80 video today saying the low light performance on the C80 is better than the FX6. Both shot in 10bit... I guess because the FX6 can't shoot RAW, internally. I dunno? I've watched the compares. I don't see a motive of the previously mentioned sites to favor Canon? They all rent both systems. It's not a bias. I own the C400. I have a bias. I'm not trying to poke holes. I just don't see what you do. The images are better from the Canon C400 & C80.
Yeah I don’t have an fx3/fx6 with me to compare so please take everything I say with a grain of salt, but if I had to guess- The c80/c400 is probably cleaner than the fx6 at 12800iso in 10-bit, but I suspect it’s because the c80/c400 has a whole bunch of internal noise reduction you can’t turn off. I would love to see a comparison side by side of both cameras shooting in raw
I got the C80. And as a documentary filmmaker, it’s perfect for me. I’ve been making youtube videos about it. I have no issues exposing or with dynamic range. However, the only thing I’m still unsure about is autofocus. I’m still finding the right combo of settings for that because it has so many options. Thanks for your perspective
I'm sure it's awesome for documentary stuff, and honestly I think in most circumstances the dynamic range won't be limiting out in the field. I'm glad you're happy with it! Are you using Canon RF lenses for autofocus? What's wrong with the autofocus? It's just unreliable or slow or what? I didn't have any RF glass to test autofocus while I had the C400 in, so I'm curious.
Some Panasonic cameras also have DR boost sensors. The GH6 was the first one to have it but the G9II and GH7 sensors improved on it to beat most APSC cameras in DR.
I haven’t been able to test those newer Panny cameras that have DR boost, that’s interesting. Hopefully I can get one into the studio soon so I can test that
you’re the first person who actually did a proper image review of a camera. I love how you zoom in on the Noise so we can see exactly what it’s gonna look like. I wanna know how the camera operates at its worst. that’s what gives me a good idea of how it’s going to work. you can make any camera look good. given the time, proper lighting, proper set design and so on and so forth, but what does it take to break the image that’s important.
Good question. I personally would much rather go with an FX6 versus the C400- - Much cheaper - Comparable image quality (even versus C400 Raw) - Full size XLR - I like E mount much better (way more options) But I guess it comes down to ecosystem and taste. There are a lot of people out there that have Canon lenses and want to stay with Canon. Personally I'm still going Sony every time.
On the monitor brightness, take a moment to increase the backlight using the luminance setting. It can be very bright, even in direct sunlight. And this setting it missed by so many ops it's frustrating. (C70 has the same setting btw, but only at a +2)
It's the first thing I always do when I use a Canon camera. I feel so bad for anyone that doesn't know about that setting and tries to use the camera at default brightness
Solid video and super practical takes. Been toying with adding a Canon to my kit next year but TBH wasnt super impressed with the footage from the C400 when I recently used one - but I didnt get a chance to go this deep. Great stuff.
@@jakehayden998 You teach things in such a clear and concise way. I think it would be valueable. Thank you so much for giving your knowledge back to the filmmaking community!!
Nice video! Another quirk I noticed with my C400 is that the default sharpening for raw is cranked way up in the files. Weird default choice on Canon’s part. Did you notice this too?
The blotchiness in the hair wasn’t something just in this latest sensor from Canon. I also saw the same thing in the c70’s DGO sensor that always bothered me. Great video and great breakdown of your thoughts.
Thank you! Yeah my C200 used to be pretty blotchy and weird, but I don't remember it being as bad as it is in the C400. It's probably the ugliest looking noise I've ever seen
I use a 5D mkii with magic lantern raw. I always thought the bad digital looking noise might be a result of the custom firmware, but I'm shocked to see that the noise pattern on this camera is identical. As a reminder, the 5D mkii released in 2008.
@@jakehayden998 It's because of the DCT compression that the camera is using in RAW which is why none of these modern cameras including the RED Komodo qualify as RAW to me. DCT compression is the same "blocky" compression that h.264 and h.265 use. That's why it looks blocky to you The 5d mark ii in magic lantern mode uses no compression and the mark iii along with the digit 5 magic lantern cameras such as the eos m use the old JPEG lossless 1992 which is similar to .zip file compression. It's compressed and decompressed to its full size with no loss in data, usually about 2:1 compression ratio. Fun fact the old canon dslr cameras can shoot raw video in 14bit and even combine 2x 14bit signals into one massive 16bit file with a massive cost to resolution and usability
This is insane. Out of curiosity, were the raw shots done in RAW LT? That blotchy pattern when you zoom in that far is something I’ve seen with the c500ii in LT. Makes that mode almost unusable for me. That being said, the latitude performance here is atrocious. I don’t know if canon is doing a lower ADC readout here leading to a higher noise floor compared to a 14 bit or 16 bit readout. They claimed to do a 14 bit readout on the c200 and that image was a noisy mess at times too. So it’s hard to really tell. I’ve been wanting to get one in to compare with the c500. But honestly, my c500 does not do well with the under exposure either. Most times it’s better to actually shoot at 3200 iso in raw and pull down in post, than shoot at 800 and try to pull it up. I wonder with that in mind if canon is approaching the dynamic range spread differently depending on sensor gain (ISO)
Yeah my test shots were recorded in 6k Raw LT. I would be very curious to see the difference in detail between the 3 different flavors of Raw. The interesting thing is the Blackmagic I was comparing it to was recorded in a lower bitrate Raw codec and it still looked that much better. It's baffling.
@ ok, nice. I think the z8/z9 is probably the better camera but I’m just curious what you think about their raw. I’m assuming it will be better than the c400 due to the lack of noise reduction in camera. Keep up the great work
I truly believe this sensor is the R3 sensor. Same exact specs and believe it or not, when you change the ISO settings on the R3, cleans up at 3200, and then shockingly cleans up again at 12,800. Try renting that camera and do your tests on it and I think you’ll be surprised it’s the same.
It’s never been a better time to grab a cheap C70 of someone who jumped ships. I was disappointed they didn’t include the DGO sensor in C80 and C400 and right from the first images that came out the supposed dynamic range just didn’t seem realistic. I’ve seen a lot of OMG’s “C80 destroys C70 with its third iso of 12800” when the 12800 looked so inferior to 6400 on C70. Not even accounting for the fact that with the super 35 sensor I have an extra stop of light bolted to the front of my camera. Great exposé, dude.
Never really owned any canon cams, came close to the c70 but went with the Komodo. Have recently started to look at the C80 and 400 as really need an all in one doc cam and thought this might be Canon finally shaking things up… but everything I’m seeing is not great news and a lot of old c70 users don’t seem to be falling in love with the image on these newer cams. Thanks so much for running through these tests, incredibly eye opening.
Re: "apples and orangutans" Burano comparison. I've used the Burano on a number of projects since the Spring, from small 1-man doc style shoots, to medium sized crewed productions, and it's quickly become my favourite camera to work with. If you have a chance to rent one, I can't recommend it enough. I haven't touched any other Sony since other than an FX3 for gimbal work. Relative to this video, the latitude continues to surprise me. You almost have to try and clip the highlight. And when you do get noise in the shadows, it's so clean it's borderline acceptable, and extremely easy to clean up. I almost liken it to the digital equivalent of Arri's organic noise, and I think if Sony were to ever unlock XOCN-ST or XT the latitude would be nocking on the door of the Alexa 35. It's also quite a bit more expensive than the C80, which again brings up back to the "apples and orangutans".
@@jakehayden998 whenever I hear someone mention the slow RS or IR, I know they've never actually used the camera. The RS is a non-issue for 98% of projects and I haven't noticed IR in any shots. That's not to say some people haven't experienced it, but I've personally seen it in my footage. I think the biggest gripe is the cost because it's so unaffordable, especially for the UA-camrs reviewing it. Could the camera have been $5-8k less? 100%. But I think Sony's biggest mistake was sending it to a bunch of UA-camrs, who've never been on a proper set, during the launch. All the reviews were "I can buy X amount of FX3/6s for the same price". Meanwhile, the type of person that this camera was designed for mostly don't even have a UA-cam channel. I will say though, the inconsistent output between HDMI, SDI, format, resolution & codec is annoying as hell, but it looks like that's all being standardized in the next firmware, along with the addition of 240fps and a fast readout 4k full frame crop (95% of full frame). If you ever get the option, I highly recommend renting it. It's not hard to find it for FX9ish pricing.
@@jakehayden998 whenever I hear someone mention the slow RS or IR, I know they've never actually used the camera. The RS is a non-issue for 98% of projects and I haven't noticed IR in any shots. That's not to say some people haven't experienced it, but I've personally seen it in my footage. I think the biggest gripe is the cost because it's so unaffordable, especially for the UA-camrs reviewing it. Could the camera have been $5-8k less? 100%. But I think Sony's biggest mistake was sending it to a bunch of UA-camrs, who've never been on a proper set, during the launch. All the reviews were "I can buy X amount of FX3/6s for the same price". Meanwhile, the type of person that this camera was designed for mostly don't even have a UA-cam channel. I will say though, the inconsistent output between HDMI, SDI, format, resolution & codec is annoying as hell, but it looks like that's all being standardized in the next firmware, along with the addition of 240fps and a fast readout 4k full frame crop (95% of full frame). If you ever get the option, I highly recommend renting it. It's not hard to find it for FX9ish pricing.
The burano has a sensor that's very got very low read noise which is a godsend in itself but it's still a 12bit sensor unfortunately. While latitude tests do show the dynamic range capabilities of the sensor, they're more about the effect of read noise on the image and the leeway that gives you in exposing your image. The fact that multiple sensors with 12bit readout have such different results with the amount of noise being the biggest difference says a lot
@@vhauser416 the camera I use literally has 15-38ms of rolling shutter depending on the resolution I use and it's quite funny yes. I just drop the resolution to get a faster readout, sometimes by as much as 60% to get the result I need
I’ve got a pair of FX30’s because they are cheap and soooooo good! I mostly shoot in well lit environments, but even in low light situations, there is still enough info in shadows to make it look decent. Can’t believe how bad the C400 performs in these standard situations that we often find ourselves in. Great comparison! And loved the genuine laughter too 😂
The C500 Mark II has the same "problem". Here's the "fix": Go to the Custom Picture Menu (CP), then Edit CP file, then Noise Reduction, then make sure that the Frame Correlation is set to 1 (not 0). At least for the C500 Mii this fixes the noise issue - pretty dramatically.
ok so considering the a6700 is the same sensor as the fx30... it cant shoot raw but as we saw with the XAVC codec, it looked like the fx30 had a more natural grain pattern in the noise than canon.... so..... is the a6700 giving you better image quality than the c400 even in a compressed codec? I mean obviously for professional onset work the a6700 doesn't have anywhere near the io or functionality or even redundant recording but.... still...
It is kind of crazy to say, but I would rather have the a6700's image quality over the C400 for sure. You're right, theres a lot of other cool things about the C400, but image-quality-wise I think the a6700 is superior
You can compensate for the C80's triple base ISO with controlled lighting. You can compensate for the full frame sensor with a wider lens (to some extent of course). But nothing in the C80 can replace the DGO sensor on the C70. An absolute dynamic range king in the Canon lineup!
I agree, and I’m still buying a c80 based on all the reviews I’ve seen. Mainly because the C80 is more run and gun based camera imo. Idk if there’s any other camera I can throw in a small backpack and get cinema like picture with 24 bit audio and exposure tools built in, decent screen, and internal nd. Literally pull it out, slap on a mic, hit record and record raw video.
@@SamA-kl6pi indeed the C80 is great for run and gun situations. The full frame sensor is really something when it comes to using Canon's RF lenses' full potential. Still for the image purists, C70's DGO sensor is the closest that 4 thousand bucks will get you to an Alexa :)
Great question. SNR of 2 is a good measure of "clean stops" or stops that are totally usable. I use SNR of 1 as a measure of "total dynamic range" since in theory you can use noise reduction and dig out some extra stops even if they are a bit noisy to begin with.
I haven't gotten one in the studio for review yet, but if I had to guess, yes. The Ursa Cine 12k LF is on another level as far as image quality is concerned versus basically anything under $30k. The dynamic range ( even without highlight recovery) is really really close to Alexa level, and the resolutions/framerates/readout speeds are absolutely insane. It's kind of the ultimate camera
I’ve had a brief brush with the Canon ecosystem, but I’m Sony pretty much 100%. I like the entire cinema line that works together from FX30 to Venice. Canon I think is finally going to start rolling a consistent line up from beginning to end, but it’s too late for me. The C400 is the most exciting camera I’ve seen from them in 15 years. It looks really good!
Totally agreed. If Canon is able to bring both their prices and their noise floors down, I think they could be a force to be reckoned with in the next few years
13:08 - That noise pattern not only looks processed but I get a better looking noise profile out of the Panasonic GH5 II ($1,200 camera) with noise reduction reduced as much as possible in camera and recording to a H.265 LongGOP video format to v30 SD cards. Be aware that all Lumix cameras have a bad noise profile on the first frame of a video clip; every frame after is excellent. Canon doesn't know what they are doing here, are crippling it on purpose, or just don't care. I'm not joking ... rent a Panasonic GH5 II or find one in your local dumpster and light the scene similarly. Even with the ISO at 800, the noise profile will still look more like a per pixel sort of noise ... in H.265 LongGOP ... close in visual quality to All-I and with no apparent macro blocking / lossy compression artifacts. ISO less than 800 on a GH5 II will absolutely look better than that ... this is atrocious. Just turn down noise reduction in the profile as much as it will go (-5.0) on the GH5 II to see. Also, Neat Video will give a better result than Resolve Studio's built in temporal and spatial denoising ... but unfortunately Neat Video is CPU only and so is like 5x to 10x slower than Resolve's GPU accelerated setup. So on a quick turn around project, I use Resolve's GPU denoising. But when I have some time (like overnight), I'll have my first node (a serial with Neat Video applied) in my Power Grade render out to the Node Cache to ProRes / DNxHR for a few hours as I sleep for much better results. But again ... regardless ... your eyes aren't fooling you. That noise profile is atrocious. Neat Video will not do well with that ... thanks for the video. My only complaint? You should have had it auto post on Christmas morning. Haha. Subscriber earned!
Thank you so much for this video! That explains a lot why I am not very satisfied with my C400 which I bought as an upgrade to my Pocket 6K. I should have waited I guess, its hard to get clean shadows in sunlight sometimes, in those high dynamic range scenes.
Lol comparing the FX3 xfavc at 3 stops under and the c400 in raw at 3 stops under is ridiculous - of course a raw image will have more noise. C400 noise reduction parameters can be finely adjusted or turned off unlike many sony cameras.
The comparison wasn’t to see which image had more noise, the comparison was to see which image had a better looking noise pattern. Somehow in a compressed format the fx30 is able to a beautiful and granular noise pattern and the c400 in Raw (which should be the least processed) has a noise pattern that looks processed and digital. The whole point is the c400 weirdly has a much worse looking noise pattern vs the fx30 (and typically Sony cameras are the ones that famously have internal noise reduction that you can’t turn off). I would much rather have the fx30s noise pattern vs the c400s. In addition the fx30 has a lower noise floor, but that wasn’t really the point
FX30 owner here, so happy my little camera compares favourably, but surely you can only compare the quality of the noise between it and the C400 when DR is turned on? God knows what RAW looks like from an FX30. Anyway, I’m glad the DR myths are finally being debunked on YT - most cameras are stuck around 12 stops. Even the famous 15 stops on an FX3 is BS, but the marketing obviously works.
Key takeaways: 1) Sony's sensor fabrication is better than Canon's hence better intrinsic SNR 2) Sony uses a better H.265 based XAVC-HS codec that holds up lot's of data in limited bitrate compared to Canon's AVC and internal Debayered/compressed RAW 3) Sony has much better noise Profiling that uses much more Temporal NR than Spatial NR hence even noise texture overall.
I also wonder if Canon has issues with sensor heat. It seems like all of their cameras have the tendency to overheat, or at least run super hot. The C400s fan was the loudest I have ever heard in my life, it sounded like my PS5
I haven't been able to test out the A1 yet, but from what I can tell a lot of the newer Sony cameras have really great noise patterns/latitude. The A9iii is supposed to be crazy good
@@AstralxProd the sensors are mostly likely the same, but the internal processors and colour science are much different. In XAVC they could potentially be similar in latitude, though you'd see different results in what was pulled/pushed. However XOCN- LT will give you significantly more information to push/pull from. 10 bits shades are in the hundreds while 16 bit is in the 10s of thousands. It also see many millions more colours. I've used the Burano on a few projects now and the image is incredible, and it's such an easy camera to operate.
I love the explanation of dynamic range and signal to noise ratio! Have you seen Shagrals video on it? It seems like you have he goes DEEP. Good stuff.
@@jakehayden998 I like your take on it! Simplified and understandable. Also made me question my life right in the middle of a wedding film color grade, so thanks for that 🤣
Excellent breakdown on the C400 noise and limited latitude situation. In general buyers keep thinking next generation sensors will come to the DR rescue, and it just ain’t so. I will happily keep filming with my FX6 and FX3 for now and hope that the FX9 replacement supposedly coming early next year (not holding my breath) will provide the 6k bump I want to have for some productions. Fingers crossed it will be priced closer to the C400 than the Burano.
@@jakehayden998 We will see about the FX9. Hopefully by the time NAB rolls around. And I will tell you the other deal breaker for me with the C400. I’m probably more fastidious about cards and data than most, but all it takes is one bad card on the wrong shoot and life can get very expensive, or jobs can get scarce, or both. So having symmetrical dual slot recording is a must for me. And yes you can kinda fake it on the Canon for some codecs, but to me a commercial camera should be able to dual slot record at every frame rate and codec. That’s another key reason why I will stick with the Sonys.
You obviously haven't kept up with the R5 C specs. You can now record 8k60 RAW LT (with working AF) from an internal battery, as long as that battery is the LP-E6P, and get 55 minutes recording from a single fully charged battery. That's nearly 1TB of data. And you get 1h26m 8k24 RAW LT from a single, fully charged, LP-E6P pack.
Assuming I haven't kept up with the R5C's specs seems like a bit of a leap, but since you're interested- I'm aware of the 8k60 Raw, but when the market is flocking to cameras that shoot reliable 4k video, I think it was a goofy choice for Canon to come out with an 8k camera with a bunch of related flaws. 4k is only possible in compressed XFAVC, which means you only get C-log 3 (because Canon arbitrarily kneecapped it) so you get even less dynamic range. Your choices with the R5C are- 1. Gigantic 8k Raw files with a super high noise floor 2. Compressed 4k files with significantly kneecapped dynamic range I'm also aware of the update to the batteries, and while I always am a huge fan of brands updated cameras after release, an hour of battery life is still not a long time. For an extended shoot day you would need half a dozen of those things at least to get by. I don't want to sound too critical of a camera, especially if you own said camera, but I think it's important to call out when a company is being stingy like Canon was with the R5C
Thanks for this. Such a well researched, even handed take on the new Canon. I had a similar take after renting the C80 and using it on three jobs alongside my C70s and R6M2s. The C80 image looked closer to the R6M2 than the C70. The C70's image, for me, is the best thing out there for under $10K. Yes, it isn't as sharp but that's what I prefer. I've shot hundreds of hours on C70s now and that DGO sensor is where it's at. I'll sit up and take notice when Canon puts out a 6K DGO sensor that shoots open gate and has comparable AF to the R6M2/R5 M2/R3. Meanwhile my mix of C70s and R6M2s is the best bang for the buck.
The major problem of these high Megapixel, fast readout-sensors is heat. Im not talking about overheating, Im talking about the average operating temperature. The higher the temperature the lower the dynamic range, because electrons have their best operating temeprature between 20 - 35 Celcius. The Venice 2 have "cooling rodes" to prevent the sensor getting too hot.
Happy with my FX6 and FX3. Glad I didn’t do C70 when I purchased new 4K cameras in 2021. C70/C80 is such a weird body shape too. Canon’s C400 demo film looked goofy to me, like it was shot on 5DII and had been up-resed by Topaz. The Sony’s will do fine for me for many more years. Heck you may have just sold me on buying an FX30 to beat up. Hopeful we see an FX9II soon, I want a big shoulder camera, but have no interest in Burano. Also want a Z90V replacement. I’m a sucker for camcorders. I have an XF605 that does okay, but it is a little big for what I want to use it for, a fishing channel, so I even still use an old XF100 from time to time. The Z200 isn’t for me. I’ve seen some super weird and unexplainable artifacts coming out of that camera that I don’t even want to risk. Ideally they would put 4K in the VX2100 body!
Loved the analagy and explanation of why lattitude is important for the scope within the scene, i think this more closesly compares to an FX6 wirh RAW cababilities internally and i feel the price point and features reflect that. I think this offers what most docu shooters would rather have and the utility that comes with it. Not just the 12600 base but the 3200 base iso is the key to this camera that middle range Where 800 isnt enough. but 128 seems over kill. This would be a great a cam for me if i could justify the price tag!
Here is a funny one for you: slomo does not drop to DCI in 120fps, if you set the base fps to 60. It only does it for 30fps and 24fps. Not that it's useful, but it's still funny non the less.
IRE on the C-log 2 is generally 39 to 42 the delta depends of the sensor. Problem of judging dynamic range on nr 2 or 1 you are constantly wrong about step up in tech and algorithm, but first judging the c80 and c400 on raw... well against what exactly? Appart from red every cameras on the price range are compressed codec with even more sometimes noise reduction, aka sony for example. For the tech part yes it is fair to say 16 stop because they are there that you can clean up the image as you wish meaning not the expected DR that you are seing! There is a big problem about noise, it just exist as our eyes can display it, just being fan of a clean look to put an analog filter on it is just taste, we should assume digital sensor images for ones! And the sad thing is that push pull is not relevant about dynamic range, because as the name says it is about in an image the highlight to low light range, not the number of stops under exposed or over exposed it handle (relevant in another way but not this one) You are not an ingenier so you canno't know how to appreciate the result of a push pull test! Like the fact that much filmakers don't know how to use base iso thinking that just the base is good but that's unfortuniatly wrong under the base better low light dynamic range and over better highlightwhich is great with a triple iso sensor where even an arri canno't compete at highter iso! Man 15 unclean stop at 12800 iso in raw is that really bad, and is that really 9.6 at 12600 clean that bad, just prove something can do that ( and that mean that it can do better with controlled noise reduction)
You don’t need to be an “ingenier” as you say to appreciate a simple push pull test, I and a lot of film makers find this is the best way to to effectively test a camera’s real life dynamic range. Unless you are a beginner who drops a lut on there footage and calls it a day you need a camera with the ability to recover the highlights and shadows that most every manufacturer crushes with the included lut (especially canon), that’s just the way cameras work
@jaron_hayden To test Dynamic range you need a chart that Map the sensor entirely AT once like the xyla 21. It get you on a waveform how the sensor handle stops of light. Push pulls don't give any result to analyse, it push a sensor far in a way that can change depending on the situation or the way to achieve overexposed and underexposed image. Push pulls Also are made to put the sensor in situations never existing in Real life. As I said it shaws things but as we are not able to analyze it we can't jump to conclusions. Finally I shouldn't be a discovery that c-log2 to has 16stop with a lot of noise, canon clearly state it that's why the not very good c-log3 exist the same things exist in sony wheee s-log2 and 3 was created together to handle low lights noise an easier way with s-log 2.
@jaron_hayden just try a second to understand what I Say to begin with! It is not a reliable way to state about Dynamic range, never said it doesn't Shaw things just that it is in no way a things you can trust. Push pulls has the same effect as changing ISO value underexposing has the effect of amplifying the signal pushing the dB UP and pushing down with overexposure. So the normal Dynamic range Map when it exist will translate exaclty that without having to test yourself. Also using manual exposure tools like waveform mean that you can have numbers of filmaker seing good and normal as not the manufacturer 0ev So what you Say about manufacturer lut is basic, normal no one should use those. If you wan't to know I do my own conversion lut by camera/ Lens combo made specialy for the way I expose the image. Another problem of push pulls is that you have to shoot Always in very crap conditions of lighting for that to talk to you, like pitch Black because otherwise if you have 6 stop underexposure on few spots if you wtach AT normal speed your footage it is hard top see. It is just use nerds of image that see the noise every day that care about. And I will Say it again canon state that the 16stop in c-log2 is achieve by having a lot of noise, that IS on their c-log3 official page.
There are two problems with relying on the Xyla 21 to test for dynamic range/latitude: 1. They are really expensive so I won't be buying one anytime soon 2. They don't account for shifts in color/loss of saturation/quality of noise pattern (like I say in the video) Maybe you don't find push/pull tests useful, and that's totally fine, but I included one and will continue to include them because I find them very useful. Having a properly exposed middle gray and a skin tone in a frame and then pushing and pulling the exposure around is very helpful for figuring out where the footage is going to have breakpoints
Now, the only thing is missing it's a drop test! Nobody on earth will hire a camera operator if the footage has to be recovered in such a manner. I feel that you are chasing for some alien technology.
finally, someone talking about it... My friend introduce me to video with Canon. I invest so much , it is hard to get back. But, if i could go back in time... i would choose other brand
3 base iso is crazy lol. But budget wise I’d still go fx3 just because it’s good enough to just about do anything . I mean even the fx30 can do a ton . Idk if it’s worth it . For me if I was willing to spend that it be purely for the art of it I’d grab the Komodo x . lol
@ especially here in Africa where gear is 5x as expensive I don’t get why anyone would want to upgrade Everytime a new camera comes out , maybe it’s a Yuotube marketing thing
First of all, I like your videos ! But if I may, it lacks some footage from the camera. Even though you wanna make a point about its dynamic range and latitude, a 22 min video about the image of a camera without a single footage of it besides the test at your desk, through da vinci at 300%, is a bit disappointing. I know it's a lot of work to do (and it's only my take), but it would be less nerdy and more engaging to have even a short sequence of c400 footage at the begining or the end of the video to show what it can do in the real world. So your channel doesn't turn into Gerald's videos (which are useful) with only charts and rubic's cube footage 😅 you have a bigger artistic sensitivity, you should use it ! Keep up the good work !
I really appreciate the feedback and the thoughts, that's really helpful For this video I kind of wanted to do less of a "review" and do more of a focused "essay" about dynamic range and the C400, but I totally get wanting to see more footage. Maybe in the future I'll rent the C80 and do more of a proper review with some footage samples and color grading demos and stuff. Seriously, thank you for the feedback!
@jakehayden998 I package I'm looking at has the camera body, both EF and PL mounts, 3 cards, Arri shoulder mount, cage,3 BP batteries, Expansion unit for 8500. Throw in a Canon 24-70 EF for an extra grand, its a hard deal to pass up.
People need to understand that numbers on a spec sheet don’t actually matter. I DP’d a comedy special recently that we shot on a fleet of C300mk2’s in Clog2, our editor used the “Utopia” LUTS (Arri conversion LUTS) and we’ve had several people ask us if we shot the comedy special on Alexa 35’s. Keep in mind the C300mk2 is an 8 year old camera, and it is Netflix approved. Sure, newer cameras are sharper, and have 1 or 2 stops more dynamic range. At the end of the day it’s not the equipment, it’s how you use it.
That's awesome, I love it when old cameras win the day
where can I watch the special?
Yes, I work on shows and projects that regularly use cameras that are 10 years old. New cameras rarely get you clients. Quality work gets clients.
Unless you're making thousands of dollars with high-end clients, don't waste your money chasing gear. This is completely asinine. A few months later, they will be posting another piece of gear that everyone MUST HAVE.
Especially with a piece of gear this expensive. Cameras like this lose value on the used market super fast
Cameras aren’t about “retaining value” they are about delivering value. This camera does deliver value. Gear by itself doesn’t make the image better but gear and skill do.
Please don’t try to tamp down gear acquisition syndrome, we who buy used gear depend on these people.
@@keylanph right about cameras are just a gear, your work is delivering the value. speaking to that, this camera is delivering value exactly like a lot of 8 - 10 year old cameras you can buy used for 1000 - 3000€, depending on your needs. isnt that a great time we are livin in? just dont worry about the new gear, worry about your quality of work :) - yea, all the influencers will tell you another story, but their main income is to sell the hot shit from the manufacturer.
@noBfilm Camera is the brush for the painting. You can have a great painting with an old brush but you can also have a bad painting with a new brush
Shocking how people just forgot about the significance of a DGO sensor.
100%
They haven’t forgotten. They just know what comes with that. Notably rolling shutter and bad af.
@@SamA-kl6pi Rolling shutter and AF could be improved. Main factor is for sure costs of producing the DGO sensor.
I wouldn't necessarily relate the AF capabilities of C70/C300mk3 as bad. C300mk3 was released about the same time as R5/R6 were, and C300 had the best AF out of all C-line of cams at the time. I highly doubt that AF can't be improved on the DGO sensor, and if it's connected with rolling shutter capabilities - then every single Alexa is a good proof that DGO can have a faster readout. Besides, all recent sensors on the market seem to have a remarkably fast readout in 2024. I suppose they had to pay ARRI licensing fees to be able to use DGO, and it does seem that it didn't work out really well business-wise for them (C300mk3 didn't gain that much traction due to fierce competition on the market).
DGO sensors have high sensor readout speed and don't really help that much.
My C400 out of the box was alarmingly noisy. After I performed an auto black balance the noise seemed to be a fair bit better. Just curious if you ran a ABB by chance for these tests? Great video by the way!
Had me dying when you switched from the FX30 to the C400 hahahaha 😂
I had to cut about 20 extra seconds of laughing. The difference is so funny
@ hahaha that’s hilarious and yes it’s staggering
Crazy. Can you do the same test in comparison to the c70?
canon try not to lie challenge: impossible
I think they don't. They even rate the dynamic range of the C80 lower than the C70 on their website.
@@tgm_previewsYou mean they try to be truthful with the lies!
This is one of the best camera channel on YT.
Thank you so much, that means a lot
100%
I haven't seen such a good cam review in a while. Awesome job, mate! learned a lot. As an FX30 owner I would have loved to see more test shots from that cam but that quick noise comparison with the C400 really blew me away.
As soon as you started talking about Dynamic Range, I suspected you were talking about Shagral’s video. I already loved your content, but seeing that you’re working on growing & perfecting your craft with top notch info just made me love and respect your channel even more.
I subbed to your channel for the immaculate aesthetic that you bring, so glad to see you also run deep into the technical, even more than I had previously imagined. Hope I’m making sense, keep up the fantastic work!
Thank you so much for the kind words, I really really appreciate that. Totally made my day
Hey Jake thanks for your content.
I'm curious here. Did you performed the ABB on C400 before the test?
Holyyyyy moly. I was already shocked but then remembered it's a FF sensor! Thats wild
Yeah it's insane, I was really caught off guard by the noise results, especially from a FF sensor
I think this is your best review yet, diving more into the nerdy stuff while remaining the easiest photo/video content to consume out there
And that’s coming from a canon user
You're my favorite Canon user
C70 in a C400 body, I'd buy it in a heartbeat
It already exists and it's called C300MKIII - the C400's body is shorter but also wider. I all honesty, I didn't see a great deal of difference save for the LCD brace, which is better articulated although very plastic-y.
@@emilianoranzani6049 totally, it's purely a mount limitation. For doco I really love my fuji zoom, I don't like stills lenses, the fuji 18-55 can't adapt to ef.. other EF cine zooms are a lot bigger which is a bit intense for low footprint verite style filming
i got a used s5iix for about $1500 and it's
insane you can get similar image quality (worse frame rates and bad rolling shutter aside) for literally a fraction of the cost
i was hoping a c80 would be a good 'A Cam' to the S5iix but at this point im going to hold off for what lumix announces in 2025
Totally agree, I got the test the S5iix earlier this year and was blown away by the image quality and features vs price. On image quality alone, I would much rather grab an S5iix.
I have my fingers crossed for an EVA-2 announcement this year
ProAV and The Film Guy both do tests comparing the FX6 to the C400 @ 12,800 and the Canon looks way better and way less color cast.
Oh yeah, I watched both videos before making mine for research. On The Film Guy's video the C400 has less noise in the comparison, but that's because they were comparing compressed 10-bit. The FX6 has the ability to turn off noise reduction (which allows more visible noise but the noise looks better) and the C400 has a ton of noise reduction you can't turn off (which takes away some noise, but the noise looks much worse a very digital).
In that case, I would much prefer to have more noise, and then apply the perfect amount of noise reduction in post, rather than my camera applying a ton of noise reduction by default. In the same video the FX6 also does better in the latitude test and holds on to colors longer than the C400 (which basically immediately loses all saturation and color detail)
@@jakehayden998 You are far more technical than I.
But Northwest Camera Co just dropped a C80 video today saying the low light performance on the C80 is better than the FX6. Both shot in 10bit... I guess because the FX6 can't shoot RAW, internally. I dunno? I've watched the compares. I don't see a motive of the previously mentioned sites to favor Canon? They all rent both systems. It's not a bias. I own the C400. I have a bias. I'm not trying to poke holes. I just don't see what you do. The images are better from the Canon C400 & C80.
Yeah I don’t have an fx3/fx6 with me to compare so please take everything I say with a grain of salt, but if I had to guess-
The c80/c400 is probably cleaner than the fx6 at 12800iso in 10-bit, but I suspect it’s because the c80/c400 has a whole bunch of internal noise reduction you can’t turn off. I would love to see a comparison side by side of both cameras shooting in raw
I love the image on the FX30. I love my ZV-E1 too, but the FX30 has a look that is a home run IMO.
R5 firmware solves the overheating. I shoot 8 hours + in 4K 60FPS with 0 heat issues.
Really? I've heard the exact opposite from a lot of people for many years but I'm glad you haven't had issues
I got the C80. And as a documentary filmmaker, it’s perfect for me. I’ve been making youtube videos about it. I have no issues exposing or with dynamic range. However, the only thing I’m still unsure about is autofocus. I’m still finding the right combo of settings for that because it has so many options. Thanks for your perspective
I'm sure it's awesome for documentary stuff, and honestly I think in most circumstances the dynamic range won't be limiting out in the field. I'm glad you're happy with it!
Are you using Canon RF lenses for autofocus? What's wrong with the autofocus? It's just unreliable or slow or what? I didn't have any RF glass to test autofocus while I had the C400 in, so I'm curious.
@@jakehayden998 Ive also heard this on another videos, that autofocus is not reliable for both C80 and 400.
@@luis.hermosilloI’ve been pretty impressed with the auto focus so far. I’m using RF glass.
Some Panasonic cameras also have DR boost sensors. The GH6 was the first one to have it but the G9II and GH7 sensors improved on it to beat most APSC cameras in DR.
I haven’t been able to test those newer Panny cameras that have DR boost, that’s interesting. Hopefully I can get one into the studio soon so I can test that
So what cinema camera in comparison, in your eyes, has a way better dynamic range than this?
you’re the first person who actually did a proper image review of a camera. I love how you zoom in on the Noise so we can see exactly what it’s gonna look like. I wanna know how the camera operates at its worst. that’s what gives me a good idea of how it’s going to work. you can make any camera look good. given the time, proper lighting, proper set design and so on and so forth, but what does it take to break the image that’s important.
Solid video but what happened to the C80?!
Do you think these problems are also there in the R5C? What is the problem of the gamma curve on the R5C? Thanks!
21:30 but in that case why not go with an FX3 or FX6, if even the FX30 gives better results?
Good question. I personally would much rather go with an FX6 versus the C400-
- Much cheaper
- Comparable image quality (even versus C400 Raw)
- Full size XLR
- I like E mount much better (way more options)
But I guess it comes down to ecosystem and taste. There are a lot of people out there that have Canon lenses and want to stay with Canon. Personally I'm still going Sony every time.
@@jakehayden998 C80 is a better comparison. And it’s cheaper than the FX6.
@@jakehayden998 I agree, seems like a better deal all around. Thank you!
On the monitor brightness, take a moment to increase the backlight using the luminance setting. It can be very bright, even in direct sunlight. And this setting it missed by so many ops it's frustrating. (C70 has the same setting btw, but only at a +2)
It's the first thing I always do when I use a Canon camera. I feel so bad for anyone that doesn't know about that setting and tries to use the camera at default brightness
Solid video and super practical takes. Been toying with adding a Canon to my kit next year but TBH wasnt super impressed with the footage from the C400 when I recently used one - but I didnt get a chance to go this deep. Great stuff.
You should carve out the portion on dynamic range and post it as a stand-alone video. Lots of people would love to see such a great explanation!
Thats a fun idea, I considered doing that in the beginning but I got carried away when making this video
@@jakehayden998 You teach things in such a clear and concise way. I think it would be valueable. Thank you so much for giving your knowledge back to the filmmaking community!!
Nice video, and music😉 Thanks! just something that I didn’t get: does the ursa evf work with the c400 through the usb-c?
Your grades always look so good dude, appreciate that you show your process its super helpful
Wow thank you, I really appreciate that
I am not in the market for Canon Cameras but I loved your analysis and explanation of DR or lack of latitude in those Canon Sensors.
Thank you, I really appreciate that
Nice video! Another quirk I noticed with my C400 is that the default sharpening for raw is cranked way up in the files. Weird default choice on Canon’s part. Did you notice this too?
Yeah it's always cranked up to 10 which is really odd and looks kind of funky. My old C200 was the same way
what's the FX30 score in the highlights ?
Nice b-roll using that old box
The blotchiness in the hair wasn’t something just in this latest sensor from Canon. I also saw the same thing in the c70’s DGO sensor that always bothered me.
Great video and great breakdown of your thoughts.
Thank you!
Yeah my C200 used to be pretty blotchy and weird, but I don't remember it being as bad as it is in the C400. It's probably the ugliest looking noise I've ever seen
I use a 5D mkii with magic lantern raw. I always thought the bad digital looking noise might be a result of the custom firmware, but I'm shocked to see that the noise pattern on this camera is identical. As a reminder, the 5D mkii released in 2008.
This is crazy big L for canon on this
Wow, that's interesting. It would be fun to compare the images side by side
@@jakehayden998 It's because of the DCT compression that the camera is using in RAW which is why none of these modern cameras including the RED Komodo qualify as RAW to me. DCT compression is the same "blocky" compression that h.264 and h.265 use. That's why it looks blocky to you
The 5d mark ii in magic lantern mode uses no compression and the mark iii along with the digit 5 magic lantern cameras such as the eos m use the old JPEG lossless 1992 which is similar to .zip file compression. It's compressed and decompressed to its full size with no loss in data, usually about 2:1 compression ratio. Fun fact the old canon dslr cameras can shoot raw video in 14bit and even combine 2x 14bit signals into one massive 16bit file with a massive cost to resolution and usability
This is insane. Out of curiosity, were the raw shots done in RAW LT? That blotchy pattern when you zoom in that far is something I’ve seen with the c500ii in LT. Makes that mode almost unusable for me.
That being said, the latitude performance here is atrocious. I don’t know if canon is doing a lower ADC readout here leading to a higher noise floor compared to a 14 bit or 16 bit readout. They claimed to do a 14 bit readout on the c200 and that image was a noisy mess at times too. So it’s hard to really tell.
I’ve been wanting to get one in to compare with the c500. But honestly, my c500 does not do well with the under exposure either. Most times it’s better to actually shoot at 3200 iso in raw and pull down in post, than shoot at 800 and try to pull it up. I wonder with that in mind if canon is approaching the dynamic range spread differently depending on sensor gain (ISO)
Yeah my test shots were recorded in 6k Raw LT. I would be very curious to see the difference in detail between the 3 different flavors of Raw. The interesting thing is the Blackmagic I was comparing it to was recorded in a lower bitrate Raw codec and it still looked that much better. It's baffling.
Hey Jake, do you think you’ll ever do a video on a Nikon camera? I’m curious how nraw compares to braw.
I was actually throwing around the idea of doing a Z6III video next, I think it's about time I gave Nikon a chance
@ ok, nice. I think the z8/z9 is probably the better camera but I’m just curious what you think about their raw. I’m assuming it will be better than the c400 due to the lack of noise reduction in camera. Keep up the great work
I truly believe this sensor is the R3 sensor. Same exact specs and believe it or not, when you change the ISO settings on the R3, cleans up at 3200, and then shockingly cleans up again at 12,800. Try renting that camera and do your tests on it and I think you’ll be surprised it’s the same.
It’s never been a better time to grab a cheap C70 of someone who jumped ships. I was disappointed they didn’t include the DGO sensor in C80 and C400 and right from the first images that came out the supposed dynamic range just didn’t seem realistic. I’ve seen a lot of OMG’s “C80 destroys C70 with its third iso of 12800” when the 12800 looked so inferior to 6400 on C70. Not even accounting for the fact that with the super 35 sensor I have an extra stop of light bolted to the front of my camera. Great exposé, dude.
Hi thx for ur content wish raw profile did u used hq?
Never really owned any canon cams, came close to the c70 but went with the Komodo. Have recently started to look at the C80 and 400 as really need an all in one doc cam and thought this might be Canon finally shaking things up… but everything I’m seeing is not great news and a lot of old c70 users don’t seem to be falling in love with the image on these newer cams. Thanks so much for running through these tests, incredibly eye opening.
The r5 C is incredible. Just what I needed. The best c70 companion.
Re: "apples and orangutans" Burano comparison. I've used the Burano on a number of projects since the Spring, from small 1-man doc style shoots, to medium sized crewed productions, and it's quickly become my favourite camera to work with. If you have a chance to rent one, I can't recommend it enough. I haven't touched any other Sony since other than an FX3 for gimbal work. Relative to this video, the latitude continues to surprise me. You almost have to try and clip the highlight. And when you do get noise in the shadows, it's so clean it's borderline acceptable, and extremely easy to clean up. I almost liken it to the digital equivalent of Arri's organic noise, and I think if Sony were to ever unlock XOCN-ST or XT the latitude would be nocking on the door of the Alexa 35. It's also quite a bit more expensive than the C80, which again brings up back to the "apples and orangutans".
That's nice info from someone who is actually using the Burano. I'm basing most of what I'm saying on stuff I've seen online so this is very useful
@@jakehayden998 whenever I hear someone mention the slow RS or IR, I know they've never actually used the camera. The RS is a non-issue for 98% of projects and I haven't noticed IR in any shots. That's not to say some people haven't experienced it, but I've personally seen it in my footage. I think the biggest gripe is the cost because it's so unaffordable, especially for the UA-camrs reviewing it. Could the camera have been $5-8k less? 100%. But I think Sony's biggest mistake was sending it to a bunch of UA-camrs, who've never been on a proper set, during the launch. All the reviews were "I can buy X amount of FX3/6s for the same price". Meanwhile, the type of person that this camera was designed for mostly don't even have a UA-cam channel. I will say though, the inconsistent output between HDMI, SDI, format, resolution & codec is annoying as hell, but it looks like that's all being standardized in the next firmware, along with the addition of 240fps and a fast readout 4k full frame crop (95% of full frame). If you ever get the option, I highly recommend renting it. It's not hard to find it for FX9ish pricing.
@@jakehayden998 whenever I hear someone mention the slow RS or IR, I know they've never actually used the camera. The RS is a non-issue for 98% of projects and I haven't noticed IR in any shots. That's not to say some people haven't experienced it, but I've personally seen it in my footage. I think the biggest gripe is the cost because it's so unaffordable, especially for the UA-camrs reviewing it. Could the camera have been $5-8k less? 100%. But I think Sony's biggest mistake was sending it to a bunch of UA-camrs, who've never been on a proper set, during the launch. All the reviews were "I can buy X amount of FX3/6s for the same price". Meanwhile, the type of person that this camera was designed for mostly don't even have a UA-cam channel. I will say though, the inconsistent output between HDMI, SDI, format, resolution & codec is annoying as hell, but it looks like that's all being standardized in the next firmware, along with the addition of 240fps and a fast readout 4k full frame crop (95% of full frame). If you ever get the option, I highly recommend renting it. It's not hard to find it for FX9ish pricing.
The burano has a sensor that's very got very low read noise which is a godsend in itself but it's still a 12bit sensor unfortunately. While latitude tests do show the dynamic range capabilities of the sensor, they're more about the effect of read noise on the image and the leeway that gives you in exposing your image. The fact that multiple sensors with 12bit readout have such different results with the amount of noise being the biggest difference says a lot
@@vhauser416 the camera I use literally has 15-38ms of rolling shutter depending on the resolution I use and it's quite funny yes. I just drop the resolution to get a faster readout, sometimes by as much as 60% to get the result I need
I’ve got a pair of FX30’s because they are cheap and soooooo good! I mostly shoot in well lit environments, but even in low light situations, there is still enough info in shadows to make it look decent. Can’t believe how bad the C400 performs in these standard situations that we often find ourselves in. Great comparison! And loved the genuine laughter too 😂
FX30 is a better value than an FX3, you can get two for the price of one too.
Thank you! I'm glad you're happy with your FX30s. I've loved mine more and more as time has gone on
The C500 Mark II has the same "problem". Here's the "fix": Go to the Custom Picture Menu (CP), then Edit CP file, then Noise Reduction, then make sure that the Frame Correlation is set to 1 (not 0). At least for the C500 Mii this fixes the noise issue - pretty dramatically.
ok so considering the a6700 is the same sensor as the fx30... it cant shoot raw but as we saw with the XAVC codec, it looked like the fx30 had a more natural grain pattern in the noise than canon.... so..... is the a6700 giving you better image quality than the c400 even in a compressed codec? I mean obviously for professional onset work the a6700 doesn't have anywhere near the io or functionality or even redundant recording but.... still...
It is kind of crazy to say, but I would rather have the a6700's image quality over the C400 for sure. You're right, theres a lot of other cool things about the C400, but image-quality-wise I think the a6700 is superior
did you back shading? thx for this
Yeah both cameras were freshly black shaded. Great question
You can compensate for the C80's triple base ISO with controlled lighting. You can compensate for the full frame sensor with a wider lens (to some extent of course). But nothing in the C80 can replace the DGO sensor on the C70. An absolute dynamic range king in the Canon lineup!
I agree, and I’m still buying a c80 based on all the reviews I’ve seen. Mainly because the C80 is more run and gun based camera imo. Idk if there’s any other camera I can throw in a small backpack and get cinema like picture with 24 bit audio and exposure tools built in, decent screen, and internal nd. Literally pull it out, slap on a mic, hit record and record raw video.
@@SamA-kl6pi indeed the C80 is great for run and gun situations. The full frame sensor is really something when it comes to using Canon's RF lenses' full potential. Still for the image purists, C70's DGO sensor is the closest that 4 thousand bucks will get you to an Alexa :)
When it comes to imatest DR, why not care more about the SNR of 2 result to judge the camera capability ?
Great question. SNR of 2 is a good measure of "clean stops" or stops that are totally usable. I use SNR of 1 as a measure of "total dynamic range" since in theory you can use noise reduction and dig out some extra stops even if they are a bit noisy to begin with.
How would you recommend Ursa Cine 12K LF for a person who's aim is just image quality?
I haven't gotten one in the studio for review yet, but if I had to guess, yes. The Ursa Cine 12k LF is on another level as far as image quality is concerned versus basically anything under $30k. The dynamic range ( even without highlight recovery) is really really close to Alexa level, and the resolutions/framerates/readout speeds are absolutely insane. It's kind of the ultimate camera
mostly excited this is gonna cut the cost of the c70
That is true, hopefully it does cause the C70 and C300 mkiii to drop in price
I’ve had a brief brush with the Canon ecosystem, but I’m Sony pretty much 100%. I like the entire cinema line that works together from FX30 to Venice. Canon I think is finally going to start rolling a consistent line up from beginning to end, but it’s too late for me. The C400 is the most exciting camera I’ve seen from them in 15 years. It looks really good!
Totally agreed. If Canon is able to bring both their prices and their noise floors down, I think they could be a force to be reckoned with in the next few years
13:08 - That noise pattern not only looks processed but I get a better looking noise profile out of the Panasonic GH5 II ($1,200 camera) with noise reduction reduced as much as possible in camera and recording to a H.265 LongGOP video format to v30 SD cards. Be aware that all Lumix cameras have a bad noise profile on the first frame of a video clip; every frame after is excellent. Canon doesn't know what they are doing here, are crippling it on purpose, or just don't care. I'm not joking ... rent a Panasonic GH5 II or find one in your local dumpster and light the scene similarly. Even with the ISO at 800, the noise profile will still look more like a per pixel sort of noise ... in H.265 LongGOP ... close in visual quality to All-I and with no apparent macro blocking / lossy compression artifacts. ISO less than 800 on a GH5 II will absolutely look better than that ... this is atrocious. Just turn down noise reduction in the profile as much as it will go (-5.0) on the GH5 II to see. Also, Neat Video will give a better result than Resolve Studio's built in temporal and spatial denoising ... but unfortunately Neat Video is CPU only and so is like 5x to 10x slower than Resolve's GPU accelerated setup. So on a quick turn around project, I use Resolve's GPU denoising. But when I have some time (like overnight), I'll have my first node (a serial with Neat Video applied) in my Power Grade render out to the Node Cache to ProRes / DNxHR for a few hours as I sleep for much better results. But again ... regardless ... your eyes aren't fooling you. That noise profile is atrocious. Neat Video will not do well with that ... thanks for the video. My only complaint? You should have had it auto post on Christmas morning. Haha. Subscriber earned!
I love your reviews!
Thank you! I appreciate that
Hard to know where your test method went wrong without being there or at least seeing the files for oneself.
What makes you say my testing method went wrong?
Thank you so much for this video! That explains a lot why I am not very satisfied with my C400 which I bought as an upgrade to my Pocket 6K. I should have waited I guess, its hard to get clean shadows in sunlight sometimes, in those high dynamic range scenes.
Such a banger of a video! Spot on!
Lol comparing the FX3 xfavc at 3 stops under and the c400 in raw at 3 stops under is ridiculous - of course a raw image will have more noise. C400 noise reduction parameters can be finely adjusted or turned off unlike many sony cameras.
The comparison wasn’t to see which image had more noise, the comparison was to see which image had a better looking noise pattern. Somehow in a compressed format the fx30 is able to a beautiful and granular noise pattern and the c400 in Raw (which should be the least processed) has a noise pattern that looks processed and digital.
The whole point is the c400 weirdly has a much worse looking noise pattern vs the fx30 (and typically Sony cameras are the ones that famously have internal noise reduction that you can’t turn off). I would much rather have the fx30s noise pattern vs the c400s. In addition the fx30 has a lower noise floor, but that wasn’t really the point
@@jakehayden998 C400 xfavc with noise reduction on would be a relevant comparison.
That was the point through. Putting noise reduction on would make the noise pattern look even worse and more processed
@@jakehayden998 You can fine tune the noise reduction on the C400 or turn it off.
FX30 owner here, so happy my little camera compares favourably, but surely you can only compare the quality of the noise between it and the C400 when DR is turned on? God knows what RAW looks like from an FX30. Anyway, I’m glad the DR myths are finally being debunked on YT - most cameras are stuck around 12 stops. Even the famous 15 stops on an FX3 is BS, but the marketing obviously works.
Key takeaways:
1) Sony's sensor fabrication is better than Canon's hence better intrinsic SNR
2) Sony uses a better H.265 based XAVC-HS codec that holds up lot's of data in limited bitrate compared to Canon's AVC and internal Debayered/compressed RAW
3) Sony has much better noise Profiling that uses much more Temporal NR than Spatial NR hence even noise texture overall.
I also wonder if Canon has issues with sensor heat. It seems like all of their cameras have the tendency to overheat, or at least run super hot. The C400s fan was the loudest I have ever heard in my life, it sounded like my PS5
Did you by any chance come across my comments? 💀 It's good to feel seen 😂
I see you, my friend
@@jakehayden998 You should've seen my face when I realized it. Almost flew out of my chair lmao
Check out the Sony A1 latitude. It's insanely clean in the shadows.
I haven't been able to test out the A1 yet, but from what I can tell a lot of the newer Sony cameras have really great noise patterns/latitude. The A9iii is supposed to be crazy good
@@jakehayden998 It's the same sensor as the Burano, so besides not having RAW it should have similar latitude.
@@AstralxProd the sensors are mostly likely the same, but the internal processors and colour science are much different. In XAVC they could potentially be similar in latitude, though you'd see different results in what was pulled/pushed. However XOCN- LT will give you significantly more information to push/pull from. 10 bits shades are in the hundreds while 16 bit is in the 10s of thousands. It also see many millions more colours. I've used the Burano on a few projects now and the image is incredible, and it's such an easy camera to operate.
can you do a review on the sony f55?
I would love to, it might be kind of hard to rent since it's a bit older but I would love to do a review at some point
Earned a subscription for me! Unbelievable video I don't like the crop CRT look but hey that your artistic choice lol
I love the explanation of dynamic range and signal to noise ratio! Have you seen Shagrals video on it? It seems like you have he goes DEEP. Good stuff.
Yeah I have seen Shagral's video, pretty good stuff. Glad you liked all the DR stuff. I was worried people wouldn't connect with it
@@jakehayden998 I like your take on it! Simplified and understandable. Also made me question my life right in the middle of a wedding film color grade, so thanks for that 🤣
Am just interested in the high end grear to see if it trickles down to more affordable like the R7 lol
If the R7ii has C-LOG2 🤯
C log 3 is ok, but h.265 sucks for editing and transcoding is a pain.
Yeah it's annoying they haven't put clog 2 in all cameras yet, I'm hoping they put it in the R7ii
Excellent breakdown on the C400 noise and limited latitude situation. In general buyers keep thinking next generation sensors will come to the DR rescue, and it just ain’t so. I will happily keep filming with my FX6 and FX3 for now and hope that the FX9 replacement supposedly coming early next year (not holding my breath) will provide the 6k bump I want to have for some productions. Fingers crossed it will be priced closer to the C400 than the Burano.
I've heard rumors of an FX9 successor coming soon too. I have my fingers crossed for the same price as it is now
@@jakehayden998 We will see about the FX9. Hopefully by the time NAB rolls around. And I will tell you the other deal breaker for me with the C400. I’m probably more fastidious about cards and data than most, but all it takes is one bad card on the wrong shoot and life can get very expensive, or jobs can get scarce, or both. So having symmetrical dual slot recording is a must for me. And yes you can kinda fake it on the Canon for some codecs, but to me a commercial camera should be able to dual slot record at every frame rate and codec. That’s another key reason why I will stick with the Sonys.
You obviously haven't kept up with the R5 C specs. You can now record 8k60 RAW LT (with working AF) from an internal battery, as long as that battery is the LP-E6P, and get 55 minutes recording from a single fully charged battery. That's nearly 1TB of data. And you get 1h26m 8k24 RAW LT from a single, fully charged, LP-E6P pack.
Assuming I haven't kept up with the R5C's specs seems like a bit of a leap, but since you're interested-
I'm aware of the 8k60 Raw, but when the market is flocking to cameras that shoot reliable 4k video, I think it was a goofy choice for Canon to come out with an 8k camera with a bunch of related flaws. 4k is only possible in compressed XFAVC, which means you only get C-log 3 (because Canon arbitrarily kneecapped it) so you get even less dynamic range. Your choices with the R5C are-
1. Gigantic 8k Raw files with a super high noise floor
2. Compressed 4k files with significantly kneecapped dynamic range
I'm also aware of the update to the batteries, and while I always am a huge fan of brands updated cameras after release, an hour of battery life is still not a long time. For an extended shoot day you would need half a dozen of those things at least to get by.
I don't want to sound too critical of a camera, especially if you own said camera, but I think it's important to call out when a company is being stingy like Canon was with the R5C
Thanks for this. Such a well researched, even handed take on the new Canon.
I had a similar take after renting the C80 and using it on three jobs alongside my C70s and R6M2s. The C80 image looked closer to the R6M2 than the C70. The C70's image, for me, is the best thing out there for under $10K. Yes, it isn't as sharp but that's what I prefer. I've shot hundreds of hours on C70s now and that DGO sensor is where it's at.
I'll sit up and take notice when Canon puts out a 6K DGO sensor that shoots open gate and has comparable AF to the R6M2/R5 M2/R3. Meanwhile my mix of C70s and R6M2s is the best bang for the buck.
Bmcc 6k FF !!!
Yeah the C80 feels pretty far off from the nice organic image coming out of the C70. I'd be really excited to see Canon do a full frame DGO too
finally someone’s calling Canon out
Dynamic is more important than resolution...I hope the next version of the FX30 has 2 more stops of dynamic range even at 4k....
Couldn't agree more. I wish most cameras were 4k instead of 6k/8k and we got slightly better DR or readout instead
The major problem of these high Megapixel, fast readout-sensors is heat. Im not talking about overheating, Im talking about the average operating temperature. The higher the temperature the lower the dynamic range, because electrons have their best operating temeprature between 20 - 35 Celcius. The Venice 2 have "cooling rodes" to prevent the sensor getting too hot.
Yeah I suspect Canon is having heat issues too. The fan on the C400 was by far the loudest I have ever heard in my life
@ All CMOS sensors with such a fast read out have it. Sony, Nikon and Canon too.
The noise and performance of this camera is HORRIBLE. This, I am afraid to say, makes my BMPCC 4k look outstanding.
Very interesting indeed! Thanks for the in-depth examples. Just found ya. Subbed buddy.
Thanks a bunch, I really appreciate that! Thanks for the sub
Great video, I wanted this camera to be awesome but I guess I'm 'stuck' with FX3.
The FX3 is so awesome, I miss mine all the time
@@jakehayden998 What is your A cam if I may know?
@@EndlessEnergyI'm pretty sure it's the BM ursa 4.6k g2
@@kubazawisza6066 Thanks man! 🙌
It was an URSA 4.6k G2 up until about 2 weeks ago and I switched to an URSA Broadcast G2. I’ll have a video on it pretty soon
Happy with my FX6 and FX3. Glad I didn’t do C70 when I purchased new 4K cameras in 2021. C70/C80 is such a weird body shape too. Canon’s C400 demo film looked goofy to me, like it was shot on 5DII and had been up-resed by Topaz. The Sony’s will do fine for me for many more years. Heck you may have just sold me on buying an FX30 to beat up. Hopeful we see an FX9II soon, I want a big shoulder camera, but have no interest in Burano. Also want a Z90V replacement. I’m a sucker for camcorders. I have an XF605 that does okay, but it is a little big for what I want to use it for, a fishing channel, so I even still use an old XF100 from time to time. The Z200 isn’t for me. I’ve seen some super weird and unexplainable artifacts coming out of that camera that I don’t even want to risk. Ideally they would put 4K in the VX2100 body!
You’re not wrong, that noise pattern on the Canon is just weird, Prores raw on the Sonys looked more natural, you should do one on Nraw
I would love to test out NRaw. I've never gotten to test a Nikon camera but I'm looking into getting a Z6iii into the studio soon
Loved the analagy and explanation of why lattitude is important for the scope within the scene, i think this more closesly compares to an FX6 wirh RAW cababilities internally and i feel the price point and features reflect that. I think this offers what most docu shooters would rather have and the utility that comes with it. Not just the 12600 base but the 3200 base iso is the key to this camera that middle range Where 800 isnt enough. but 128 seems over kill. This would be a great a cam for me if i could justify the price tag!
I'm glad the analogy was helpful!
I totally agree, it almost perfectly resembles a Canon FX6 with Raw
Here is a funny one for you: slomo does not drop to DCI in 120fps, if you set the base fps to 60. It only does it for 30fps and 24fps. Not that it's useful, but it's still funny non the less.
IRE on the C-log 2 is generally 39 to 42 the delta depends of the sensor.
Problem of judging dynamic range on nr 2 or 1 you are constantly wrong about step up in tech and algorithm, but first judging the c80 and c400 on raw... well against what exactly?
Appart from red every cameras on the price range are compressed codec with even more sometimes noise reduction, aka sony for example.
For the tech part yes it is fair to say 16 stop because they are there that you can clean up the image as you wish meaning not the expected DR that you are seing!
There is a big problem about noise, it just exist as our eyes can display it, just being fan of a clean look to put an analog filter on it is just taste, we should assume digital sensor images for ones!
And the sad thing is that push pull is not relevant about dynamic range, because as the name says it is about in an image the highlight to low light range, not the number of stops under exposed or over exposed it handle (relevant in another way but not this one)
You are not an ingenier so you canno't know how to appreciate the result of a push pull test!
Like the fact that much filmakers don't know how to use base iso thinking that just the base is good but that's unfortuniatly wrong under the base better low light dynamic range and over better highlightwhich is great with a triple iso sensor where even an arri canno't compete at highter iso!
Man 15 unclean stop at 12800 iso in raw is that really bad, and is that really 9.6 at 12600 clean that bad, just prove something can do that ( and that mean that it can do better with controlled noise reduction)
You don’t need to be an “ingenier” as you say to appreciate a simple push pull test, I and a lot of film makers find this is the best way to to effectively test a camera’s real life dynamic range. Unless you are a beginner who drops a lut on there footage and calls it a day you need a camera with the ability to recover the highlights and shadows that most every manufacturer crushes with the included lut (especially canon), that’s just the way cameras work
How do you test Dynamic range? Or do you not care and not like when other people do?
@jaron_hayden To test Dynamic range you need a chart that Map the sensor entirely AT once like the xyla 21. It get you on a waveform how the sensor handle stops of light.
Push pulls don't give any result to analyse, it push a sensor far in a way that can change depending on the situation or the way to achieve overexposed and underexposed image.
Push pulls Also are made to put the sensor in situations never existing in Real life.
As I said it shaws things but as we are not able to analyze it we can't jump to conclusions.
Finally I shouldn't be a discovery that c-log2 to has 16stop with a lot of noise, canon clearly state it that's why the not very good c-log3 exist the same things exist in sony wheee s-log2 and 3 was created together to handle low lights noise an easier way with s-log 2.
@jaron_hayden just try a second to understand what I Say to begin with!
It is not a reliable way to state about Dynamic range, never said it doesn't Shaw things just that it is in no way a things you can trust.
Push pulls has the same effect as changing ISO value underexposing has the effect of amplifying the signal pushing the dB UP and pushing down with overexposure. So the normal Dynamic range Map when it exist will translate exaclty that without having to test yourself.
Also using manual exposure tools like waveform mean that you can have numbers of filmaker seing good and normal as not the manufacturer 0ev
So what you Say about manufacturer lut is basic, normal no one should use those.
If you wan't to know I do my own conversion lut by camera/ Lens combo made specialy for the way I expose the image.
Another problem of push pulls is that you have to shoot Always in very crap conditions of lighting for that to talk to you, like pitch Black because otherwise if you have 6 stop underexposure on few spots if you wtach AT normal speed your footage it is hard top see.
It is just use nerds of image that see the noise every day that care about.
And I will Say it again canon state that the 16stop in c-log2 is achieve by having a lot of noise, that IS on their c-log3 official page.
There are two problems with relying on the Xyla 21 to test for dynamic range/latitude:
1. They are really expensive so I won't be buying one anytime soon
2. They don't account for shifts in color/loss of saturation/quality of noise pattern (like I say in the video)
Maybe you don't find push/pull tests useful, and that's totally fine, but I included one and will continue to include them because I find them very useful. Having a properly exposed middle gray and a skin tone in a frame and then pushing and pulling the exposure around is very helpful for figuring out where the footage is going to have breakpoints
I’d love to see a video of the C80 compared to the XH2S
Yeah that would be really interesting. I would really like to get a XH2S in the studio again and test latitude against some other stuff
Now, the only thing is missing it's a drop test! Nobody on earth will hire a camera operator if the footage has to be recovered in such a manner. I feel that you are chasing for some alien technology.
Love you bro
love you too man
we need a video on the goated a7IV
Yeah I don't know how I haven't gotten my hands on one yet. Hopefully soon
finally, someone talking about it... My friend introduce me to video with Canon. I invest so much , it is hard to get back. But, if i could go back in time... i would choose other brand
What Canon cameras/lenses did you end up going with?
@@jakehayden998 RF 24-70, EF 24-70, RF 35, EF 100-200 and I got an RF Artisan 50mm for my Canon R7. I shoot a lot of podcast and conferences
Another certified banger.
I see you've seen the dynamic range video of Shagral...
Yeah for sure, that video helped me brush up on ADCs last week
Sorry for not listening to your video 90% of the time. Did I understand right that DGO sensor has better dynamic range and worse noise?
3 base iso is crazy lol. But budget wise I’d still go fx3 just because it’s good enough to just about do anything . I mean even the fx30 can do a ton . Idk if it’s worth it . For me if I was willing to spend that it be purely for the art of it I’d grab the Komodo x . lol
Yeah I agree. There's too many other options at this price point with better image quality
is it that they're huge?
It's simply absurd to have sd card slots in a camera like this. Even the R1 has dual cfexpress.
Yeah I definitely wish they would have done dual cf and included redundant recording in all codecs
i honestly find all these new camera releases to be way too pricey for what they offer , after $3500 alot of shortcuts shouldnt be acceptable
I very much agree. It feels like returns start to diminish significantly at the $3-$4k mark while the $2k-$3k cameras have so much bang for the buck
@ especially here in Africa where gear is 5x as expensive I don’t get why anyone would want to upgrade Everytime a new camera comes out , maybe it’s a Yuotube marketing thing
Ah everyone has seen the Shagral videos, nice
@ganondorf772 Are they on UA-cam? I couldn’t find by searching Shangral
First of all, I like your videos ! But if I may, it lacks some footage from the camera. Even though you wanna make a point about its dynamic range and latitude, a 22 min video about the image of a camera without a single footage of it besides the test at your desk, through da vinci at 300%, is a bit disappointing. I know it's a lot of work to do (and it's only my take), but it would be less nerdy and more engaging to have even a short sequence of c400 footage at the begining or the end of the video to show what it can do in the real world. So your channel doesn't turn into Gerald's videos (which are useful) with only charts and rubic's cube footage 😅 you have a bigger artistic sensitivity, you should use it !
Keep up the good work !
I really appreciate the feedback and the thoughts, that's really helpful
For this video I kind of wanted to do less of a "review" and do more of a focused "essay" about dynamic range and the C400, but I totally get wanting to see more footage. Maybe in the future I'll rent the C80 and do more of a proper review with some footage samples and color grading demos and stuff.
Seriously, thank you for the feedback!
I might stick with a used C500 Mark II.
I didn't realize the price had come down so much on C500 Mark IIs, I'm seeing some as low as $6k on eBay. That's pretty interesting
@jakehayden998 I package I'm looking at has the camera body, both EF and PL mounts, 3 cards, Arri shoulder mount, cage,3 BP batteries, Expansion unit for 8500. Throw in a Canon 24-70 EF for an extra grand, its a hard deal to pass up.
19:23 Joker 3: Folie à Bruit
Nice Sleep Token drop
I see how Canon achieves their colors... they push magenta noise in the shadows!
Thanks for doing the tests. Yeah, not the greatest low light performance. Oh well...
There's always next time
dont explain dynamic range on a cinema camera video. please for the love of god
Uh why? What do you have against dynamic range?
Expose properøy next time, so tired of click baits. Allways 1-2 stops ofer and introduce enough lights, if not, expect noise.