"EE-so" is the pronunciation in most non-English speaking countries, and the teacher I learned the term from was not a native English speaker. This is an older video, and because I get SOOOO many comments on the pronunciation, I make a point of saying "eye-so" now. Check ISO's official About page for an explanation of why ISO isn't an acronym (it's derived from the Greek word isos). Or, just google, "ISO is not an acronym". See, you can still learn a thing or two, even after 38 years!
And did Paul Gauguin happen to abandon painting because of the critics? Don't waste yourself on small things. I used to say "ISO" with an old Portuguese accent.
Wow, most of the comments are about the correct way of pronouncing the ISO? Thank you for the video! I'm about to upgrade my 600D to a budget full frame body. The 5D mk ii still kicks ass.
Glad to hear it. My Mark III is in the shop so I've returned to using my Mark II. Image quality is the same, and it's getting the job done, even for focusing on moving subjects. If you don't *really* need the improved autofocus, get the Mark II and save the extra $1500 for lenses.
Depends what you're shooting and how you're using it. If you do get the 5D2, you'd also need to upgrade to full-frame lenses, which are bigger, heavier, and more expensive.
I just checked, and I can see. If you want a second opinion on the image quality, check DxO. Their objective tests determine that the 5D Mark III's image quality is 3% better than the 5D Mark IIs.
The name of the group is International Organization for Standardization. Look it up on Wikipedia: "The organization says that ISO is not an acronym or initialism for the organization's full name in either official language; rather, recognizing that its initials would be different in different languages, it adopted ISO, based on the Greek word isos (ἴσος, meaning equal), as the universal short form of its name."
Well, it does sort of make sense, because if they used an acronym it would need to be different in different languages. So, they chose a three-letter word. The part that doesn't make sense is why they capitalize it, making everyone think it's an acronym when it's just a word. However, the organization is quite clear that ISO is not an acronym and thus should never be spelled out.
Yes, and that's part of the risk of doing a real-world test rather than a lab test. Check the links in the description for RAW files done in a lab environment. Draw your own conclusions, but mine are that the 5D3 is equal to the 5D2 until you get above ISO 50,000.
Thanks! The 24-105 f/4 is way more flexible than the 24-70, mostly because it has IS. Still, I often want the f/2.8 speed, and I've been waiting for for years for a 24-70 f/2.8 with IS. Tamron has one, but it doesn't have great image quality. If you're a pro and you don't need the IS, get the 24-70. The Mark II isn't out yet, but it's priced at $2300, which is more than twice the 24-105. For everyone who isn't a pro, or pros who do low-light hand-holding, get the 24-105 and a 70-200 f2.8.
BTW, recently learned that ISO is not an acronym. It should be pronounced "eye-so". It's a form of the Greek word "Isos" which means equality, and is the same as the prefix for many words such as "isometric."
Yes, but it's a challenge because the live view is too dark. I describe night focusing techniques in Chapter 10 of Stunning Digital Photography (links in the description) but on the Mk II it amounts to estimation and then trial-and-error.
Often flying birds are easier than moving subjects on the ground because they have a clean background (the sky). With ground subjects cameras tend to latch onto the background. Sandi's light colored fur might not have been as contrasty as a bird against a blue sky. Also, as subjects get closer to you the camera has to do more work to focus, so it's easier to focus on farther away moving subjects.
I've had dozens of people make this comment to me, and I make them all the same offer: show me the raw files where the 5D3 is better than the 5D2. Not only will I do a follow-up, but I will personally be thrilled to discover that the 5D3 has substantially better IQ, because I plunked down my own cash for the camera and IQ is extremely important to me. I definitely only have bias toward the 5D3, but I need some evidence.
They'd be about the same, and I like the 6D a touch better, so I'd go with that. Note that used 5D Mark IIs are selling for $1200, so consider that, as well...
Bongo is right that the focus system is the weak link on the 5D2. It's absolutely fine for everything except action shots, and if you use the center autofocus point, you can do an OK job with action.
Thanks! I'm definitely going to check out the D800--the higher MP would be really useful. I do love the 5D3. My only problem is that Canon is marketing it as having a sensor that dramatically improves low-light performance, and my testing shows no improvement whatsoever.
I have a tour of my studio that describes the equipment, and some outdoor lighting videos with off-camera flash, but otherwise, no. We're planning to do the studio videos this winter while we're shut in from the cold.
You're definitely not the only one :D. We alternated filming with the 5D Mark II and 5D Mark III, because those are the only two bodies we have that can film, so while I shot with one, we filmed with the other. Most notably, the night scene was filmed with the 5D Mark III. I can tell a big difference between the cameras at night, but I don't really notice much difference when the lighting is good. However, other people more picky about video quality than I say the 5D3 is much better at video.
I was quite surprised too, and not in a pleasant way! No, I'm working with raw files, which aren't processed by any in-camera software. I'm afraid we're stuck with it; firmware isn't likely to improve it. It still does great work, though, and I love the 5D3 more every day I use it.
You're looking for tethering software. I use Adobe Lightroom (which is necessary software, anyway). Just connect your camera to your Mac with a USB cable.
The Nikon D4 and the Canon 1D are specialized professional bodies. They're big, heavy tanks. They're built without any compromises (outside of weight, cost, and often resolution). So, I use the smaller full-frame body because I travel a great deal (it's much lighter than the 5D3/D4) and the 5D3 is durable enough for the abuse I give it. I use a Canon instead of Nikon for no good reason other than I've always used Canon equipment and it's expensive to switch systems.
I zoom to 4:1 to exaggerate the noise. It's a technique I learned from the stock photo agencies. With commercial work, this allows me to identify areas that need to be de-noised. Even at ISO 100, I always need to remove noise in solid areas of primary colors (R/G/B) and in shadows. Just because you can't see noise at 1:1 on your monitor doesn't mean it's not significant; it shows up in large prints and on sharp monitors, and stock agencies will reject it. Working on the ISO pronunciation.
Check the other comments for info about the first test. There's no review or test that shows the 5D Mark III has noticeably better raw image quality than the 5D Mark II. DXO tests show an improvement of 2-3%. Image quality is essentially unchanged.
This is one area where the 5D2 and 5D3 differ. The 5D3 works great with all autofocus points on (assuming there's nothing in the background it can focus on--you want to be shooting up into the sky or just use the center focus point). With the 5D2, only the center focus point is cross-type. Basically, only the center focusing point on the 5D2 is fast enough to even hope to keep up with a moving subject.
Yes, everyone says it but it only applies to JPG files processed in-camera. People who care shoot RAW. I've shot over 15,000 frames with the 5D3, and more than 100,000 on my 5D2. I care about the noise because I have to fix it (even at ISO 100) or my commercial clients reject the photos. Re: the first set of pictures being darker, read the other comments. Check the other examples and the lab files linked in the description for a controlled test. Still waiting for your RAW files.
Raw image quality is the same, but the 5D3 is functionally better at everything, it's just more expensive. If you're doing still portraits and landscapes, go with the 5D2. If you're doing weddings, sports, wildlife, or night photography, go with the 5D3
I didn't have a flash. Note that while the 5D3 is definitely better in the dark, and the focus works in darker environments, there are definitely still going to be times when it's too dark for it to focus.
Sorry about that! Sounds like something came loose. I'm sure it can be repaired. Just send it to Canon for repairs--I've had them do several repairs on different bodies and lenses and it always works out just fine.
Good call. Especially since 5D2s are available now used for about $1100, I very rarely recommend the 5D3 to people. It really only makes sense for wedding or sports photographers... Everyone else should spend the extra cash on lenses and lighting. Even wildlife photographers, who need good AF, are usually better off with a 7D and the higher pixel density.
The light leak issue isn't really a problem. It would only appear if you had the LCD light on while taking a long exposure (say 20 seconds) at night. Pictures from the 3 look better if you shoot JPG. However, if you care about image quality at all, you should be shooting RAW, and the RAW files are basically indistinguishable. DxO's objective tests put the 5D3 at 3% better than the 5D2, which nobody would ever notice. Anyway, there are many good reasons to get a 5D3, but not image quality.
I don't know what type of video you plan to do or what your total budget is, but for almost everyone I would recommend the 5D2. In fact, I'd push you to buy a used one. Put the money you save towards some nice mics (I love my Sennheiser wireless lav mic), lenses, lighting, tripods (watch my tripod video), and software. I travel with both the 5D2 and 5D3, and the only time I make a point of using the 5D3 for video is at night, but I think Magic Lantern might negate even that benefit now.
The 5D3 has a great AF system; every reviewer has agreed on this one point. No system is perfect with moving subjects, however. I can't tell you whether it's better than the 1D3, but given that it's five years newer, I imagine it is.
Yeah, I'd love medium format for the studio, but they don't support higher ISO, telephoto lenses, or even good autofocus, besides being way bigger to travel with. Anyway, you're right, I dream of the Hassy.
Yeah, it's fantastic. It's one of the sharpest zooms ever made, and at f/2.8, every headshot has a beautiful blur. Check my other portrait videos for some examples. It's also $2,300, though, and if you're not doing commercial work where people literally investigate every pixel, you can get by with the Sigma version of the lens ($1,300 at Amazon) or even grab a used Mark I version of the Canon on ebay.
It wasn't intentional. To the best of my knowledge, the lighting didn't change. The two bodies just seemed to meter the situation differently. Incidentally, now that myself and other people have done controlled lab tests, the overall results hold up: the 5D3's raw image quality is indistinguishable from the 5D2's (within 2-3%) at all reasonable ISOs (say, ISO 3200 and below).
I shot in raw and processed the images in Adobe Lightroom 4.1 RC. You're right, the claimed image quality improvement applies only to the in-camera JPEG processing, so it doesn't impact those of us who shoot raw (which is everyone who really needs good image quality). The AF system will really help your low-light work, at least compared to the 5D2. I did try the 5D3 video at night, but it didn't turn out well, as you can see--but it is probably still be better than the 5D2.
At the same price, the 6D. However, I push people towards used 5D Mark II's lately, because they're available so inexpensively. There are a ton of 5D Mark iI's on ebay for $1300, whereas the 6D is going for $2100. For that price, I'd grab the Mark II and have $800 for an awesome lens.
For me, upgrading to a UDMA 7 card allowed me to take 33 shots without it slowing down. With an SD card, I was only able to get 13. Plus, after the buffer filled up, it slowed down much less than with the SD card. My SD card is a class 10.
Decision made!! I currently own a 5DMKii and a T7i. I was literally about to sell both and get a 5DMKiii but decided to do a bit a research. Your comparisons made it easy for me to keep what I have and avoid the hassle of selling. Both cams serve me well based on what I shoot. PLUS I can keep the clicks down by splitting the workload :-D! Thanks for the very informative review! Off topic, the T7i is an excellent choice for low light shooting as well..
Not sure what other gear you have, but unless you have an unlimited budget, for weddings, I'd get the 5D Mark II used and a 70-200 f/2.8. Yes, the 5D3's AF is helpful during weddings in low-light; it'll snap a hair faster. But the 5D2 always got the job done for me, and it still does. You can buy 3 used 5D2s for the price of one 5D3. The higher FPS is useful for wildlife and sports, but for those, I would stay with the 7D because the crop helps. Why not keep the 7D as a backup to the 5D2?
Sure, it depends on how much contrast there is in the scene, how fast the subject is moving, and how much light you have. I wasn't trying to define how many shots the 5D2 misses, just trying to determine how much of an improvement the 5D3 is.
Actually, that's an awesome idea. I do have a 7D for wildlife and full-frame bodies for everything else... but you make a compelling argument against the 5D3. Your solution also gives you a backup body if one fails.
The video is measurably better, according to people who analyze these sorts of things, but I haven't found that it's noticeably better for the types of video that I do (at least, below ISO 1600). Particularly if you install Magic Lantern on your 5D2, the 5D2 is still great for video.
Agreed, and a few people have mentioned this. I shot them in the same minute, but maybe the sun went behind a cloud. It only impacts the one test, though, and the overall conclusion was that at all but the highest ISOs (50k and above) the 5D2 and 5D3 have similar image quality. Check the description of the video for links to the "high ISO, low-light lab tests" that were done under controlled lighting conditions and confirm my conclusion (but feel free to judge for yourself).
Definitely get the Mark II. My Mark III has been in the shop for the last month, so I'm using the Mark II again, and it's getting the job done just fine. Spend the extra ~1400 on a nice lens and flash. Maybe even check ebay for a good deal on a used Mark II from someone who sold it to upgrade :)
Yeah, you'll notice I took the 5D2's win in that area with a grain of salt. I shot the images back-to-back, but you know how natural light is: perhaps the sun went behind a cloud or something. I'll repeat the low-light portrait test today under artificial lights and post a link. Do note that I tested image quality with wildlife, landscapes, and night photography without any lighting variations and the image quality seemed identical.
I agree at ISO 51,000+ the 5D3 is (slightly) noticeably better, and DXO says the improvement is about 15%, but at lower ISOs (every ISO anyone would want to use) there's no noticeable improvement, and DXO supports that. Is that what you're finding? Would you send me the raw files showing the improved performance from the 5D3? If you read through the comments here, I've had a lot of people saying the 5D3 is far better, but nobody has actually submitted proof, so I'm excited to see some.
Just wanted to follow-up. Do you have a raw comparison? I've had dozens of people tell me the 5D3 is way better and that I'm wrong, but literally zero people have shown me an actual example.
The 5D3 has much better low-light video performance, and it has a headphone jack for monitoring the audio. If that's important to you, get the 5D3, otherwise get the 5D2.
This has been addressed over and over again in the comments. This video was the real-world subjective test. I also did a controlled test; check the description for the link to the lab files. The controlled test supports the findings in the real-world test, but you can draw your own conclusions. Everyone is open to your point of view, but if you want anyone to take you seriously, you're going to need to produce some raw files supporting your claims.
Yeah, lenses hold their value better than bodies. I don't do primes except for telephoto and macro lenses. Zoom lenses rule, and the Canon zooms (especially the Ls) are as sharp as primes.
OK, that makes sense. I think my previous recommendations work. If your funds are unlimited, the 5D Mark III's AF would be useful, but the 5D2 always worked for me. Also, you'll need a good bounce flash if you don't have one. Save some cash for software, because you'll definitely need Lightroom and Photoshop. Check Chapter 7 in Stunning Digital Photography (link in the description) for an overview of shooting weddings, including the type of gear you need and practices to help you get ready.
A Sennheiser EW112PG3-A wireless lavalier mic directly into the camera. Very expensive ($630) but I love it. My earlier videos were made with a corded lav mic and it was such a nuisance, and the sound wasn't as good.
Check the other comments, but maybe. I used autoexposure for both cameras, same metering type, so there's no good explanation about why they exposed differently. Who knows. Either way, just one test, and the overall conclusion is that's there's not an appreciable difference in image quality.
You don't mention what you're shooting. Unless it's wildlife or sports, I'd push you to get a used 5D Mark II from ebay, and save the rest for lenses. I'm not sure why you want the primes or the 28-135, unless you shoot with two bodies simultaneously. I'd go with a 5D2, the 24-70, and 70-200. Anyway, let me know what you shoot and I can make a better recommendation.
Yeah, get the 5D2 and install Magic Lantern. I have a zoom but unless I need to record from multiple mics I just record directly into the camera from an external mic so I don't have to sync later. The Canon 24-105 f/4 or Tamron SP 24-70 mm f/2.8 Di VC USD are ideal for video.
Hi, Brenden. Check the other comments for a discussion of the first set of images. I used AI servo mode with the default settings for both cameras. I have several thousand frames of moving subjects with the 5D2 and 5D3 since, and I can confirm that the 5D3 is far better. All pictures were shot in RAW and processed using LR 4.1 default settings. So, picture styles don't apply. If you're interested in a more objective, technical comparison, download the lab test files in the description.
If you're tight, get the Mark II and save the cash on lenses (links in the description of this video). Better yet, find a used Mark II on ebay; they work great and you can get them for around $1100.
Hey, and thanks for writing. The indoor portrait is the only one I noticed where the 5D3 seemed darker. But, that was just one of many tests, and it didn't affect my final conclusion. Professional photographers use single point AF because it's always more accurate than manual focus, especially with moving subjects. The goal was to provide a comparison of real-world uses. If you want a lab comparison controlled conditions, check the description for the links to the lab test files.
I'd be happy to look at your comparisons, but as you can see I did quite a detailed study. I also did another review featuring more cameras, and once again the 5D2 and 5D3 had similar noise. DxO's objective tests also show similar noise (with the 5D3 being about 3% better overall). Anyway, show me some evidence and I'll take a look at it.
Yeah the eso got my attention, I was thinking surely tony doesn’t pronounce it like that anymore, I’d never noticed it and I’ve watched a heap of the newer content
I used to have professional imaging monitors, 2 NEC Multisync 2180UXs, and I switched to the consumer Dell 30" (2 of them). More screen space, but lower image quality. I can't say the drop in image quality has affected my photography. Even when I'm using a laptop (like my XP13), it gets the job done fine. So I guess my recommendation would be to get two monitors, because that's terribly useful with Lightroom, and I personally prefer anti-glare rather than glass, and the more pixels, the better.
Yeah, that was the first video I shot with the 5D3, and it looked pretty awful. But there was no moon that night and I was lit only by a streetlight about 20 feet away. It wasn't pitch black, but it was dark enough that I couldn't see the buttons on the camera (and I wasn't yet familiar enough with the controls).
By chance are you comparing the JPGs instead of the RAW files? While I was (to the best of my knowledge) the first person to notice the lack of RAW image quality improvement, lots of people have since repeated the raw tests and had the exact same results.
The D800 is an amazing camera. The image quality is better than the 5D Mark II or 5D Mark III, and it's less expensive than the 5D Mark III. However, it falls behind the 5D Mark III in some areas, especially auto-focus. I have Canon lenses, so the D800 isn't an option for me. But if it were, I'd probably get the D800. Instead, I'll just wait for Canon to release something competitive.
The 5D Mark II or 5D Mark II, whichever I'm not currently holding in my hand. You can't really tell the difference for this type of video, with the exception of the night video, which is much cleaner on the 5D3 than the 5D2 (though still basically unusable in this example).
I do all my night photography at ISO 100 to keep the noise down (check Chapter 10 of my book). However, you can check this video's description for a link to the "high ISO, low-light lab test" files where I shot both camera bodies in a dark room at a range of ISOs. Feel free to make your own judgement, but I did a blind test on DPReview and nobody thought the 5D3 was better until it got to ISO 50,000+. Anyway, the 5D3 is amazing in other ways, but not image quality.
Glad to hear it. Adding up just the comments like this that I've heard, I've saved people at least $30,000 with this review, and I imagine most people don't bother adding a comment. FYI, I've been switching between the 5D2 and 5D3 since I made this video (Chelsea and I often swap cameras) and the 5D2 still holds up. I've even done a lot of action work with it and the AF system is quite workable, even though it's the weakest part of the comparison. Anyway, enjoy, and check ebay for used bodies!
My god man, it's NOT an acronym. Forgive me for growing weary of addressing this OVER and OVER again. I don't really care how people say it, but everyone on UA-cam seems to care. "The organization says that ISO is not an acronym or initialism for the organization's full name in either official language; rather, recognizing that its initials would be different in different languages, it adopted ISO, based on the Greek word isos (ἴσος, meaning equal), as the universal short form of its name."
That's a good combo. Use the 7D for anything that requires fast AF, and for it's crop factor with wildlife. Use MK2 for everything else. BTW, I've been travelling with both the 5D2 and the 5D3 since this review, and I've taken thousands of pictures with each of them side-by-side now, and I stand behind everything in the review. In other words, experience hasn't yielded any big surprises.
Yeah, I love the D800's sensor (though it doesn't stand up to the 5D3 in other ways). I'd be jazzed if Canon released a 40MP sensor, but I'll believe it when I see it.
Good review I own both 5D mark II and III and its true it really depends on what you shoot I cant stand reviews of people that buy a 5D run both cameras down because when really they wanted a 1DX.
Thanks, John. However, I've become somewhat of an expert on this useless bit of trivia about ISOs name, and it's definitely not an acronym. Search the web for "ISO is not an acronym", check the ISO about page, and look up some official videos produced by the ISO organization to hear their pronunciation.
I want to see the raw files you're using as evidence for your claim that the 5D3 has substantially better IQ. DXO supports my findings. At ISO 100, the 5D2 has an SNR of 43.1. The 5D3 has an SNR of 44.4. That's an improvement of 3%. ISO 800, it's 32.6 to 33.3, for a 2% improvement. Even at ISO 12,800 (way beyond what any professional could use). it's a 15% improvement. Note that I acknowledge that improvement at unusably high ISOs in the video. The 5D2 has (slightly) better dynamic range.
great video! thanks. i'm a photography student, just starting this year and I need to buy a new camera, i'm leaning towards the mark II because i'm not sure what kinds of photography i'm gonna be doing, so it makes sense to save money now. but i'm still a little torn so for someone in school who's going to be experimenting with lots of different kinds of photography would you recommend the mark III or the mark II?
📷Get the Canon 5D Mark II here: help.tc/5d2
📷And the Mark III here: help.tc/5d4
Tony & Chelsea Northrup would you buy a used 5d Mk2 or a used Nikon D7100 ? I shoot portraits family stuff and would like to get into weddings.
Would you still recommend the Mark II in 2018 for Pro Photographers
Do you know what camera this video was shot with? I mean when you were in the studio, this is 2012 and that video is brilliant!
"EE-so" is the pronunciation in most non-English speaking countries, and the teacher I learned the term from was not a native English speaker. This is an older video, and because I get SOOOO many comments on the pronunciation, I make a point of saying "eye-so" now.
Check ISO's official About page for an explanation of why ISO isn't an acronym (it's derived from the Greek word isos). Or, just google, "ISO is not an acronym".
See, you can still learn a thing or two, even after 38 years!
And did Paul Gauguin happen to abandon painting because of the critics? Don't waste yourself on small things. I used to say "ISO" with an old Portuguese accent.
I just picked up a used 5D II last weekend. So far so good!
Wow, most of the comments are about the correct way of pronouncing the ISO?
Thank you for the video! I'm about to upgrade my 600D to a budget full frame body. The 5D mk ii still kicks ass.
theizza68 Yes I'm buying it for 250$
Glad to hear it. My Mark III is in the shop so I've returned to using my Mark II. Image quality is the same, and it's getting the job done, even for focusing on moving subjects. If you don't *really* need the improved autofocus, get the Mark II and save the extra $1500 for lenses.
Depends what you're shooting and how you're using it. If you do get the 5D2, you'd also need to upgrade to full-frame lenses, which are bigger, heavier, and more expensive.
I just checked, and I can see. If you want a second opinion on the image quality, check DxO. Their objective tests determine that the 5D Mark III's image quality is 3% better than the 5D Mark IIs.
So hard to go back to his older video's and hearing him say ESO...
Yea! what's with that? ESO
Rude@@RobBob555
RockstahRolln he means I S O... Who knows, maybe he had a teacher who spoke spanish or something.
@@eddmaster9 You both complain a lot! Try saying DIN, das ist Deutsch. Ou tente dizer ISO em português.
@@hannibalcosta lol, im not complaining bro. Just telling him why the different pronunciation.
The name of the group is International Organization for Standardization. Look it up on Wikipedia: "The organization says that ISO is not an acronym or initialism for the organization's full name in either official language; rather, recognizing that its initials would be different in different languages, it adopted ISO, based on the Greek word isos (ἴσος, meaning equal), as the universal short form of its name."
Well, it does sort of make sense, because if they used an acronym it would need to be different in different languages. So, they chose a three-letter word.
The part that doesn't make sense is why they capitalize it, making everyone think it's an acronym when it's just a word. However, the organization is quite clear that ISO is not an acronym and thus should never be spelled out.
Yes, and that's part of the risk of doing a real-world test rather than a lab test. Check the links in the description for RAW files done in a lab environment. Draw your own conclusions, but mine are that the 5D3 is equal to the 5D2 until you get above ISO 50,000.
Thanks!
The 24-105 f/4 is way more flexible than the 24-70, mostly because it has IS. Still, I often want the f/2.8 speed, and I've been waiting for for years for a 24-70 f/2.8 with IS. Tamron has one, but it doesn't have great image quality.
If you're a pro and you don't need the IS, get the 24-70. The Mark II isn't out yet, but it's priced at $2300, which is more than twice the 24-105.
For everyone who isn't a pro, or pros who do low-light hand-holding, get the 24-105 and a 70-200 f2.8.
BTW, recently learned that ISO is not an acronym. It should be pronounced "eye-so". It's a form of the Greek word "Isos" which means equality, and is the same as the prefix for many words such as "isometric."
Yes, but it's a challenge because the live view is too dark. I describe night focusing techniques in Chapter 10 of Stunning Digital Photography (links in the description) but on the Mk II it amounts to estimation and then trial-and-error.
Often flying birds are easier than moving subjects on the ground because they have a clean background (the sky). With ground subjects cameras tend to latch onto the background. Sandi's light colored fur might not have been as contrasty as a bird against a blue sky. Also, as subjects get closer to you the camera has to do more work to focus, so it's easier to focus on farther away moving subjects.
I've had dozens of people make this comment to me, and I make them all the same offer: show me the raw files where the 5D3 is better than the 5D2. Not only will I do a follow-up, but I will personally be thrilled to discover that the 5D3 has substantially better IQ, because I plunked down my own cash for the camera and IQ is extremely important to me. I definitely only have bias toward the 5D3, but I need some evidence.
They'd be about the same, and I like the 6D a touch better, so I'd go with that. Note that used 5D Mark IIs are selling for $1200, so consider that, as well...
Bongo is right that the focus system is the weak link on the 5D2. It's absolutely fine for everything except action shots, and if you use the center autofocus point, you can do an OK job with action.
Thanks! I'm definitely going to check out the D800--the higher MP would be really useful. I do love the 5D3. My only problem is that Canon is marketing it as having a sensor that dramatically improves low-light performance, and my testing shows no improvement whatsoever.
I have a tour of my studio that describes the equipment, and some outdoor lighting videos with off-camera flash, but otherwise, no. We're planning to do the studio videos this winter while we're shut in from the cold.
You're definitely not the only one :D.
We alternated filming with the 5D Mark II and 5D Mark III, because those are the only two bodies we have that can film, so while I shot with one, we filmed with the other. Most notably, the night scene was filmed with the 5D Mark III.
I can tell a big difference between the cameras at night, but I don't really notice much difference when the lighting is good. However, other people more picky about video quality than I say the 5D3 is much better at video.
Yes, the Canon 70-200 f2.8 Mark II.
Thanks, and best wishes from the US!
Sorry, I'm in the US, and I usually recommend bhphotovideo and Amazon. I'm not sure if they're the best choices in the Uk, however.
I was quite surprised too, and not in a pleasant way!
No, I'm working with raw files, which aren't processed by any in-camera software. I'm afraid we're stuck with it; firmware isn't likely to improve it. It still does great work, though, and I love the 5D3 more every day I use it.
You're looking for tethering software. I use Adobe Lightroom (which is necessary software, anyway). Just connect your camera to your Mac with a USB cable.
Don't let the fan boys get you down. Keep making these shows. You did a fantastic job.
The Nikon D4 and the Canon 1D are specialized professional bodies. They're big, heavy tanks. They're built without any compromises (outside of weight, cost, and often resolution).
So, I use the smaller full-frame body because I travel a great deal (it's much lighter than the 5D3/D4) and the 5D3 is durable enough for the abuse I give it. I use a Canon instead of Nikon for no good reason other than I've always used Canon equipment and it's expensive to switch systems.
I zoom to 4:1 to exaggerate the noise. It's a technique I learned from the stock photo agencies. With commercial work, this allows me to identify areas that need to be de-noised. Even at ISO 100, I always need to remove noise in solid areas of primary colors (R/G/B) and in shadows.
Just because you can't see noise at 1:1 on your monitor doesn't mean it's not significant; it shows up in large prints and on sharp monitors, and stock agencies will reject it.
Working on the ISO pronunciation.
Check the other comments for info about the first test.
There's no review or test that shows the 5D Mark III has noticeably better raw image quality than the 5D Mark II. DXO tests show an improvement of 2-3%. Image quality is essentially unchanged.
This is one area where the 5D2 and 5D3 differ. The 5D3 works great with all autofocus points on (assuming there's nothing in the background it can focus on--you want to be shooting up into the sky or just use the center focus point). With the 5D2, only the center focus point is cross-type. Basically, only the center focusing point on the 5D2 is fast enough to even hope to keep up with a moving subject.
Yes, everyone says it but it only applies to JPG files processed in-camera. People who care shoot RAW.
I've shot over 15,000 frames with the 5D3, and more than 100,000 on my 5D2. I care about the noise because I have to fix it (even at ISO 100) or my commercial clients reject the photos.
Re: the first set of pictures being darker, read the other comments. Check the other examples and the lab files linked in the description for a controlled test.
Still waiting for your RAW files.
Raw image quality is the same, but the 5D3 is functionally better at everything, it's just more expensive. If you're doing still portraits and landscapes, go with the 5D2. If you're doing weddings, sports, wildlife, or night photography, go with the 5D3
I didn't have a flash. Note that while the 5D3 is definitely better in the dark, and the focus works in darker environments, there are definitely still going to be times when it's too dark for it to focus.
Sorry about that! Sounds like something came loose. I'm sure it can be repaired. Just send it to Canon for repairs--I've had them do several repairs on different bodies and lenses and it always works out just fine.
Good call. Especially since 5D2s are available now used for about $1100, I very rarely recommend the 5D3 to people. It really only makes sense for wedding or sports photographers... Everyone else should spend the extra cash on lenses and lighting. Even wildlife photographers, who need good AF, are usually better off with a 7D and the higher pixel density.
When he pulled out his 500mm at 3:44 with a straight face I laughed my ass off. What a monster HAHAHAHA
THAT MONSTER COST 10K TOO LOL
So expensive....this guy has made some big time investments. What I would do for a lens like that!
Yeah, the NEX 7 is my mirrorless recommendation if image quality is your priority. You might also look at the Olympus OM-D E-M5.
The light leak issue isn't really a problem. It would only appear if you had the LCD light on while taking a long exposure (say 20 seconds) at night.
Pictures from the 3 look better if you shoot JPG. However, if you care about image quality at all, you should be shooting RAW, and the RAW files are basically indistinguishable. DxO's objective tests put the 5D3 at 3% better than the 5D2, which nobody would ever notice.
Anyway, there are many good reasons to get a 5D3, but not image quality.
It brackets, but it doesn't have HDR built in. If you install Magic Lantern, it has some fairly remarkable bracketing capabilities.
I don't know what type of video you plan to do or what your total budget is, but for almost everyone I would recommend the 5D2. In fact, I'd push you to buy a used one. Put the money you save towards some nice mics (I love my Sennheiser wireless lav mic), lenses, lighting, tripods (watch my tripod video), and software.
I travel with both the 5D2 and 5D3, and the only time I make a point of using the 5D3 for video is at night, but I think Magic Lantern might negate even that benefit now.
The 5D3 has a great AF system; every reviewer has agreed on this one point. No system is perfect with moving subjects, however. I can't tell you whether it's better than the 1D3, but given that it's five years newer, I imagine it is.
Yeah, I'd love medium format for the studio, but they don't support higher ISO, telephoto lenses, or even good autofocus, besides being way bigger to travel with. Anyway, you're right, I dream of the Hassy.
Yeah, it's fantastic. It's one of the sharpest zooms ever made, and at f/2.8, every headshot has a beautiful blur. Check my other portrait videos for some examples. It's also $2,300, though, and if you're not doing commercial work where people literally investigate every pixel, you can get by with the Sigma version of the lens ($1,300 at Amazon) or even grab a used Mark I version of the Canon on ebay.
Both the 5D Mark II and the Mark III (whichever one isn't currently in my hand)
It wasn't intentional. To the best of my knowledge, the lighting didn't change. The two bodies just seemed to meter the situation differently.
Incidentally, now that myself and other people have done controlled lab tests, the overall results hold up: the 5D3's raw image quality is indistinguishable from the 5D2's (within 2-3%) at all reasonable ISOs (say, ISO 3200 and below).
I shot in raw and processed the images in Adobe Lightroom 4.1 RC. You're right, the claimed image quality improvement applies only to the in-camera JPEG processing, so it doesn't impact those of us who shoot raw (which is everyone who really needs good image quality). The AF system will really help your low-light work, at least compared to the 5D2. I did try the 5D3 video at night, but it didn't turn out well, as you can see--but it is probably still be better than the 5D2.
At the same price, the 6D. However, I push people towards used 5D Mark II's lately, because they're available so inexpensively. There are a ton of 5D Mark iI's on ebay for $1300, whereas the 6D is going for $2100. For that price, I'd grab the Mark II and have $800 for an awesome lens.
For me, upgrading to a UDMA 7 card allowed me to take 33 shots without it slowing down. With an SD card, I was only able to get 13. Plus, after the buffer filled up, it slowed down much less than with the SD card.
My SD card is a class 10.
Decision made!! I currently own a 5DMKii and a T7i. I was literally about to sell both and get a 5DMKiii but decided to do a bit a research. Your comparisons made it easy for me to keep what I have and avoid the hassle of selling. Both cams serve me well based on what I shoot. PLUS I can keep the clicks down by splitting the workload :-D! Thanks for the very informative review! Off topic, the T7i is an excellent choice for low light shooting as well..
Which camera do you like better? Im currently thinking about buying 77D(its pretty much the same as 800D, sensor, af system etc) or 5Dmk2
Not sure what other gear you have, but unless you have an unlimited budget, for weddings, I'd get the 5D Mark II used and a 70-200 f/2.8.
Yes, the 5D3's AF is helpful during weddings in low-light; it'll snap a hair faster. But the 5D2 always got the job done for me, and it still does. You can buy 3 used 5D2s for the price of one 5D3.
The higher FPS is useful for wildlife and sports, but for those, I would stay with the 7D because the crop helps. Why not keep the 7D as a backup to the 5D2?
Sure, it depends on how much contrast there is in the scene, how fast the subject is moving, and how much light you have. I wasn't trying to define how many shots the 5D2 misses, just trying to determine how much of an improvement the 5D3 is.
Actually, that's an awesome idea. I do have a 7D for wildlife and full-frame bodies for everything else... but you make a compelling argument against the 5D3. Your solution also gives you a backup body if one fails.
Read through the comments; you'll see that addressed. Feel free to dismiss that test if you like. The other tests are still valid.
The video is measurably better, according to people who analyze these sorts of things, but I haven't found that it's noticeably better for the types of video that I do (at least, below ISO 1600). Particularly if you install Magic Lantern on your 5D2, the 5D2 is still great for video.
Agreed, and a few people have mentioned this. I shot them in the same minute, but maybe the sun went behind a cloud. It only impacts the one test, though, and the overall conclusion was that at all but the highest ISOs (50k and above) the 5D2 and 5D3 have similar image quality. Check the description of the video for links to the "high ISO, low-light lab tests" that were done under controlled lighting conditions and confirm my conclusion (but feel free to judge for yourself).
Definitely get the Mark II. My Mark III has been in the shop for the last month, so I'm using the Mark II again, and it's getting the job done just fine. Spend the extra ~1400 on a nice lens and flash.
Maybe even check ebay for a good deal on a used Mark II from someone who sold it to upgrade :)
Yeah, you'll notice I took the 5D2's win in that area with a grain of salt. I shot the images back-to-back, but you know how natural light is: perhaps the sun went behind a cloud or something. I'll repeat the low-light portrait test today under artificial lights and post a link.
Do note that I tested image quality with wildlife, landscapes, and night photography without any lighting variations and the image quality seemed identical.
I agree at ISO 51,000+ the 5D3 is (slightly) noticeably better, and DXO says the improvement is about 15%, but at lower ISOs (every ISO anyone would want to use) there's no noticeable improvement, and DXO supports that.
Is that what you're finding? Would you send me the raw files showing the improved performance from the 5D3? If you read through the comments here, I've had a lot of people saying the 5D3 is far better, but nobody has actually submitted proof, so I'm excited to see some.
Great review! Honest and in depth!
Yes, for landscapes and stills, go with the 5D Mark II.
That's on our to-do list. Sorry we haven't completed it yet!
Just wanted to follow-up. Do you have a raw comparison? I've had dozens of people tell me the 5D3 is way better and that I'm wrong, but literally zero people have shown me an actual example.
The 5D3 has much better low-light video performance, and it has a headphone jack for monitoring the audio. If that's important to you, get the 5D3, otherwise get the 5D2.
This has been addressed over and over again in the comments. This video was the real-world subjective test. I also did a controlled test; check the description for the link to the lab files. The controlled test supports the findings in the real-world test, but you can draw your own conclusions.
Everyone is open to your point of view, but if you want anyone to take you seriously, you're going to need to produce some raw files supporting your claims.
Yeah, lenses hold their value better than bodies. I don't do primes except for telephoto and macro lenses. Zoom lenses rule, and the Canon zooms (especially the Ls) are as sharp as primes.
Depends on what lenses you already have, what you're shooting, and who your customer is...
OK, that makes sense. I think my previous recommendations work. If your funds are unlimited, the 5D Mark III's AF would be useful, but the 5D2 always worked for me. Also, you'll need a good bounce flash if you don't have one. Save some cash for software, because you'll definitely need Lightroom and Photoshop. Check Chapter 7 in Stunning Digital Photography (link in the description) for an overview of shooting weddings, including the type of gear you need and practices to help you get ready.
Agreed; if they're going to compete against the D800 or even the Mark II, they need to get the body price below $3000.
No, I haven't, sorry. I'm using PocketWizards, a 580EX (original!) and a couple other odd flashes.
A Sennheiser EW112PG3-A wireless lavalier mic directly into the camera. Very expensive ($630) but I love it. My earlier videos were made with a corded lav mic and it was such a nuisance, and the sound wasn't as good.
Check the other comments, but maybe. I used autoexposure for both cameras, same metering type, so there's no good explanation about why they exposed differently. Who knows. Either way, just one test, and the overall conclusion is that's there's not an appreciable difference in image quality.
You don't mention what you're shooting. Unless it's wildlife or sports, I'd push you to get a used 5D Mark II from ebay, and save the rest for lenses.
I'm not sure why you want the primes or the 28-135, unless you shoot with two bodies simultaneously. I'd go with a 5D2, the 24-70, and 70-200.
Anyway, let me know what you shoot and I can make a better recommendation.
Yeah, get the 5D2 and install Magic Lantern. I have a zoom but unless I need to record from multiple mics I just record directly into the camera from an external mic so I don't have to sync later.
The Canon 24-105 f/4 or Tamron SP 24-70 mm f/2.8 Di VC USD are ideal for video.
Hi, Brenden. Check the other comments for a discussion of the first set of images.
I used AI servo mode with the default settings for both cameras. I have several thousand frames of moving subjects with the 5D2 and 5D3 since, and I can confirm that the 5D3 is far better.
All pictures were shot in RAW and processed using LR 4.1 default settings. So, picture styles don't apply.
If you're interested in a more objective, technical comparison, download the lab test files in the description.
If you're tight, get the Mark II and save the cash on lenses (links in the description of this video). Better yet, find a used Mark II on ebay; they work great and you can get them for around $1100.
Hey, and thanks for writing.
The indoor portrait is the only one I noticed where the 5D3 seemed darker. But, that was just one of many tests, and it didn't affect my final conclusion.
Professional photographers use single point AF because it's always more accurate than manual focus, especially with moving subjects.
The goal was to provide a comparison of real-world uses. If you want a lab comparison controlled conditions, check the description for the links to the lab test files.
Search the web for "ISO is not an acronym" (or just check the official ISO website and videos)
I'd be happy to look at your comparisons, but as you can see I did quite a detailed study. I also did another review featuring more cameras, and once again the 5D2 and 5D3 had similar noise. DxO's objective tests also show similar noise (with the 5D3 being about 3% better overall).
Anyway, show me some evidence and I'll take a look at it.
eso eso eso eso is all i heard
Please watch sdp.io/iso
He pronounces it better nowadays .
Yeah the eso got my attention, I was thinking surely tony doesn’t pronounce it like that anymore, I’d never noticed it and I’ve watched a heap of the newer content
@@TonyAndChelsea no thanks, this video hurt my brain enough with all the eeso
You obviously know what you are doing and are good at it. I would be immensely grateful if you could do a video comparison as well.
I used to have professional imaging monitors, 2 NEC Multisync 2180UXs, and I switched to the consumer Dell 30" (2 of them). More screen space, but lower image quality. I can't say the drop in image quality has affected my photography. Even when I'm using a laptop (like my XP13), it gets the job done fine.
So I guess my recommendation would be to get two monitors, because that's terribly useful with Lightroom, and I personally prefer anti-glare rather than glass, and the more pixels, the better.
10 yr ago Wow time flies.
Yeah, that was the first video I shot with the 5D3, and it looked pretty awful. But there was no moon that night and I was lit only by a streetlight about 20 feet away. It wasn't pitch black, but it was dark enough that I couldn't see the buttons on the camera (and I wasn't yet familiar enough with the controls).
Yeah, the brand matters little to me. I bought a Canon Elan II-E in 1996 or so and just haven't bothered to sell all my gear :).
Hands down the best comparison I've seen on the net. Majority of users/testers beats around the bush but this vid was extensively convincing.
By chance are you comparing the JPGs instead of the RAW files? While I was (to the best of my knowledge) the first person to notice the lack of RAW image quality improvement, lots of people have since repeated the raw tests and had the exact same results.
The D800 is an amazing camera. The image quality is better than the 5D Mark II or 5D Mark III, and it's less expensive than the 5D Mark III. However, it falls behind the 5D Mark III in some areas, especially auto-focus.
I have Canon lenses, so the D800 isn't an option for me. But if it were, I'd probably get the D800. Instead, I'll just wait for Canon to release something competitive.
The 5D Mark II or 5D Mark II, whichever I'm not currently holding in my hand. You can't really tell the difference for this type of video, with the exception of the night video, which is much cleaner on the 5D3 than the 5D2 (though still basically unusable in this example).
I do all my night photography at ISO 100 to keep the noise down (check Chapter 10 of my book). However, you can check this video's description for a link to the "high ISO, low-light lab test" files where I shot both camera bodies in a dark room at a range of ISOs. Feel free to make your own judgement, but I did a blind test on DPReview and nobody thought the 5D3 was better until it got to ISO 50,000+. Anyway, the 5D3 is amazing in other ways, but not image quality.
Go DIZEMAN! You rule, buddy.
I'll let you rest but leave the link in the description for other people following this. Best UA-cam thread ever!
Glad to hear it. Adding up just the comments like this that I've heard, I've saved people at least $30,000 with this review, and I imagine most people don't bother adding a comment.
FYI, I've been switching between the 5D2 and 5D3 since I made this video (Chelsea and I often swap cameras) and the 5D2 still holds up. I've even done a lot of action work with it and the AF system is quite workable, even though it's the weakest part of the comparison.
Anyway, enjoy, and check ebay for used bodies!
@10:30 for the summary and save yourself time!
My god man, it's NOT an acronym. Forgive me for growing weary of addressing this OVER and OVER again. I don't really care how people say it, but everyone on UA-cam seems to care.
"The organization says that ISO is not an acronym or initialism for the organization's full name in either official language; rather, recognizing that its initials would be different in different languages, it adopted ISO, based on the Greek word isos (ἴσος, meaning equal), as the universal short form of its name."
Eh, the Mark III only has 5% more pixels. I think the normalization process would create artifacts worth more than 5%.
That's a good combo. Use the 7D for anything that requires fast AF, and for it's crop factor with wildlife. Use MK2 for everything else.
BTW, I've been travelling with both the 5D2 and the 5D3 since this review, and I've taken thousands of pictures with each of them side-by-side now, and I stand behind everything in the review. In other words, experience hasn't yielded any big surprises.
Yeah, I love the D800's sensor (though it doesn't stand up to the 5D3 in other ways). I'd be jazzed if Canon released a 40MP sensor, but I'll believe it when I see it.
Good review I own both 5D mark II and III and its true it really depends on what you shoot I cant stand reviews of people that buy a 5D run both cameras down because when really they wanted a 1DX.
Thanks, John. However, I've become somewhat of an expert on this useless bit of trivia about ISOs name, and it's definitely not an acronym. Search the web for "ISO is not an acronym", check the ISO about page, and look up some official videos produced by the ISO organization to hear their pronunciation.
I want to see the raw files you're using as evidence for your claim that the 5D3 has substantially better IQ.
DXO supports my findings. At ISO 100, the 5D2 has an SNR of 43.1. The 5D3 has an SNR of 44.4. That's an improvement of 3%. ISO 800, it's 32.6 to 33.3, for a 2% improvement. Even at ISO 12,800 (way beyond what any professional could use). it's a 15% improvement. Note that I acknowledge that improvement at unusably high ISOs in the video.
The 5D2 has (slightly) better dynamic range.
great video! thanks. i'm a photography student, just starting this year and I need to buy a new camera, i'm leaning towards the mark II because i'm not sure what kinds of photography i'm gonna be doing, so it makes sense to save money now. but i'm still a little torn so for someone in school who's going to be experimenting with lots of different kinds of photography would you recommend the mark III or the mark II?