7D vs. 5D Mark II...I know it's been done to death...but...

[danielleetaylor]

<p>Mark - <em>Keep the 40D, and RENT a 5DMKII (and whatever lens) for your vacation trip(s) and give it a test drive. Costco makes brilliant 16x20 and 20x30 poster prints from my 40D!</em></p>

<p>I'm going to use your comment to illustrate something. I have used Costco for printing > 8x10 in the past, but I now own an Epson 3880 for prints up to 16x24.</p>

<p>On my most demanding images I consider the difference between the Epson 3880 and the Noritsu printers Costco uses to be much greater than the difference between a 5D mkII file and a 7D file. It's not that Costco is bad. But I could easily and consistently pick between prints from those two printers, especially when using a paper like Epson Hot Press at 2880 dpi. And it wouldn't just be the paper that would give it away. The Epson is clearly better in terms of sharpness, fine detail, tonality, and both highlight and shadow detail.</p>

<p>By contrast I cannot reliably discern between 5D mkII and 7D files printed to 16x24 when they are optimally processed and printed. That's how small the difference really is. Basically just about any other difference you can think of in lenses, technique, post processing, and printing will be larger than the differences between those two sensors at low to mid ISO.</p>

[danielleetaylor]

<p>Philip - while I agree that the 5D mkII is better at high ISO, I don't feel your test crops are convincing either way. The landscape crops in particular don't impress me one way or the other even though the 5D mkII has the better settings (aperture).</p>

<p>As for DxO, let's leave them out of this. They rank the 5D mkII and Sony A900 as better than medium format digital backs! I don't own a MFDB, but I have studied prints from a few and you would need to stitch at least 3 frames from a 5D2 or 7D to begin to approach their IQ, at least on large prints. The many flaws in DxO's testing methodology and results have been beaten to death. They're not a reliable source of information.</p>

<p><em>Similarly while Photozone does not have tests with the 7D they do have tests of the 85 F1.8 with the 5DII and 50D.</em></p>

<p>Photozone explicitly states that you cannot compare results between camera bodies due to the nature of their tests. DPReview performs a resolution test that is designed for sensor comparison, and you'll note that the differences are rather small between the 5D2 and 7D. Indeed, the 7D in RAW has a higher extinction resolution than the 5D2 in JPEG.</p>

[danielleetaylor]

<p><em>Okay...so back to me people ;)</em><br>

<em> My dear brother-in-law just handed over his 7D and 5D Mark II.</em></p>

<p>Zvia - just a quick note: pixel peeping out of camera the 5D mkII will often be sharper and have less noise. I do concede this point. But give each file optimal (i.e. not identical, but optimal for the sensor) processing in terms of local contrast enhancement, sharpness, and NR if necessary, and see how they compare on screen and in print after processing. Then of course you can weigh whatever differences you find against prices and features.</p>

<p>Good luck!</p>

[danielleetaylor]

<p>Zvia - I finally got through all the pages and saw your observations.</p>

<p>My only comment would be that at ISO 1600 and 3200 7D noise appears worse on screen than in print. I'm one of the people who always preaches about pixel peeping vs. print. Yet even I have had a "ho hum" reaction to some ISO 3200 files on screen, followed by a "wow" reaction after printing.</p>

<p>The first time this happened was when reviewing some casual, hand held portrait shots of my 8 year old cousin at her older sister's HS basketball game (7D + 85 f/1.8, HS gym lighting, no flash). They were really cute and when I first opened them I thought "ah, I hope the noise doesn't irritate me." I did some post work, didn't over do the NR since I hate the plastic look even more, then printed. On screen the noise worried me. In print it was tighter and less noticeable than grain from scanned, low ISO 35mm portrait film. (And it even had the look of tight grain with zero banding.) I was thrilled with the prints and so was her family. Same for the basketball shots.</p>

<p>Would the 5D mkII have had even less noise in that situation? Yes. But it would not have been a huge difference in print. I might have been able to tell, but no one in my family would have noticed. And of course the 7D had the better AF and fps for the basketball shots which I had been invited to take.</p>

<p>I'm sure you can predict this, but if I were in your shoes I would just get a 7D, including for your trips. And I would save up for a 5D mkIII. I know you said you might rent a 5D mkII for your trips, but really think about whether or not you will gain anything for the cash vs. just putting it towards the future.</p>

<p>I am a big fan of the 7D, but I am looking forward to seeing what Canon does with FF in the next generation. If they manage to take the next 5D into the 30 MP territory and improve noise yet again I will probably end up with both myself.</p>

[danielleetaylor]

<p>Dave - no surprises with your IR samples. I know I sound like a broken record, but there just aren't any huge differences between these sensors at low to mid ISO. I don't know where the post is now, but I recall being challenged to do a Google search and all the top results/reviews came to the same conclusion. Differences, yes. Huge differences, or even print visible differences? No. I don't know why it's so hard to convince others of this, or what the psychology is behind the "must have FF" meme.</p>

<p>Zvia - it just occurred to me that rather than buy or rent a 5D mkII for your trips it would be less expensive to simply get a pano head such as the Panosaurus ($90) and stitch the best landscape scenes. Multi row stitching is a pain. But you can very quickly shoot and stitch a single row of three frames. 3 stitched 7D frames will produce a file equal to or better than 6x7 MF film on a high end scanner, and will yield beautiful, detailed prints at very large sizes. You can't stitch every scene. But generally speaking if it's a landscape and you have time to setup a tripod, you have time for a 3 frame stitch.</p>

[keith reeder]

<blockquote>

<p>Following is a pre-dawn shot with the 5D MkII and the 70-200mm f/4L IS with the 1.4x TC at ISO 6400:</p>

</blockquote>

<p>David, it <em>looks </em>NRd - obviously, and heavily - especially at <a href=" size</a> on your Flickr stream. The fir tree on the left looks painted.</p>

<p>As soon as NR is that blatant, any argument that "<em>camera a is better at high ISO than camera b</em>" goes right out of the window because all we're looking at is the result of (DxO) NR, not the inherent ability of the camera, and DxO is a "blunt instrument" in NR terms.</p>

<p>Again, blurring away every hint of noise and detail proves nothing about a camera's noise capabilities. I can easily do that with my 7D, but I like detail, and for my pictures not to look like oil paintings.</p>

<p>Here's 6400 ISO from my 7D (in less light than your shot too, judging by the exif), resized to the size of the elk image: http://www.capture-the-moment.co.uk/tp/tfu29//upload/Lr-2382_2048.jpg</p>

<p>I know which one looks better to me...</p>

<p> </p>

[mike_hitchen]

<p>Zvia - an excellently presented summary, not only content but how you explained it. Have you ever thought of a sidelines as a gear reviewer? :)<br>

You seem to talk about the 5D in terms of 'artisitc' quality, and the 7D functional terms. And I agree completely with Scott's comments on the so-called tele advantage. Have you taken successively tighter crops from the 3 cameras to see at what point the image breaks up? That could be a significant issue for wildlife shots.<br>

I have seen some great action and wildlife shots taken with the 5D, so the AF and frame rate are obviously not barriers to getting memorable shots. Just that the 7D makes it a bit easier.</p>

 

[danielleetaylor]

<p>Keith<em> - David, it looks NRd - obviously, and heavily - especially at <a rel="nofollow" href=" target="_blank">large size</a> on your Flickr stream. The fir tree on the left looks painted.</em></p>

<p>I don't like trying to make judgments from photos which are so radically different as David's and yours. But I will say that in David's shot the NR seemed to do a number on the out of focus areas. More than that though I'm surprised by the color noise in the shot. That is not better than what I would expect out of my 7D. I dislike color noise more than luminance noise and what has impressed me with the 7D is that in print through 3200 there really isn't any color blotching so long as you nail exposure.</p>

<p>From online test samples and the few tests I did at high ISO I reached the conclusion that the 5D mkII does have a high ISO advantage. But reading your posts makes me want to pit the two against each other again with the latest versions of various RAW converters, just to see what can be squeezed out of each and if the gap narrowed thanks to software. I haven't done it because I haven't felt any pressing need. My 7D has more than impressed me through 3200, which is about as high as I need to go.</p>

[danielleetaylor]

<p>David - <em>Also remember that you can crop the 5D2 image so that the subject is the same size as on the 7D and there'll be approximately the same number of pixels on the subject. Hence, there'll be no loss in image quality, even after cropping.</em></p>

<p>Am I misunderstanding what you're saying here? If you crop the APS-C portion of a 5D mkII file you will end up with an 8 MP image. The 7D is 18 MP. That's a big difference. For subjects like sports and wildlife you may not notice the difference in small to medium sized prints. But if you print large or have to crop further the difference will become apparent.</p>

<p>I've got 16x20 surfing prints made from cropped 7D images where the final resolution was about 10 MP. That's sufficient for the subject matter. But if I had used a 5D mkII I would have been down to 4 or 5 MP, and that would not be sufficient for a 16x20 IMHO.</p>

[keith reeder]

<blockquote>

<p>I don't like trying to make judgments from photos which are so radically different as David's and yours.</p>

 

</blockquote>

<p>Oh, quite right too, Daniel.</p>

<p>But the fact remains that there's <em>a lot</em> more detail in the fabric and "fur" of the toy than in the elk or its surroundings (and there'd be more if I sharpened it properly) and no "noise problem" whatsoever, demonstrating that it's certainly possible to have low noise <em>and </em>detail, and that - in terms of the balance between noise and detail - the 7D lacks for nothing in practical terms compared to the 5D Mk II: you just have to know how to make it happen.</p>

[keith reeder]

<blockquote>

<p>But reading your posts makes me want to pit the two against each other again with the latest versions of various RAW converters, just to see what can be squeezed out of each and if the gap narrowed thanks to software. I haven't done it because I haven't felt any pressing need. My 7D has more than impressed me through 3200, which is about as high as I need to go.</p>

</blockquote>

<p>This is the nub of what I've been saying all along, Daniel - the 7D can be extraordinarily good at high ISO, but there's a right way to do it.</p>

<p>I have a very straightforward approach, but it simply <em>works: </em>conversion in Lr 3.3 with Chroma NR always at default of 25 (which works perfectly right up to 12800 ISO); and Luma at around 10, give or take.</p>

<p>Some slight capture sharpening (making use of the Masking slider as necessary) and I'm done.</p>

<p>Then if I need to, I'll apply some selective NR using Topaz Denoise in CS4 - apply on a layer then erase from where it's not needed. The interesting point here is that my 7D files need very little "help" at this stage - there's no additional NR whatsoever on the lion toy for example, and yet Chroma is entirely absent, and - in this case anyway - any Luma (and - be honest - there's not much) isn't observable because of the texture of the fabric.</p>

<p>But I've had the same quality of result shooting a wine bottle, where Luma no where to hide.</p>

<p>And - for for the absence of doubt - noise in deep shadows is managed every bit as effectively.</p>

<p>I've done blind-testing with high ISO 7D files, asking friends (including a couple of UK-based pro wildlife 'togs, one of whom is shooting with Mk IVs and the 5D Mk II for landscape and macro) to guess the camera and ISO: they normally get the camera because it's what I shoot; but nobody has been <em>even close</em> to getting that they're at 6400 and even 12800 ISO.</p>

<p>Not even the 5D Mk II shooter, which says it all for me.</p>

[number95]

<p>Zvia,<br>

Although lots of response (some of them are really not related to what you asked or a few misleading the facts/arguments) came, I can tell you that you can't make wrong either getting a 7D or 5D mkII, they are both gr8 cameras. They have many similarities just like your 40D, however, they also have some differences and serve for different purposes as well. If you get a 7D, the difference over 40D will not be as strong/tangible as you get a 5DmkII. That is for sure, no ifs no buts. You may like/dislike these differences however, it is a different thing to consider. I recommend keeping your 40D which is still a fantastic camera for the purpose and get a 5DmkII. By this way, you will have a still good APS-C camera, and on top of it, a fantastic FF body.<br>

5d mkII viewfinder as it is a FF, is almost 15% larger than 7d's viewfinder. Even for some, the viewfinder size is just enough reason to move to FF.<br>

7D is second fastest DSLR after 1D series. 5D mkII is slow however. But you already have a 40D which is also very fast (6.5 frames/sec), so moving to 8 f/s may not be the main reason there.<br>

Both cameras have fantastic image qualities, to me, the IQ of them (even including 40D) exceed the demands/abilities of 90% of the total DLSR users over the planet. Having said, if you are mostly concerned about big prints and/or excessive image quality, assuming you have a good technique and excellent glasses, a current FF is always better than a current cropped frame, no ifs no buts, it is the law of physics. It is not mainly the pixel size, no of pixels, but the area/size of the sensors of film plane. 35mm sensor has almost 2.5 times more size than APS-C, meaning 2.5 times more light for the same subject/framing. More information is always better to depict/describe a subject, and photography is not an exception. But don't expect to see tangible/realizable differences between two photos taken by a 7D and 5D plotted on an ordinary computer screen or on a small print with a mediocre printer.<br>

For me, I keep my 40D and my main body is a 5D mkII. Both are fantastic for the purpose. While photographing wildlife, sports, actions, mostly I go with 40D. The rest, 5dD mkII does it.<br>

Dont forget to invest on good lenses with either 7D or 5D mkII.<br>

Hakan</p>

 

[Dave Luttmann]

<p>Some good points Daniel. I'm interested in hearing from the few that talked of big and striking differences between the two cameras at the print size I mentioned. They seem to have fallen silent. Amazing what an image can do opposed to quoting LP/MM. ;-)</p>

[Dave Luttmann]

<p>Oh Daniel, as to what the psychology is behind the "must have FF," it has everything to do with the fact that people just like to parrot back what they've read on the net. They read an opinion from someone who actually has never made a comparison. They then become convinced after reading a few hundred such uninformed opinions that it must be true.</p>

<p>When confronted with the evidence by way of image, they either fall silent, or accuse you of doing something wrong via manipulations, etc. It seems to be very common. That sad part is how far the discussions go without the naysayer simply going and trying themselves. It's easier to simply state the other person is wrong as opposed to spending some time working with the images and actually learning something in the process. Many are simply too anxious to be seen as "right" or with the majority of other uninformed opinion holders, as opposed to expanding what they actually know and uncovering the truth.</p>

<p>Anyone who has the most basic post processing knowledge (and that number is fewer than many would think) understands that both cameras need different processing in post. This tends to result in slight differences between the two, but again, nothing that will really show in print. With good glass on each, both are capable of producing stunning 16x24 and 20x30 prints. The variation of output is actually greater from differences in the photographers capture ability and post processing ability then will ever appear from differences in the sensor....unless you shoot all your landscape at greater than 800iso ;-)</p>

<p> </p>

[keith reeder]

<blockquote>

<p> In a 16"x24" print, will there be a noticeable difference in quality to the naked eye between the 7D vs. 5D?</p>

 

</blockquote>

<p>As long as you don't mess up the conversion and PP of the 7D file, no there will not.</p>

[keith reeder]

<blockquote>

<p>Anyone who has the most basic post processing knowledge (<em>and that number is fewer than many would think</em>)</p>

</blockquote>

<p>That's becoming very obvious from some of the evidence on this thread..!</p>

[dcstep]

<p ><a name="00Y1k0"></a><a href="../photodb/user?user_id=2129602">Daniel Lee Taylor</a> <a href="../member-status-icons">said:</a></p>

<p >David - <em>Also remember that you can crop the 5D2 image so that the subject is the same size as on the 7D and there'll be approximately the same number of pixels on the subject. Hence, there'll be no loss in image quality, even after cropping.</em></p>

 

<blockquote>

<p>Am I misunderstanding what you're saying here? If you crop the APS-C portion of a 5D mkII file you will end up with an 8 MP image. The 7D is 18 MP. That's a big difference. For subjects like sports and wildlife you may not notice the difference in small to medium sized prints. But if you print large or have to crop further the difference will become apparent.</p>

</blockquote>

<p>Yes, you misunderstand. I was suggesting taking the full 5D2 file and cropping so that the subject is the same size as in the original 7D image. Some call this, "digital crop." You'll end up with essentially the same number of pixels on the subject.</p>

 

[dcstep]

 

<p ><a name="00Y1gN"></a><a href="../photodb/user?user_id=1055250">Geoff Francis</a> <a href="../member-status-icons"></a>, Jan 14, 2011; 12:31 a.m. said:</p>

 

<blockquote>

<p>Cropping is one thing and I don't want to enter into a debate about whether 5D crops match 7D crops, but I find having viewfinder framing that more closely matches the final picture better for composition than a viewfinder that covers a much bigger area and then having to think what it will look like heavily cropped.<br />I guess that is one of the reasons why I like having both and APS-C and a FF camera.</p>

</blockquote>

<p>I agree whole heartedly with this. I shoot birds and most wildlife with my 7D for essentially this reason. The 8-fps helps a lot also. I wouldn't suggest taking images with the 5D2 with the anticipation that you'd crop every one. However, if you take your 70-200mm and add a 1.4x TC on the 5D2, then you've got both the reach and the ability to crop.</p>

 

 

[dcstep]

 

<p ><a name="00Y1gN"></a><a href="../photodb/user?user_id=1055250">Geoff Francis</a> <a href="../member-status-icons"></a>, Jan 14, 2011; 12:31 a.m. said:</p>

 

<blockquote>

<p>Cropping is one thing and I don't want to enter into a debate about whether 5D crops match 7D crops, but I find having viewfinder framing that more closely matches the final picture better for composition than a viewfinder that covers a much bigger area and then having to think what it will look like heavily cropped.<br />I guess that is one of the reasons why I like having both and APS-C and a FF camera.</p>

</blockquote>

<p>I agree whole heartedly with this. I shoot birds and most wildlife with my 7D for essentially this reason. The 8-fps helps a lot also. I wouldn't suggest taking images with the 5D2 with the anticipation that you'd crop every one. However, if you take your 70-200mm and add a 1.4x TC on the 5D2, then you've got both the reach and the ability to crop.</p>

 

 

[matthijs]

<p>Dear Zvia,</p>

 

<blockquote>

<p>Overall look: Ok, so this is kinda hard to explain, but the 5D had a different look. Better bokeh and something else too....maybe depth or polish or something; not sure what to call it, but I liked it.</p>

</blockquote>

<p>I know exactly what you mean. Rationally the differences are minimal but after using my three camera's (XTi/400D, 50D and 5Dii) over a period of time I just favor the results of the 5Dii. My guess is that it's because of subtle color and contrast differences.</p>

<p>So I keep using all three. The XTi/400D because it's small and light and the results are way good enough for a "you never know what you see" or "take it to a work related event" camera. The 50D because it's so snappy for action and the 5Dii for artsy shots.</p>

<p>Have fun shooting,</p>

<p>Matthijs.</p>

<p>P.S. I wonder if you'll regret buying a 5Dii with a possible 5Diii on the horizon. A great lot of people are still very happy with their 5D classics and that camera is a lot older than the current 5Dii...</p>

[dcstep]

 

<p ><a name="00Y1jE"></a><a href="../photodb/user?user_id=1774085">Keith Reeder</a> <a href="../member-status-icons"></a>, Jan 14, 2011; 05:18 a.m. said:</p>

 

 

 

<blockquote>

<p>Following is a pre-dawn shot with the 5D MkII and the 70-200mm f/4L IS with the 1.4x TC at ISO 6400:</p>

<p>David, it <em>looks </em>NRd - obviously, and heavily - especially at <a rel="nofollow" href=" target="_blank">large size</a> on your Flickr stream. The fir tree on the left looks painted.</p>

<p>As soon as NR is that blatant, any argument that "<em>camera a is better at high ISO than camera b</em>" goes right out of the window because all we're looking at is the result of (DxO) NR, not the inherent ability of the camera, and DxO is a "blunt instrument" in NR terms.</p>

<p>Again, blurring away every hint of noise and detail proves nothing about a camera's noise capabilities. I can easily do that with my 7D, but I like detail, and for my pictures not to look like oil paintings.</p>

<p>Here's 6400 ISO from my 7D (in less light than your shot too, judging by the exif), resized to the size of the elk image: <a rel="nofollow" href="http://www.capture-the-moment.co.uk/tp/tfu29//upload/Lr-2382_2048.jpg" target="_blank">http://www.capture-the-moment.co.uk/tp/tfu29//upload/Lr-2382_2048.jpg</a></p>

<p>I know which one looks better to me...</p>

</blockquote>

<p>First, my shot was an attempt to show Zvia what she could do with a 5D2, a 70-200mm f/4L IS and a 1.4x TC before dawn. I hope that I succeeded.</p>

<p>Everyone is welcome to pixel-peep, but your comparison is laughable, not just because the lion doll is stupid, but because it's evenly lit in a studio with a much wider lens. It's not challenging the camera.</p>

<p>Many of us call that OOF in my image "bokeh". When you shoot a telephoto lens wide open, this "painterly" pattern can show. It'll vary from lens to lens, but this is typical of my 70-200mm.</p>

<p>This image has not been hit hard with NR. Look at the antlers at 100% and you'll see chroma because I did NOT process that all out. This was meant for internet viewing on Flickr and, at the time, I didn't have a software that would NR without too much loss of detail. Also, based on the file size, it looks like I cropped this one around 20% to frame the animal. (There's that digital crop we've been talking about). BTW, the 7D was on the car seat beside me. I wasn't thinking "comparison test" so I didn't tranfer the lens over to do a comparison. Sorry...</p>

<p>Taking pictures of dolls in even light can be a valid comparison when you shoot the same shot with both cameras and examine them at 100%, but comparing a controlled studio shot to a hand held, pre-dawn, uneven light, telephoto shot and reaching a conclusion, is folly.</p>

<p>I'm please to see that Zvia noted the elk shot and took it for what it was meant to convey, the possibility of shooting useful shots in an imperfect world.</p>

<p> </p>

 

 

 

[keith reeder]

<p>All the shot showed is what heavy DxO NR does to 5D Mk II images, David - it tells us absolutely nothing whatsoever about the 5D Mk II.</p>

<p>Furthermore, a 7D image with that much NR on it would look <em>exactly the same</em>, so again, your picture says nothing about the 5D Mk II.</p>

<p>And - personally - I don't think it's a look to aspire to at all.</p>

[jim_ducey]

<p>Ziva, Canon does have a camera that is full frame with the same focusing system as the 7D, it is the 1Ds Mark lll. You can look at the Canon site and do the comparison, but the price is a deal killer.<br>

Jim</p>

[number95]

 

<p ><a name="00Y1n7"></a><a href="../photodb/user?user_id=5442351">Jim Ducey</a> <a href="../member-status-icons"></a>, Jan 14, 2011; 10:02 a.m.</p>

 

<p>Ziva, Canon does have a camera that is full frame with the same focusing system as the 7D, it is the 1Ds Mark lll. You can look at the Canon site and do the comparison, but the price is a deal killer.<br />Jim</p>

<p>It must be a joke. They have completely different focusing system.</p>

 

 

[dcstep]

 

<p ><a name="00Y1n4"></a><a href="../photodb/user?user_id=1774085">Keith Reeder</a> <a href="../member-status-icons"></a>, Jan 14, 2011; 10:01 a.m.</p>

 

<blockquote>

<p>All the shot showed is what heavy DxO NR does to 5D Mk II images, David - it tells us absolutely nothing whatsoever about the 5D Mk II.<br>

Furthermore, a 7D image with that much NR on it would look <em>exactly the same</em>, so again, your picture says nothing about the 5D Mk II.<br>

And - personally - I don't think it's a look to aspire to at all.</p>

</blockquote>

<p>Well Keith, you can denigrate my image all you like. Once again, I didn't put it up to show an example of something that's going to make National Geographic, but something that Zvia might like to do herself some early morning or after sunset.</p>

<p>DxO did apply a Default level of NR to that image, but, as I've pointed out, I didn't run it through their "High-ISO" noise reduction module or do anything extra ordinary. The pixel-peepers can see plenty of noise remaining if they look at 100% and look particularly at the antlers. The BG is in bokeh, so it's very hard to evaluate that. The bokeh is not due to NR, but due to using 280mm focal length wide open.</p>

<p>When I try the 7D in these situations and even better lighting, at ISOs like 1600 and 3200, I more often get a "painted" looking background. I need more +EV than with the 5D2 and it shows up in more of my 7D images than in my 5D2 images, at least that's my impression from having tried it. This is in the field and not in a test, studio situation, typically using a long focal length, up to 700mm and at least 200mm. I'm not sure what's happening, but I find that using the 5D2 and +1 or +2EV in these situations yields useable, enjoyable images.</p>

<p>I'm not ashamed that this particular image isn't NG quality. Sierra Club, Cornell Ornithology Labs and others have licensed enough of my images that I know that I'm doing at least a few things right. All I did here was show a real world, tough conditions shot, so that Zvia and others that are truly interested could see the 5D2 "at work" with a 70-200mm and a 1.4x TC.</p>