S Dilworth wrote:
I notice that the 5D Mark III does a better job of suppressing the false colour while retaining detail in the JPEGs, compared to the 5D Mark II. That's the new DIGIC in action, I suppose.
Yeah, I was surprised. I had kind of written off 5D3 jpgs as utter crap after seeing the early examples, but at least some things about the engine are definitely better, CA suppression (I had it off here and it wouldn't make a difference anyway center frame with a 300 2.8), it fights moire better and does a cleaner de-bayer it seems and extracts more fine detail actually (surprising since all the talk has been about how it makes softer jpgs). Some have said that using Neutral turns off some of the extra hidden NR on the 5D3, maybe it's because I used neutral?? Or it might just be that the 5D3 jpg engine only softens things up when the details have less contrast, these had extreme contrast.
Now I'm wondering how the 7D would look if you shot that banknote again, this time from about 1.6× farther away.
splathrop wrote:
I would like to see some equal-size big enlargements. By equal-size I mean the images scaled to the same dimensions, not the same megapixels.
I did do that for the ISO6400 shots. I will post them later.
7D shows better detail, less artifacting than from either and a little less noise than from 5D2 in that scenario. I think it showed a little more noise than from the 5D3 though.
pKai wrote:
Are you saying the 5D3 has faster AF than the 1D4? What about the loss of 4 FPS?
.
Your confusing faster AF with fps. They are not the same.
FWIW I have tried to like the 7D, owning numerous copies, but I didn't like the IQ compared to my 1DMK3 must less my 1DMK4.
I would love to use the 1.6x for wildlife and birds but the 7D is going to have to be improved in alot of areas for me to consider giving up my 5DMK3 or1DMK4 for one.
I shoot mostly at dawn and dusk so I never see ISO 100. I'm lucky to see ISO800 and the 7D doesn't compare to the others past ISO800. The noise, even on properly exposed shots, starts to destroy any resolution advantage offered by the pixel density of the 7D
gdanmitchell wrote:
Online comparisons of differences between cameras using different sensor formats are fraught with problems. A few examples:
Comparing 100% crops is equivalent to comparing flim/or slides at different - sometimes very different - magnifications.
umm that was the whole point for the 7D vs 5 cameras, to show that the 7D when viewed at a different mag does show a heck of a lot more detail grabbed and does have the potential to give you a lot more effective reach
How do you account of the difference in lens performance on the different format cameras? A lens (at FL and apertures you select) will have X line pairs/millimeter resolution - but the two systems have different line pair per picture width resolution as a result. (All else being equal, the larger format system always has the potential to resolve more detail.)
Why do I want to account for it? The whole point here is to let everything play out as it does.
I'm not trying to compare sharpness of different lenses using different bodies to test different lens or something.
In the end, there are great cropped sensor systems and great reasons for many photographers to choose them. There are also great full frame systems and wonderful reasons for some photographers to chose them. (The fact is that the large majority of people buying DSLRs today will get better results than they need with either format.)
Rather than getting into the impossible "Crop is better than FF! No it isn't! FF is better than crop!" silliness - which never leads anywhere useful in the end - it is a whole lot more useful to understand the ways that the two systems operate and the objective facts regarding their different performance, at least if you are doing a sort of photography in which it even matters. ...Show more →
Who was getting into silly FF vs crop rants??
I exactly did compare it in a real world scenario. I was comparing them how they might be used. You'd yse a 5D2 and 5D3 the same way and a 7D when distance limited.
But I guess wildlife photography is not real photography that even matters.
5D2 and 5D3 will deliver basically the same detail. 7D has potential to deliver a lot more WHEN REACH LIMITED. Of course when not reach limited and framing same FOV the 7D will show a little bit less total detail and more noise.
For example, some benefits of cropped sensor cameras can include: lower cost for a camera with otherwise equivalent features, faster burst rate in lower cost cameras, the potential to use smaller and (in some cases) less expensive lenses for the same tasks, excellent image quality that exceeds the needs of almost all DSLR photographers, the potential to "fill the frame" more with small or distant subjects, etc.
The thing is "fill the frame" more isn't really an advantage. A 10D can fill teh frame more with the subject than a 5D3 but it will put less detail on it so what the heck good does that do? How is that an advantage? It's not an advantage for noise or detail. That said, the 7D clearly can have a sizeable advantage than a 5D2 when shooting stuff too far away to frame fully at FF, it has much denser photosites and the lens/AA/etc. do NOT conspire to make the advantage zero or near zero.
There is no "better" in a general universal sense, any more than a truck is better than a compact car or vice versa. It is a question of which tool is most suited to the tasks at hand.
Dan
When has anyone in this thread said anything about crop or FF are universally better
You do realize I owned both side by side for 4 years....
Jim Victory wrote:
I shoot mostly at dawn and dusk so I never see ISO 100. I'm lucky to see ISO800 and the 7D doesn't compare to the others past ISO800. The noise, even on properly exposed shots, starts to destroy any resolution advantage offered by the pixel density of the 7D
Jim.
I thought it did starting at ISO1600 or so, but if you expose well and a subject that is fairly light in color I actually found that even compared to the 5D3 even at ISO6400! the 7D still maintains a sizeble resolution advantage. For a black crow on a sidewalk maybe not, but for a mockingbird on pavement it would.
If you care about noise above all it does seem a little worse than 5D3 after normalization but it's better than 5D2 even at ISO6400.
gallery with giant images from the ISO6400 'test' ('test' because it was a quick after thought, things are not normalized for different gains per ISO between cameras, ACR cooks books slightly different for each camera in terms of just about everything, our power doesn't seem to be 100% stable in the summer and I noticed that same exposure, same camera, same everything that the image brightness changed slightly shot to shot and when I did the rescales to 8MP I didn't first apply filters and I didn't use the most advanced re-scale algorithm just a simple photoshop bicubic and nothing more, etc. etc. so the results a bit rough):
the image titles tell you which shot from which camera
when it says "full" that means the full size image with no re-sizing
when it say 8MP that means it was downsampled to 8MP using just a simple bicubic with no pre-filtering
when it says (for the 7D shot) at FFScale that means I very roughly tried to rescale the bill shot by the 7D down to the size to match the one shot by the 5D2/5D3, not quite right, but close enough
The 5D3 is put at a bit disadvantage here as it applied more gain than the 5D2 and especially 7D.
Anyway it's a super rough 'test', still you can get the idea that the 5D3 will have just slightly less noise than the 5D2 will have less noise than the 7D when compared at normalized scale. The larger difference between 5D3 and 5D2 might be in the high ISO DR since the shadows don't start banding and they retain detail a bit better. Anyway it's not really a great test for this. I ws just doing a sanity check to make sure my 5D3 wasn't bum (someone on another forum got a copy that had 1.5 stops worse than 5D2 for noise) and more looking into whether you might still ahve reach advantage at very high iso with 7D and a bright subject.
Here is a comparison of ISO6400 reach for a bright, well-lit, high contrast subject (the reach advantage for 7D will be less if the subject has dark tones or is in a darker part of the image or you underexpose in general, while it's still very clear, surprisingly, for a bright subject well lit it might be barely there for a black crow in a dark corner of an image most likely; I may do mid-tonality reach comparison later since many subjects might fit that scenario):
top image is the 7D re-scaled to match the bill size on the 5D2 - you notice that it delivers BOTH less noise, less artifacts and more detail than the 5D2 and slightly more detail and less artifacts than the 5D3 with maybe a touch more noise
second image from top is the 5D2 - you notice it retains a bit less detail than the 5D3 bill
thrid from top is the 5D3 bill - you notice that it and the 5D2 show a less detail than the 7D bill even at ISO6400, the 5D2 shows tons less
bottom is the native size 7D bill - bright subject, well-lit the 7D still has a sizeable reach advantage over the 5 series cams, especially compared to the 5D2
but again for a subject that is not so super contrasty and is made up of many tones much farther from white the reach advantage may prove to be much less
if you use firefox make sure to click all of these images to make sure you are seeing the full 100% view
and note that the 5D3 looks brightest and the 7D darkest so at ISO6400 the bodies are probably not applying the same gain so the effective ISO compared here is highest for 5D3 and lowest for 7D most likely which would give 5D3 a slight disadvantage and the 7D a slight advantage
I will do an ISO6400 test with a darker, less contrasty subject to see if the 7D still shows reach advantage there or not and to see if the 5D3 pulls ahead more compared to 5D2 and then be done with it.
If you want to test performance differences between cameras, 100% crops are just about useless for such comparisons. That was my main point. (If you consider how large the compared images would be if printed "at 100%" - which we could take to mean the same resolution value in the range of typical screen resolutions - I think you'll see the main reason this is the case.)
When it comes to the resolution aspect of image quality, if you want to test functional performance for your photography, there are two important ways you can evaluate samples and make useful comparisons among them:
1. Make prints at the sizes you anticipate producing, using all of your normal, optimized post-processing workflow. If you will print up to 13" x 19", then evaluate carefully made 13" x 19" prints from the systems under consideration. If you regularly print at 24" x 36", then evaluate prints at that size, or at least small sections cropped from such large size print files. This will produce images that reflect the real-world best that the systems you are looking at can produce.
Or...
2. If you are a person who doesn't print much or at all, simply evaluate the images at the largest realistic image size at which you will potentially view them. If that means full screen on a 19" monitor, then compare that way. If you regularly share your photographs on a 30" computer monitor or on a HD TV monitor, compare that way.
p.2 #10 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
Thanks for all these tests skibum5.
I've yet to do any real testing like you have done, but I still keep getting the impression when I shoot my 7D back to back with my 5D3 that the 5D3 just seems to pull more detail and seems better to crop the 5D3 vs shoot the 7D native?? I guess this is what you were trying to prove/disprove to start with. Again, I've never done any real tests. The closest test I did was shoot a very far off eagle perched on a tree with both the 5D3 and 7D. I shot about 3-4 shots with both cameras from a tripod using the 100-400 lens at 400mm f/5.6. This eagle was filling maybe 5% of the frame even on the 7D and obviously less on the 5D3. Never was going to be a shot one would keep but upon reviewing the images the 5D3 looked equal to the 7D once I cropped them both to the same size (this involved cropping the 7D also but the 5D3 much more).
Because of this basic test I did and my overall impression on shooting landscapes and seeing much better small tree detail from the 5D3, I was starting to write off the 7D and only wanting to shoot my 5D3 and 5D2. I really need to go back and seriously evaluate the 7D as I have some amazingly detailed shots from it when shooting the 100-400 very close to the lens's MFD. But any type of landscape shot or when the subject is small in the frame the 7D just lags behind what I've seen out of my 5D2 and 5D3??
The one reason I always pick up my 5D3 now over the 7D when I go out birding is that the AF is just so superior. With strong confidence I can say that trying to focus on small fast flying birds (kingfishers and swallows), the 7D is next to useless, the 5D2 is completely useless and the 5D3 is amazing (don't want to think about how the 1DX would do!! Too big of a hit to the bank account)
Bottom line is I think after seeing your tests, I need to go and reevaluate my 7D vs my 5D3 to see if I should start using it more for larger birds or mammals that are far away. For small birds, there is no way I will go back to the 7D. I have been seriously considering selling the 5D2 and 7D to either buy another 5D3 or a 1D4. So many questions and decisions to be made.
p.2 #11 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
Jim Victory wrote:
Your confusing faster AF with fps. They are not the same.
FWIW I have tried to like the 7D, owning numerous copies, but I didn't like the IQ compared to my 1DMK3 must less my 1DMK4.
I would love to use the 1.6x for wildlife and birds but the 7D is going to have to be improved in alot of areas for me to consider giving up my 5DMK3 or1DMK4 for one.
I shoot mostly at dawn and dusk so I never see ISO 100. I'm lucky to see ISO800 and the 7D doesn't compare to the others past ISO800. The noise, even on properly exposed shots, starts to destroy any resolution advantage offered by the pixel density of the 7D
My question above dealt with the 5D3 VS the 1D4, not the 7D. You said in your previous post: "I already
have the 5D3 and am very pleased with the improvement in IQ AND AF performance compared to the 1D4 and 5D2, both of which I use."
You are basically saying that the 5D3 performs better than the 1D4 -- which I find surprising...... better IQ, I can believe, UNLESS you are forced to crop the 5D3 image to get the same FOV as the 1D4.... or the 7d for that matter.... I've tried this with the 5D2 and found that a cropped 5D2 image to 7D FOV is visibly poorer in several areas than a native 7D image.
I am well aware that both the 1D4 and 5D3 outperform the 7D. I am also well aware of the difference between AF and FPS..... Both should be much faster on the 1D4 versus all the other bodies mentioned.
p.2 #12 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
I only directed this sentence to your statement. "Your confusing faster AF with fps. They are not the same."
Everything else was a response to the general discussion about the 7D and its performance outlined by skibum's test. It had nothing to do with anything you said.
Your quoting someone elses post in your narrative but with my quote as a subset. I didn't state any comparison of the 5DMK3 or 1DMK4. However I have both and the 5DMK3 does have a better AF than the 1DMKIV. It acquires focus faster in general and better in low light and low contrast situations.
"Comparing a lens without TC on 7D vs lens with 1.4x TC on FF, the 7D gives slightly better image quality - more detail and less chromatic aberration. "
p.2 #14 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
gdanmitchell wrote:
If you want to test performance differences between cameras, 100% crops are just about useless for such comparisons. That was my main point. (If you consider how large the compared images would be if printed "at 100%" - which we could take to mean the same resolution value in the range of typical screen resolutions - I think you'll see the main reason this is the case.)
have you EVER shot wildlife or at least sports
1. Make prints at the sizes you anticipate producing, using all of your normal, optimized post-processing workflow. If you will print up to 13" x 19", then evaluate carefully made 13" x 19" prints from the systems under consideration. If you regularly print at 24" x 36", then evaluate prints at that size, or at least small sections cropped from such large size print files. This will produce images that reflect the real-world best that the systems you are looking at can produce.
so to compare the 5D2 and 5D3 I was to print out a 13x19" from each for each and every member and mail one to them all
i was just checking if the rumors that the 5D3 deliver crisper and more detail because of lesser AA were true or if the rumors that it was softer because of stronger AA and split-greens and so on were true, it seems to me that neither rumor is true, at least not to any reasonably real world degree
and as for the 7D I was just checking if it gets you more reach, I already knew it did at lower ISO, but hadn't done such a careful test of it before and hadn't tried it on bright objects at super high iso in comparison
p.2 #15 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
arbitrage wrote:
Thanks for all these tests skibum5.
I've yet to do any real testing like you have done, but I still keep getting the impression when I shoot my 7D back to back with my 5D3 that the 5D3 just seems to pull more detail and seems better to crop the 5D3 vs shoot the 7D native?? I guess this is what you were trying to prove/disprove to start with. Again, I've never done any real tests. The closest test I did was shoot a very far off eagle perched on a tree with both the 5D3 and 7D. I shot about 3-4 shots with both cameras from a tripod using the 100-400 lens at 400mm f/5.6. This eagle was filling maybe 5% of the frame even on the 7D and obviously less on the 5D3. Never was going to be a shot one would keep but upon reviewing the images the 5D3 looked equal to the 7D once I cropped them both to the same size (this involved cropping the 7D also but the 5D3 much more).
Because of this basic test I did and my overall impression on shooting landscapes and seeing much better small tree detail from the 5D3, I was starting to write off the 7D and only wanting to shoot my 5D3 and 5D2. I really need to go back and seriously evaluate the 7D as I have some amazingly detailed shots from it when shooting the 100-400 very close to the lens's MFD. But any type of landscape shot or when the subject is small in the frame the 7D just lags behind what I've seen out of my 5D2 and 5D3??
The one reason I always pick up my 5D3 now over the 7D when I go out birding is that the AF is just so superior. With strong confidence I can say that trying to focus on small fast flying birds (kingfishers and swallows), the 7D is next to useless, the 5D2 is completely useless and the 5D3 is amazing (don't want to think about how the 1DX would do!! Too big of a hit to the bank account)
Bottom line is I think after seeing your tests, I need to go and reevaluate my 7D vs my 5D3 to see if I should start using it more for larger birds or mammals that are far away. For small birds, there is no way I will go back to the 7D. I have been seriously considering selling the 5D2 and 7D to either buy another 5D3 or a 1D4. So many questions and decisions to be made....Show more →
I'd think it would be even better for small birds since they are likely to be even harder to fill the frame with. My warblers up in tree tops def show more detail with 7D than 5D3.
But for landscapes where you are free to frame, yeah for sure the 5D2 is better than the 7D.
p.2 #16 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
Jim Victory wrote:
FWIW I have tried to like the 7D, owning numerous copies, but I didn't like the IQ compared to my 1DMK3 must less my 1DMK4.
I would love to use the 1.6x for wildlife and birds but the 7D is going to have to be improved in alot of areas for me to consider giving up my 5DMK3 or1DMK4 for one.
I shoot mostly at dawn and dusk so I never see ISO 100. I'm lucky to see ISO800 and the 7D doesn't compare to the others past ISO800. The noise, even on properly exposed shots, starts to destroy any resolution advantage offered by the pixel density of the 7D
Certainly not my experience. I found the 7D's IQ better than the Mk3 and on par with the Mk4. As for shooting at ISO 1600, I had no problems with noise on the 7D. Any noise cleaned up beautifully with Neat Image.
The only reason I went from a 7D to a Mk4 was because of the more consistent AF accuracy in AI Servo.
If a 7D2 or 70D comes along with improved AF, I will sell my Mk4 and get one.
p.2 #17 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
Imagemaster wrote:
Certainly not my experience. I found the 7D's IQ better than the Mk3 and on par with the Mk4. As for shooting at ISO 1600, I had no problems with noise on the 7D. Any noise cleaned up beautifully with Neat Image.
The only reason I went from a 7D to a Mk4 was because of the more consistent AF accuracy in AI Servo.
If a 7D2 or 70D comes along with improved AF, I will sell my Mk4 and get one.
Tony
I agree with the AF comment. The 7D is pretty good - a really busy background will fool it though whereas I never had this issue, even with my 1DM2.
p.2 #18 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
Jim Victory wrote:
I only directed this sentence to your statement. "Your confusing faster AF with fps. They are not the same."
Everything else was a response to the general discussion about the 7D and its performance outlined by skibum's test. It had nothing to do with anything you said.
Your quoting someone elses post in your narrative but with my quote as a subset. I didn't state any comparison of the 5DMK3 or 1DMK4. However I have both and the 5DMK3 does have a better AF than the 1DMKIV. It acquires focus faster in general and better in low light and low contrast situations.
p.2 #19 · 5D2 vs 5D3 vs 7D for detail/reach/noise:
Skibum, I am following your posts with interest. I want a good reason to upgrade, but having a 7D and 5D2, am not finding your posts all tooo helpful with that pursuit. 7D is still making a strong showing...
)
7D is still making a strong showing...