agree with the pixel density reason, we need a efs 30mm that is 1.6x sharper than 50 f1.4 on 7d to do the comparison with 5dII?.
And the EF 50 1.4 USM isn't exactly a killer performer, so the designers won't have to injure themselves trying. Of course, the 15MP 50D files were uprezzed to 18MP and they looked slightly better than the 7D from ISO 100-800. Maybe a little softer lens should be used for the 50D--just to make things fair and even.
The 5D Mark II images looked significantly sharper with better local contrast at all ISO. In theory, there should not be a great difference in perceived resolution between the 18MP sensor of the 7D and 21.1 MP sensor of the 5D Mark II. I do not know why the 5D Mark II images look sharper, but if I were to guess, I think maybe the 5D Mark II has a significantly weaker anti-aliasing filter than the 7D and that could be the reason why the 5D Mark II images looked sharper.
There are a couple of things I've observed in my own testing: If using DPP, at the default settings some NR is applied to 7D files even at ISO 100. This is not the case for 5D2 files. However, I think this only accounts for part of the difference.
The other factor seems to be RAW conversion algorithms. For some reaosn, Capture One Pro 5 seems to produce significantly sharper output from the 7D, with better micro-contrast, at low ISO than DPP with all NR disabled. See alundeb's comment above; there is also some discussion of this on this thread: https://www.fredmiranda.com/forum/topic/829160/0?keyword=capture,one#7700355
Based on this I'd certainly recommend that all 7D owners at least take a look at the the Capture One 5 trial.
Roland Lim wrote:
I have finally finished my user review of the 7D. It has taken much longer than I originally anticipated due to various factors. Those who have had my previous reviews on the 40D and 50D will know my style of review. I hope it will help those still on the fence.
Nope AGeoJO. I want video, a high-res display, toggle-able grid lines in view-finder on the new transmissive display etc in a small form factor. 1DsMKIII is also a little much money for some of us, and starting to feel a bit dated in terms of pure "technology."
Now the 1Ds MKIV, whenever it shows up, is calling my name... must... resist....
I can already hear sounds of fire (flames everywhere) but can we please have a simple camera with good FF sensor, good focus, good viewfinder and nothing else (for instance - no video)!?
1Ds Mark III ohhhh now it's too expensive....what is your excuse this time ?.
mfurman wrote:
I can already hear sounds of fire (flames everywhere) but can we please have a simple camera with good FF sensor, good focus, good viewfinder and nothing else (for instance - no video)!?
kewlcanon wrote:
1Ds Mark III ohhhh now it's too expensive....what is your excuse this time ?.
As a matter of fact it is on top of my list but still a bit too expensive for me and too big. I hear some complaints about outer focus points and fast lens focusing issues.
Well that's it....I guarantee you won't buy ANY camera for the next 20 years...well maybe P&S.
mfurman wrote:
As a matter of fact it is on top of my list but still a bit too expensive for me and too big. I hear some complaints about outer focus points and fast lens focusing issues.
n0b0 wrote:
I wonder if perhaps the 5D2 samples in the review look noticeably sharper because it's been sized down to 7D resolution?
I knew I should not have been lazy, and should have made a set of comparison with 7D upsized to match 5D II resolution. I have actually made such a comparison on my computer, and I can tell you the results I got was just the same, but 5D Mark II crops still looked significantly sharper than those from 7D. I will try to add that section in the review later when I have time.
alundeb wrote:
A very extensive review, Roland, thanks for posting!
I have a couple of questions, hope you have time to answer
What NR and sharpening settings did you use in the RAW conversion?
In the comparison vs 5DII, ISO 100, how would you explain that the white on red details in the background of the crops look sharper for the 7D than the 5DII? Or maybe I'm wrong. The details in the red in the background, lower left, looks at least as detailed for the 7D, but more sharpened for the 5DII.
Thanks in advance.
No sharpening and noise reduction were applied to any of the images during and after RAW conversion for all 3 cameras.
The red background was a red color table runner and yes, on the 7D crops, the red table runner did looked sharper. The reason is simple. The same lens on the 7D when moved back to a point to allow the same FOV with 5D Mark II has more depth of field. The 5D Mark II has appreciably shallower DOF even at f8.
pity still can not open the link you show above...
Unfortunately, main China blocks many of blog sites like wordpress. However, you get try use one of the many free proxy sties around to bypass the block and you will then be able to read my review or almost any other sites blocked by China.
There are a couple of things I've observed in my own testing: If using DPP, at the default settings some NR is applied to 7D files even at ISO 100. This is not the case for 5D2 files. However, I think this only accounts for part of the difference.
The other factor seems to be RAW conversion algorithms. For some reaosn, Capture One Pro 5 seems to produce significantly sharper output from the 7D, with better micro-contrast, at low ISO than DPP with all NR disabled. See alundeb's comment above; there is also some discussion of this on this thread: https://www.fredmiranda.com/forum/topic/829160/0?keyword=capture,one#7700355
Based on this I'd certainly recommend that all 7D owners at least take a look at the the Capture One 5 trial....Show more →
Capture One Pro 5 was only released after I finished processing all the comparison shots. So I did not use it for the review. However, I have tested it myself and although 7D files processed by Capture One Pro 5 does llok a bit better at 100% view than by DPP, especially concerning local contrast, the 5D Mark II crops processed by Capture One Pro still look sharper.
There is a major confusion going on about the resolution and sharpness of the 7D.
Most of this confusion is related to RAW converters like DPP and ACR that renders images from the 7D soft because there is something not yet fully understood about the green channels spectral response difference.
The microcontrast can be enhanced by sharpening in posts, but testers don't do that, because they need to keep the baseline for comparison. For the 7D this goes wrong, because of the special algorithms required to handle the green channel imbalance.
The only review so far that doesn't destroy the inherent sharpness of the 7D, is from imgaing-resource.com. They have RAW files available for download.
Then most tests are done at aperture F8 for both FF and APS-C cameras. The sharpness of the 50/1.4 peaks at F/4, this is measured by photozone.de even with an 8MP APS-C sensor. The sharpness of the 70-200/4 IS peaks at F/5.6.
Testing APS-C cameras at F/8 introduces loss in sharpness due to diffraction, and biases the results in favor of FF cameras. Even imaging-resource.com does this.
Don't believe everything you see, the 7D has full 18MP resulotion. You need to use lenses that ares sharp enough, at a small enough aperture, and processed with software that doesn't stumble over the green channel imbalance.
alundeb wrote:
Most of this confusion is related to RAW converters like DPP and ACR that renders images from the 7D soft because there is something not yet fully understood about the green channels spectral response difference.
Canon is the manufacturer of both DPP and 7D. They are in the best position to fully understand the green channels spectral response difference. However, it seems the DPP images are still soft.
If Canon's software cannot deliver best IQ, we could presume the problem lies not with the software, but with the hardware.
?