lct Posted February 17, 2008 Share #1  Posted February 17, 2008 Advertisement (gone after registration) M8 vs Nikanopus sharpness by Chasseur d'Images  The last iteration of the French mag Chasseur d'Images (#301, 15 Feb. 2008, page 133) compares various digital bodies in terms of sharpness and resolution (so called 'fine detail rendition') taking into account smearing and various digital artefacts. Not sure how they do this actually (raws? jpgs? with what lenses?) and there are apparently mistakes in the resolution charts but the sharpness results might interest some M8 users possibly. FWIW.  Link to post Share on other sites More sharing options...
Advertisement Posted February 17, 2008 Posted February 17, 2008 Hi lct, Take a look here M8 vs Nikanopus sharpness by Chasseur d'Images. I'm sure you'll find what you were looking for!
robsteve Posted February 17, 2008 Share #2 Â Posted February 17, 2008 It is not surprising that the M8 did so well at 640iso. This is a very usable speed, with a little noise, but not much noise reduction to soften the image. Â Robert Link to post Share on other sites More sharing options...
carstenw Posted February 17, 2008 Share #3 Â Posted February 17, 2008 It does surprise me that the M8 placed 4th at ISO 160. I wonder what went wrong there. It must have been a really close fight, but still, I would have expected the AA filters of the other cameras to have penalised them more. The E-3 apparently sucked across the board, in comparion, which also surprises me. I have seen D300 and E-3 shots compared, and the D300 shots are way softer, so they must have really pushed the sharpening. I wonder which camera shows more detail when you take M8 and E-3 shots unsharpened and then D300 shots and sharpen them. Maybe there might even be some of the 'false' detail which Simon claims for the Foveon. Link to post Share on other sites More sharing options...
sdai Posted February 17, 2008 Share #4 Â Posted February 17, 2008 It does surprise me that the M8 placed 4th at ISO 160. I wonder what went wrong there. It must have been a really close fight, but still, I would have expected the AA filters of the other cameras to have penalised them more. The E-3 apparently sucked across the board, in comparion, which also surprises me. I have seen D300 and E-3 shots compared, and the D300 shots are way softer, so they must have really pushed the sharpening. I wonder which camera shows more detail when you take M8 and E-3 shots unsharpened and then D300 shots and sharpen them. Maybe there might even be some of the 'false' detail which Simon claims for the Foveon. Â Makes you wonder what the guys in Paris are smoking, right? Link to post Share on other sites More sharing options...
billh Posted February 18, 2008 Share #5 Â Posted February 18, 2008 Hi LCT, Â Did they say which lenses they were using? Also, did they mention the possibility of focus inaccuracies with the rangefinder? Link to post Share on other sites More sharing options...
lct Posted February 18, 2008 Author Share #6 Â Posted February 18, 2008 ...Did they say which lenses they were using? Also, did they mention the possibility of focus inaccuracies with the rangefinder? Hi Bill, no just comparative charts showing quite homogeneous results from 160 to 1250 iso for the M8. At 160 iso for instance, the M8 (5.2 pts out of 10) is outclassed by the D3 (7.2) and the 1Ds MkIII (6.2) at 200 iso, but its results remain very close at 320 (5.2), 640 (5.5) and 1250 iso (5.1). Chasseur d'Image's comment is the M8 delivers fine and detailed images from 160 to 1250 iso but its smaller sensor limits the sharpness of very structured details as well as resolution of very fine ones which tend to be 'ignored' by the sensor. My feeling is Chasseur d'Image compare fine jpegs at standard in-camera settings but i'm not 100% sure of this because i don't read this mag regularly. Link to post Share on other sites More sharing options...
jrgeoffrion Posted February 19, 2008 Share #7 Â Posted February 19, 2008 Advertisement (gone after registration) Makes you wonder what the guys in Paris are smoking, right? Â Not really. The reason the M8 is doing more poorly at ISO 160 is that it's "only" got 10MP to capture fine details while the 1Ds III has 22MP. As noises and artifacts increase, the effective resolution of the 1Ds III (and others) are diminished -- which reduces the difference between MP... Link to post Share on other sites More sharing options...
pascal_meheut Posted February 19, 2008 Share #8 Â Posted February 19, 2008 No mention of the lens used but with the M8, it does not matter. All the one tested by CdI had the maximum note. I guess the reason for the relative performance of the M8 is the fact that they test in JPEG. And we know that Leica's implementation is not the best. I also read that the M8's jpeg compression tries to act like an AA filter. Â BTW, I know the guys doing the tests for CdI personnally and they are nice, competent and always willing to discuss their procedure when not insulted first. And they have some reasons not to use RAW: because it depends on the software, on the sharpening chosen... Also, remember that most of their readers are normal "amateurs", shooting mainly in jpeg. So their target is not to show what can be extracted from a body after hours of post-processing but what one can get "out of the box". Of course, this approach can be critisized when testing high-end cameras but the other ones have their flaws too. Â Another example: when testing sharpness, they compute the results based on an A3 print. At this size, the difference between a D3 and an Eos1DsIII is not obvious. But once again, many of their readers will not print at this size very often and almost never in A2 or A1. Â Anyway, they were excellent when testing lenses and bodies in the old analog times. They try to reach the same level in the digital world but this is even more complex and even if I appreciate them, I am now more careful when reading their conclusions. Â P.S: they are not in Paris BTW. Link to post Share on other sites More sharing options...
pascal_meheut Posted February 19, 2008 Share #9 Â Posted February 19, 2008 Not really. The reason the M8 is doing more poorly at ISO 160 is that it's "only" got 10MP to capture fine details while the 1Ds III has 22MP. As noises and artifacts increase, the effective resolution of the 1Ds III (and others) are diminished -- which reduces the difference between MP... Â You are right of course but the 12 MP D3 gets results almost as good as the Eos1DsIII in their test. So part of the M8 performance comes from the jpeg mode. Â Nice to see you here again BTW. Link to post Share on other sites More sharing options...
sdai Posted February 19, 2008 Share #10 Â Posted February 19, 2008 And they have some reasons not to use RAW: because it depends on the software, on the sharpening chosen... Â With all due respect to your friends at CDI, Pascal ... this doesn't make sense and seems to be contradicting their belief as well. Â If they compare camera produced JPEGs then they're only comparing the in camera algorithms. Â Also, how do they judge "sharpness"? simply by visual inspection or by calculating SFR? Â If by calculation, there's gonna be some numbers to convince their readership, right? Link to post Share on other sites More sharing options...
pascal_meheut Posted February 19, 2008 Share #11 Â Posted February 19, 2008 I'm well aware of the limitation of their method and as I said, I do not find it adequate for high-end cameras. Â But I know of no way to compare RAW results without introducing bias too. Â BTW, they do not inspect visually but use a battery of automated tools including the ones from DxO. Link to post Share on other sites More sharing options...
carstenw Posted February 19, 2008 Share #12 Â Posted February 19, 2008 One way of comparing JPGs is simply to use the same RAW developer. It could be said that the profiles are not all equally good, but neither are the JPG engines in the cameras, so that isn't really a solid argument. There is no need to spend hours processing. A simple conversion with default values in ACR would be fair enough. Link to post Share on other sites More sharing options...
pascal_meheut Posted February 19, 2008 Share #13 Â Posted February 19, 2008 Life would be so simple if one could use the solutions proposed by those who never had to deal with the problem in the first place. Link to post Share on other sites More sharing options...
carstenw Posted February 19, 2008 Share #14 Â Posted February 19, 2008 Are you commenting on my reply? It seems a little harsh. Using the JPG engines in cameras doesn't test their full potential, and it is not only the M8 which doesn't measure up here. Taking one step into RAW is already an improvement. Â I don't know what you mean with me never having had to deal with the problem in the first place. If you mean that I am not a magazine tester, sure, you are right. Both with my 5D and my M8 I quickly realised that if I wanted good results, I would have to deal with RAW. I don't know why this should be any different for a tester. Since you have access to them, why not ask them, instead of defending them blindly. I am sure there are problems with using ACR as a standard, for example, but it is already better than using JPGs from the camera. Link to post Share on other sites More sharing options...
Ozkar Posted February 19, 2008 Share #15  Posted February 19, 2008 Are you commenting on my reply? It seems a little harsh. Using the JPG engines in cameras doesn't test their full potential, and it is not only the M8 which doesn't measure up here. Taking one step into RAW is already an improvement. I don't know what you mean with me never having had to deal with the problem in the first place. If you mean that I am not a magazine tester, sure, you are right. Both with my 5D and my M8 I quickly realised that if I wanted good results, I would have to deal with RAW. I don't know why this should be any different for a tester. Since you have access to them, why not ask them, instead of defending them blindly. I am sure there are problems with using ACR as a standard, for example, but it is already better than using JPGs from the camera.  Very true! I recently compared my D300/50f1.4 with my M8/50f2.0. When shot in RAW and processed using Lightroom, the M8 is every bit as sharp as the D300 at base ISO. They're so close that you would struggle to pick one from the other. Link to post Share on other sites More sharing options...
Ozkar Posted February 19, 2008 Share #16 Â Posted February 19, 2008 It does surprise me that the M8 placed 4th at ISO 160. I wonder what went wrong there. It must have been a really close fight, but still, I would have expected the AA filters of the other cameras to have penalised them more. The E-3 apparently sucked across the board, in comparion, which also surprises me. I have seen D300 and E-3 shots compared, and the D300 shots are way softer, so they must have really pushed the sharpening. I wonder which camera shows more detail when you take M8 and E-3 shots unsharpened and then D300 shots and sharpen them. Maybe there might even be some of the 'false' detail which Simon claims for the Foveon. Â I'm not sure how you compared the D300/E3, but I used an E-3/14-54 for about a week alongside my D300/18-70 and found little difference between the two in terms of detail (when shot in RAW and processed accordingly). The D3 and D300 apply little sharpening by default, but sharpen nicely in PP. Link to post Share on other sites More sharing options...
lct Posted February 19, 2008 Author Share #17 Â Posted February 19, 2008 I'm not sure how you compared the D300/E3, but I used an E-3/14-54 for about a week alongside my D300/18-70 and found little difference between the two in terms of detail (when shot in RAW and processed accordingly)... Hardly significant Ozkar with all respect. I own the Nikkor 18-70 as well, it is a great little zoom lens at the price but it is not in par with any good prime IMHO, even a 100$ one like the Nikkor 50/1.8 which is way sharper. Link to post Share on other sites More sharing options...
pascal_meheut Posted February 19, 2008 Share #18 Â Posted February 19, 2008 I don't know why this should be any different for a tester. Â I know you don't know. That's the reason for my reply. But I gave at least one good reason: because when 95% of your readers use jpeg, you might want to test using this format. Â If you are editing a general public magazine about everyday cars, you'll probably write about the handling on open roads. And not talk that much of the absolute performance on track. Â This is the same here. Â Another detail: when editing a photo review, one need to publish tests as soon as possible for obvious economic reasons. What do you do if ACR does not support the new camera at the time of your testing ? Do you switch to the manufacturer software and break your procedure ? Â As I said, when you are dealing with a few hundred of thousands of readers, 30 to 50% of them are occasional and you need to satisfy them to pay the bills at the end of the months, the constraints are not the same as when you post a message on a forum. Link to post Share on other sites More sharing options...
Ozkar Posted February 19, 2008 Share #19 Â Posted February 19, 2008 Hardly significant Ozkar with all respect. I own the Nikkor 18-70 as well, it is a great little zoom lens at the price but it is not in par with any good prime IMHO, even a 100$ one like the Nikkor 50/1.8 which is way sharper. Â I don't understand the context of your reply. I don't recall saying the 18-70 is on par with a prime. I have many primes. I used the 18-70 on the D300 in my comparison with the E3 because I felt it would be on par with the Zuiko 14-54. What's the point of using my 50/1.4 on the D300 while using a zoom on the E3? Â You may also see my later post where I compare my M8 and D300 and in both cases I use 50mm primes. Link to post Share on other sites More sharing options...
lct Posted February 20, 2008 Author Share #20 Â Posted February 20, 2008 Using modest lenses for comparisons boils down to compare lenses because they are inferior to the bodies. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.