[quote name='Class A' timestamp='1290949984' post='4517']
Can you please elaborate?
Do you not agree that capture sharpening is adequate and can restore detail to the extent the sensor is capable of recording?
The K-5's AA filter is not too strong, i.e., it doesn't destroy information that the sensor would have been able to record without it. This holds true for colour photography. Known B&W patterns are a different issue. I invited a discussion about this difference a couple of times, but so far without any success.
To summarise, don't you agree that regarding colour photography it is inadequate to call the K-5's AA filter "strong" or to speak of an "AA filter issue"?
To what extent the lack of an AA filter can help improve lens test and what the dangers might be in doing so, are interesting questions which, I think, should be explored.
[/quote]
I do not agree that a way-above-average capture sharpening should be applied to images used to measure lens performance. Frankly it is even debatable whether sharpening should be disabled for this purpose altogether (but then it has an impact on the scaling only). I believe that Imatest splits the different color channels for the slanted edge analysis so color artifacts are meaningless here (it works just fine with the M9 as mentioned) . It needs to do so to separate lateral CAs from the MTF analysis anyway (IMHO). That's all that counts for my application. As for the rest I don't overly care personally. Cameras come and go.
Whether post-sharpening is a valid approach for "field images" - again, that's a matter of taste. If you want to do it to recover details that's Ok for me and for the rest of the world. However, for my personal photography I don't require a strong AA filter. As mentioned diffraction serves as a natural AA filter from f/8 (APS-C, 16mp) and I'm mostly into ... handheld ... landscape/travel photography and I'm primarily using zoom lenses which are, for most of the image field, not sharp enough to cause serious problems. Frankly that's an average Joe's situation and not a science project. An AA filter will ultimately soften all images including those where diffraction has eliminated all relevant moiree problems anyway. Post sharpening will increase noise and produce sharpening artifacts. If the K5 would have the K7 noise characteristic aggressive post sharpening wouldn't really be an option at ISO 800, don't you think ? From my perspective this is not about one side or the other - it is about balancing all the negative effects based on your personal preferences.
And to answer your primary question - as of now I don't see the slightest indication that the K5 has a COMPARATIVELY weak AA filter (or RAW NR). You may debate whether "it's just right in conjunction with post-sharpening" (BTW, Pentax does not apply it to JPEGs - why ... ?) but I have to recognize that most other system cameras simply have a COMPARATIVELY weaker AA filter. Frankly, the discussion with Falk just reconfirmed my viewpoint here.
Can you please elaborate?
Do you not agree that capture sharpening is adequate and can restore detail to the extent the sensor is capable of recording?
The K-5's AA filter is not too strong, i.e., it doesn't destroy information that the sensor would have been able to record without it. This holds true for colour photography. Known B&W patterns are a different issue. I invited a discussion about this difference a couple of times, but so far without any success.
To summarise, don't you agree that regarding colour photography it is inadequate to call the K-5's AA filter "strong" or to speak of an "AA filter issue"?
To what extent the lack of an AA filter can help improve lens test and what the dangers might be in doing so, are interesting questions which, I think, should be explored.
[/quote]
I do not agree that a way-above-average capture sharpening should be applied to images used to measure lens performance. Frankly it is even debatable whether sharpening should be disabled for this purpose altogether (but then it has an impact on the scaling only). I believe that Imatest splits the different color channels for the slanted edge analysis so color artifacts are meaningless here (it works just fine with the M9 as mentioned) . It needs to do so to separate lateral CAs from the MTF analysis anyway (IMHO). That's all that counts for my application. As for the rest I don't overly care personally. Cameras come and go.
Whether post-sharpening is a valid approach for "field images" - again, that's a matter of taste. If you want to do it to recover details that's Ok for me and for the rest of the world. However, for my personal photography I don't require a strong AA filter. As mentioned diffraction serves as a natural AA filter from f/8 (APS-C, 16mp) and I'm mostly into ... handheld ... landscape/travel photography and I'm primarily using zoom lenses which are, for most of the image field, not sharp enough to cause serious problems. Frankly that's an average Joe's situation and not a science project. An AA filter will ultimately soften all images including those where diffraction has eliminated all relevant moiree problems anyway. Post sharpening will increase noise and produce sharpening artifacts. If the K5 would have the K7 noise characteristic aggressive post sharpening wouldn't really be an option at ISO 800, don't you think ? From my perspective this is not about one side or the other - it is about balancing all the negative effects based on your personal preferences.
And to answer your primary question - as of now I don't see the slightest indication that the K5 has a COMPARATIVELY weak AA filter (or RAW NR). You may debate whether "it's just right in conjunction with post-sharpening" (BTW, Pentax does not apply it to JPEGs - why ... ?) but I have to recognize that most other system cameras simply have a COMPARATIVELY weaker AA filter. Frankly, the discussion with Falk just reconfirmed my viewpoint here.