Oh how this debate confuses me
OK, I am not going to try and enter the debate with knowledge or insight, however, I do have one contribution to make (that might add to the discussion).
Some of my highly regarded FF lenses (e.g. 135mm - F, K, SMC v2) exhibit some serious PFing when used wide open through to f5.6ish. I have read many say that this is because the lenses are not optimised to digital, but I have not entirely bought this argument.
These lenses do not have APO elements. This is probably the main reason for seeing PF in many of the resultant APS-C images. However, I have a sneaky thought that the 'magnification' of the cropped sensor is also a contributing factor. The APS-C camera takes the 'cropped' image then enlarges it by 1.5X to give me an FF equivalent image. In so doing, it enhances the flaws of the FF lens.
If the above logic is true, then a FF sensor should lessen the PF/ aberrations otherwise observed when using legacy FF glass on a APS-C camera. Although, I acknowledge that this rule would only hold true if the FF sensor has the same sensor density as the comparable APS-C sensor (i.e. K-3 = 24MP sensor ...... K?FF = 36MP sensor).
Perhaps this logic also applies to the corner softness debate a well.
Anyway, I am hanging out for FF no doubt. My APS-C cameras will remain in my back too, especially for wildlife shooting and macro.
---------- Post added 04-04-15 at 11:39 AM ----------
Originally posted by osv since the picture is cropped, the light would have to be falling outside of the crop sensor.
i think that you can redirect the lost light back towards the center with a speedbooster type of rig, but the cheap ones make the photo look like it was shot through the bottom of a coke bottle.
Your comment has made me wonder if the use of FF lenses on APS-C cameras is delivering 'extra' light in close proximity to the sensor thus causing IQ losses .... contrast, dynamic range???