12-03-2011, 05:04 PM
Hi Klaus, Markus,
A while back, Klaus, you alluded to the discussions you were having about geeting rid of the numbers on the lens resolution scales. The rationale, if I've remembered and understood it right, was that the numbers change from one (sensor) system to another and that they therefore introduce confusion when people insist - despite your admonitions not to - on comparing resukts across systems. Simple Poor to Excellent resolution scales with no numbers would remove this confusion and would mean that comparison of lens resolution across systems would be slightly more valid (?).
I was wondering whether the 'cutoff' points (Poor - Good - Excellent, etc) was based on a % of the max LW/PH resolution obtainable from each sensor. I also wondered whether these cutoff points are derived from Imatest or are PZs own, based on the sensor resolution (though I know the max LW/PH is different for Pentax K-5 and Nikon D7000 though they use the same sensor (a suggested explanaation for that was differences between the AA filters).
I've been looking at the scales across the different Nikon and Canon systems:
[ATTACHMENT NOT FOUND]
On this basis there is some consistency in the cutoff points:
I wondered if there's a reason for this and whether it would be having any influence on the star ratings on D7000? These generally seem to be lower, although I realise that this is being attributed, at least in part, to the higher pixel density being more challenging to lens resolution.
I was also going to point out that most of the cutoff points for the Canon tests are consistently lower, but that might start arguments here, so I won't..!
<img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='
' /><img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/tongue.gif' class='bbc_emoticon' alt='
' />
A while back, Klaus, you alluded to the discussions you were having about geeting rid of the numbers on the lens resolution scales. The rationale, if I've remembered and understood it right, was that the numbers change from one (sensor) system to another and that they therefore introduce confusion when people insist - despite your admonitions not to - on comparing resukts across systems. Simple Poor to Excellent resolution scales with no numbers would remove this confusion and would mean that comparison of lens resolution across systems would be slightly more valid (?).
I was wondering whether the 'cutoff' points (Poor - Good - Excellent, etc) was based on a % of the max LW/PH resolution obtainable from each sensor. I also wondered whether these cutoff points are derived from Imatest or are PZs own, based on the sensor resolution (though I know the max LW/PH is different for Pentax K-5 and Nikon D7000 though they use the same sensor (a suggested explanaation for that was differences between the AA filters).
I've been looking at the scales across the different Nikon and Canon systems:
[ATTACHMENT NOT FOUND]
On this basis there is some consistency in the cutoff points:
- Poor starts ~ 30=33% to 40-45% of max. resolution (anything less would be "Hopeless" - or other chosen adjective, I guess!)
- Fair starts ~ 45%
- Good ~ 55%
- V.Good ~ 70%
- Excellent ~ 80-85% and above
I wondered if there's a reason for this and whether it would be having any influence on the star ratings on D7000? These generally seem to be lower, although I realise that this is being attributed, at least in part, to the higher pixel density being more challenging to lens resolution.
I was also going to point out that most of the cutoff points for the Canon tests are consistently lower, but that might start arguments here, so I won't..!
<img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='
' /><img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/tongue.gif' class='bbc_emoticon' alt='' />