[quote name='Klaus' timestamp='1332751281' post='17052']
However, the results are not really cross comparable.
[/quote]
I know - your workflow is not calibrated for this (not saying it would be easy to do). There will be differences due to micro-lensing, AA-filter, the phase of the moon, ...
However if the sensor largely outresolves the lens, these details should matter less and less. With a not so great lens, as some point we should see that the figures essentially stay when replacing the camera with a higher resolving one. I am wondering whether this is what is happening here.
In the end an MTF50 is a strong signal. What your numbers mean is, that for 2000lph (the boarder value of the 16mm) you are still observing a strong signal.
In general when you retest lenses (typically aftermarket like Tamron, Sigma, ...) on different systems I find the numbers you publish a typically within a 200lpph of each other and very rarely totally different. I take it that there is some physics still shining through. For sure that only holds if the optical train is not severely affected by the sensor properties (AA-filter, micro lenses, ...).
However, the results are not really cross comparable.
[/quote]
I know - your workflow is not calibrated for this (not saying it would be easy to do). There will be differences due to micro-lensing, AA-filter, the phase of the moon, ...
However if the sensor largely outresolves the lens, these details should matter less and less. With a not so great lens, as some point we should see that the figures essentially stay when replacing the camera with a higher resolving one. I am wondering whether this is what is happening here.
In the end an MTF50 is a strong signal. What your numbers mean is, that for 2000lph (the boarder value of the 16mm) you are still observing a strong signal.
In general when you retest lenses (typically aftermarket like Tamron, Sigma, ...) on different systems I find the numbers you publish a typically within a 200lpph of each other and very rarely totally different. I take it that there is some physics still shining through. For sure that only holds if the optical train is not severely affected by the sensor properties (AA-filter, micro lenses, ...).
enjoy