10-02-2011, 08:06 PM
[quote name='emarm' timestamp='1317550407' post='12019']
Wow, I've obtained a 100% Off Topic Replies (not taking into account Klaus reply on the sanity check)!!!
[/quote]
I don't see the replies as 100% OT.
[quote name='emarm' timestamp='1317550407' post='12019']
Anyway, it seems nobody knows how testing is performed...
[/quote]
There are two more valuable replies by Klaus. 1st: testing lenses as we do is overkill for your purposes. And even if you tried to you wouldn't be able to exactly match our results until we disclosed all details and parameters of our workflow (which for obvious reasons we won't do).
2nd: if you still insist to use a similar testing approach as we do, there's the lens test FAQ which explains some details of the procedure. To summarize: The chart we use is a slanted edge chart with quadrants only in the places we measure (not an ISO or USAF chart). Lighting is not static but depends on the focal length used. You need to make sure the chart is evenly lit and shows no reflections. We use daylight steady lights (no flashes) combined with reflectors to fill from additional angles. The measurement requires careful manual focus bracketing (don't rely on AF for lens testing) and a fixed software workflow (to get comparable results). Our MTF measurements are done with RAW files converted to JPGs which are then fed into Imatest.
Your biggest challenge will not be to get to know the workflow and software (especially Imatest), but to align the camera to the test chart. You will also need a very solid tripod and tripod head. Nonetheless use MLU and let the camera rest for a while after the mirror moved up. Weak camera support will render your MTF attempts useless right from the start.
-- Markus
Wow, I've obtained a 100% Off Topic Replies (not taking into account Klaus reply on the sanity check)!!!
[/quote]
I don't see the replies as 100% OT.
[quote name='emarm' timestamp='1317550407' post='12019']
Anyway, it seems nobody knows how testing is performed...
[/quote]
There are two more valuable replies by Klaus. 1st: testing lenses as we do is overkill for your purposes. And even if you tried to you wouldn't be able to exactly match our results until we disclosed all details and parameters of our workflow (which for obvious reasons we won't do).
2nd: if you still insist to use a similar testing approach as we do, there's the lens test FAQ which explains some details of the procedure. To summarize: The chart we use is a slanted edge chart with quadrants only in the places we measure (not an ISO or USAF chart). Lighting is not static but depends on the focal length used. You need to make sure the chart is evenly lit and shows no reflections. We use daylight steady lights (no flashes) combined with reflectors to fill from additional angles. The measurement requires careful manual focus bracketing (don't rely on AF for lens testing) and a fixed software workflow (to get comparable results). Our MTF measurements are done with RAW files converted to JPGs which are then fed into Imatest.
Your biggest challenge will not be to get to know the workflow and software (especially Imatest), but to align the camera to the test chart. You will also need a very solid tripod and tripod head. Nonetheless use MLU and let the camera rest for a while after the mirror moved up. Weak camera support will render your MTF attempts useless right from the start.
-- Markus
Editor
opticallimits.com
opticallimits.com