KelvinCJ From United Kingdom, joined Mar 2012, 34 posts, RR: 0 Posted (1 year 4 months 1 week 5 days 7 hours ago) and read 4552 times:
I have thought that my L lens seems to suffer from some sort of front/back focus problem, and believe my theory may be correct.
I have attempted to test the focus by running a MFA test, though am not sure if I have done so correctly and hoped to post my test results here in case any of you kind people may have other recommendations, advice or tips.
I shot at 400mm, using f/6.3, iso of 500 and for all but the last (+10) image, 1/400 shutter speed.
The object was approximately 11m away from the camera (not quite the 50x focal length rule), and the area of focus was the barcode.
It's a small investment, compared to your camera + lenses, and it will make calibration much easier (the scale on the right makes it a bit clear whether the lens is front focusing, back focusing, etc.).
Nikon: we don't want more pixels, we want better pixels.
You print the target and then let the software do what it's programmed to do
Keep in mind that the focus shift changes as the focusing distance changes and as the focal length changes(in case of zoom lenses). This can obviously lead to the point where you have positive correction on one side and negative on the other. It's up to you decide which correction to use.
Moreover for very long lenses this process can be really difficult, due to the focusing distance (that can be as much as 50-100 m )
ckw From UK - England, joined Aug 2010, 730 posts, RR: 16
Reply 3, posted (1 year 4 months 1 week 5 days 7 hours ago) and read 4531 times:
You don't say, but was your camera on a solid tripod? 1/400th at 400mm is not going to ensure that micro-shake won't effect the results. You should also have IS turned off.
I'm also guessing you mean the 100-400. Unless you have a very recent body, then calibrating zooms is a bit tricky as you are (I think) allowed only one calibration per lens, and of course performance can vary through-out the zoom range.
A good target is a rule set at a 45 degree angle. If you focus (using spot mode if you have it) on the center of the ruler, then you can easily see how much your lens is front or back focusing.
Calibrating requires careful set up of the target and camera (so neither can move) and then a systematic process of shots at say 100, 200 and 400mm. You may find you have to compromise at one focal length to get best results at another. And of course remember.
Finally, I cannot stress enough how important the set up for the testing is. You must eliminate all possible "other causes" otherwise you are highly likely to misset your calibration.
KelvinCJ From United Kingdom, joined Mar 2012, 34 posts, RR: 0
Reply 4, posted (1 year 4 months 1 week 5 days 5 hours ago) and read 4500 times:
Thank you all for the replies.
Tony - regarding the calibration equipment - I realise this is perhaps the best way to go - and agree that it is a small investment for (hopefully) worthwhile results, but I merely resent the fact that you are effectively paying for mounted paper and plastic - is there no way this can be done without spending a relative large sum of money on such a simple object?
- I have not used any calibration software / objects before - hence my reasoning for this topic, personally, is it worthwhile? At the moment I only have my (as Colin correctly guessed) 100-400mm L lens, so am not sure if it is a worthwhile investment when I have only one lens to utilise at this point?
ilpavone2004, thank you for the link - I was looking at that earlier - though do not think I can obtain it unfortunately as I do not have a Paypal account.
My apologies Colin - I was indeed using a tripod - but do share your view that it is perhaps not the most determinant of tests under such conditions.
I have read / heard of people using the Canon software - with their camera connected to the computer, and viewing the Live View image at 200x magnification - and adjusting it that way?
Though, I have often wondered / thought that the focussing method was slightly different between live view and merely using a viewfinder and may therefore produce slightly different results?
I apologise for all my questions and bothersome nature, though I do greatly appreciate all of your advice and input.
SNATH From United States of America, joined Mar 2004, 3238 posts, RR: 22
Reply 5, posted (1 year 4 months 1 week 5 days 3 hours ago) and read 4486 times:
Quoting KelvinCJ (Reply 4): Tony - regarding the calibration equipment - I realise this is perhaps the best way to go - and agree that it is a small investment for (hopefully) worthwhile results, but I merely resent the fact that you are effectively paying for mounted paper and plastic - is there no way this can be done without spending a relative large sum of money on such a simple object?
Here are some instructions on how to make a DIY version of it: