Calibrating lens to camera .. have you done it and does it really work ?

Started Jan 20, 2013 | Questions thread
HEWCanon
Junior MemberPosts: 37Gear list
Like?
Re: Calibrating lens to camera .. have you done it and does it really work ?
In reply to HEWCanon, Jan 21, 2013

Victor Engel wrote:


Or there could be an operator error of some sort.

None I am afraid

From several resources on the internet I gathered the following, and others may differ;

  1. Calibrate a zoom using the longest focal length (the tele end of the zoom range)
  2. Use the maximum aperture (wide open)
  3. Use 25-50x focal length in mm as testing distance between camera and target
  4. Use tripod, mirror-up and remote shutter release
  5. If you print your own target sheet, do it on inkjet and not laser
  6. Do 3 shots per adjustment
  7. Do +/- adjustments and keep doing this until you narrow down your adjustment
  8. Use JPEG’s or RAW without any adjustments
  9. Use a standard target (DataColor© SpyderLensCal© or Michael Tapes Design© LensAlign©)
  10. Use computer software (Michael Tapes Design© FocusTune© or Reikan© FoCal©) to decide the best adjustment value

Before addressing these issues individually, I will suggest one general guideline that I think trumps most of them. If you have a specific project ahead, calibrate using settings similar to what will be used in the project.

Agree

1. What is the rationale?

This is where the DOF is shallowest and minute AF differences will show best and will make a difference again like number 2.

2. This guideline will fail with certain lenses, most notoriously, the Canon 50mm f/1.4 lens. That is because the plane of best focus varies with aperture. Use the aperture that will be used in the project. If you will be shooting wide open, do so for the test. If you will be stopping down, stop down for the test. The camera will be use the same aperture when autofocusing, so this is actually quite important if you use lenses such as the one just mentioned.

Again this is where an error in AF will be most obvious .. I understand focal shift .. however at smaller apertures the DOF is wider and smaller error won't be obvious

3. Again, use what will be used in the project. This general guide strives to be generic and is appropriate if you don't know what you'll be shooting.

This I would agree with and that's why I asked my question to start with .. because taking shots of buildings 500 meters away showed results so different from home test

5. I'd be interested in the rationale for this one. I can't see it making a bit of difference.

This is what FoCal advise and I read this in a forum somewhere as an explanation: the FoCal instructions recommend that you print their targets on heavyweight matte paper using an inkjet printer rather than a laser printer. Supposedly, just the reflectiveness/shininess of the toner used in laser printers can adversely affect the reliability of the results.

http://www.reikan.co.uk/focalweb/index.php/faqs/

http://photography-on-the.net/forum/showthread.php?t=1213026

6-7. Some suggest racking out before AF-ing, then racking in before AF-ing for separate shots. The camera/lens are doing different things but should have the same result. Others suggest repeated focus presses to ensure a good lock.

8. No such thing. <grin>

True .. but I meant no increased sharpness/presence/contrast etc to make the differences more obvious

9. It's sometimes recommended to use a flat piece of newsprint.

There are many opinions there and the good thing about these is they have both a flat and an angled target .. and you can make that yourself as in this excellent article

http://www.the-digital-picture.com/Photography-Tips/AF-Microadjustment-Tips.aspx

10. I've always thought the camera should have a function to automatically calibrate. It should simply alternate between contrast and phase detection focusing methods in order to calibrate the latter using the former.

Yes very true as Contrast focusing is at the sensor so there is no shift and that's what a Canon learning article suggest .. AF using contrast in LiveView then repeat using phase detection and that would give you an idea which direction is your lens focusing front/back or exact if no movement .. and as you say many people out there want Canon and Nikon to include that in their next cameras

HEWCanon

 HEWCanon's gear list:HEWCanon's gear list
Nikon D600 Nikon AF-S Nikkor 16-35mm f/4G ED VR Nikon AF Nikkor 50mm f/1.8D Tamron SP AF 90mm F/2.8 Di Macro Nikon AF-S Nikkor 24-120mm f/4G ED VR +4 more
Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow