Adventures in microadjustment

I know it's hard to make an OP interesting when the subject has been discussed to death, but here goes:

• First time playing around with microadjustment on my 7D.
  
• Five lenses: Canon 17-55, Canon 50/1.8, 100L macro, Sigma 30mm f/1.4, and an old Pentax 50mm f/1.4 that I got from my dad; probably 1970s-80s vintage.
  
• Did the whole routine - solid tripod, busy image to focus on, lens parallel to image, flood lighting, camera distance was at least 50 times the lens focal length.
  

• First, the Siggy. Good news: focus appears to be spot on. Bad news: about 1 in 7 or 8 AF attempts failed miserably - as in, it wasn't even close. Also weird is that in all cases the focus comfirm said it was focused. 2nd bad news: the backlash is horrible! I never really noticed it before because I rarely use the LCD with this lens, but for this test I was using 10X zoom and the whole image would shift about 7-8% to the side every time I reversed focus (manually). The autofocus seems to deal with the backlash OK, but it's disconcerting nonetheless. None of the other lenses exhibit any backlash btw. So, is this an issue worth sending it back to Sigma to correct? Do Sigmas typically suffer from backlash, or do I just have a bad copy?
  
• Although all the lenses appeared to focus properly, I went ahead and played with changing the MA values, but I couldn't really see any difference, even going from -20 to +20. Is the focus shift resulting from microadjustment typically dramatic?
  
  
• This is the first time I had done a head-to-head comparison of the Pentax vs. the other lenses and I immediately noticed a decided yellow tint to the images, made noticeable by the fact that I was shooting a pure black and white test pattern.
  

Are you sure you are testing with fast AF (phase-detect), not the default Live View AF (contrast-detect)? MA only matters for the phase-detect AF.

Modern day "in control" camera machining and optics assembly should be within 1/10000 and 1/1000 th of an inch respectfully and thats within the need for no MA.
So MA is good for accepting "out of control" manufacturing and "out of range" temperature.
It is also good for reducing manufactures responcibility of quality manufacturing.
I would contact Sigma on backlash issue and accept no inferior camera or lens.

I wouldn't send it that sigma in

pulsar123 wrote in post #14490985
Are you sure you are testing with fast AF (phase-detect), not the default Live View AF (contrast-detect)? MA only matters for the phase-detect AF.

Arrggghh! I followed a set of instructions I found on the internet (www.birdphotography.co​m, I think), but I didn't notice the instruction to use Quick Live View. So I redid the test, and my lenses are still within tolerance - no MA required. I guess I failed Mircroadjustment 101, but I learned something!

rick_reno wrote in post #14491654
I wouldn't send it that sigma in

Is backlash not generally considered a problem worth fixing? I can eventually achieve focus at 10X, but It's pretty annoying.