likwidplastik wrote:
FWIW, I'm having similar issues with the S35 on my 5Dmkiii...although not nearly as pronounced.
With the lensalign/focustune, I was getting +2 at 25x focal length, and 0 at 50x focal lenth.

Doesn't sound like much, but it's noticeable.

Interesting. I wasn't aware that LensAlign also has a software solution. I'm having problems with FoCal. Maybe I'll give the Focustune a shot since I already have the LensAlign.

Ok I grabbed this from something on DPR about the sigma USB lens dock

"I'm especially excited about the USB Dock that will allow the new lenses to not only be "re-chipped" to work with new cameras, but will allow a lens owner to customize the lens to optimize such things as focus speed, accuracy, the focus limits, and the front or back focus or of every lens at every zoom length and focus distance. Wow!


Sounds like it should sort out the problem

badlydrawnboy-
I feel your pain. One thing to keep in mind that I discovered the hard way is that Focus can be affected by the type of light too in addition to distance (which you have already discussed). On my 7D I have lens with big Front Focus if used with Incandescent lights, and big back focus if used in Flourescent lights.

ALso, in your ver first 24-70 vII post, I thought it was slightly back focused. Have a look, and try some MFA with it.

Lastly, at least on my 7D (not sure about your 5dmkIII, the center point is extra and more sensitive AF than surrounding points at f2.8 and lower. Thus, do try to use consistently your center point AF, even Spot AF in your testing for accuracy (placement and sensitivity).

Best of luck.

Guy

By the way the lens align software is NOT automated . You shoot the images at each given MA setting. You then run the images thru the software and it gives you a graph to show the best MA

Ian.Dobinson wrote:
By the way the lens align software is NOT automated . You shoot the images at each given MA setting. You then run the images thru the software and it gives you a graph to show the best MA

Yeah, I'm watching the video now. Seems like more work, but I'm getting some strange errors with FoCal during testing that I've been unable to resolve (working with their support). Might be that the lenses I'm trying to calibrate are so far off, but FocusTune might be worth a try.

badlydrawnboy wrote:
Ian.Dobinson wrote:
By the way the lens align software is NOT automated . You shoot the images at each given MA setting. You then run the images thru the software and it gives you a graph to show the best MA

Yeah, I'm watching the video now. Seems like more work, but I'm getting some strange errors with FoCal during testing that I've been unable to resolve (working with their support). Might be that the lenses I'm trying to calibrate are so far off, but FocusTune might be worth a try.

Yeah, it's pretty time consuming, but it offers some piece of mind knowing that the software is giving you the same (or close) results as to what your eye is seeing with just using the ruler on the lensalign.

It's pretty cheap too, and he offers $10 off or so if you own a lensalign, just shoot him a PM.

If you end up getting it, let us know what your results are. I'd like to know how they compare to what I've been getting.

likwidplastik wrote:
badlydrawnboy wrote:
Ian.Dobinson wrote:
By the way the lens align software is NOT automated . You shoot the images at each given MA setting. You then run the images thru the software and it gives you a graph to show the best MA

Yeah, I'm watching the video now. Seems like more work, but I'm getting some strange errors with FoCal during testing that I've been unable to resolve (working with their support). Might be that the lenses I'm trying to calibrate are so far off, but FocusTune might be worth a try.

Yeah, it's pretty time consuming, but it offers some piece of mind knowing that the software is giving you the same (or close) results as to what your eye is seeing with just using the ruler on the lensalign.

It's pretty cheap too, and he offers $10 off or so if you own a lensalign, just shoot him a PM.

If you end up getting it, let us know what your results are. I'd like to know how they compare to what I've been getting.

Thanks. I saw that coupon and did buy it for $20. I plan to do some testing tomorrow. Will definitely post results.

Ian.Dobinson wrote:
Ok I grabbed this from something on DPR about the sigma USB lens dock

"I'm especially excited about the USB Dock that will allow the new lenses to not only be "re-chipped" to work with new cameras, but will allow a lens owner to customize the lens to optimize such things as focus speed, accuracy, the focus limits, and the front or back focus or of every lens at every zoom length and focus distance. Wow!


Sounds like it should sort out the problem

Either that, or drive me absolutely nuts trying to figure out all of the permutations.

Well, now I'm even more confused. I just did some testing with FocusTune. Since I already knew the lens was front-focusing, I did an initial test at +5, +10, +15 and +20. It said +15 was the best result with this chart:



So then I did a bunch of shots from +13 through +19. It told me +17 was the best result, with this chart:



I looked at the actual files, though, and +17 was inconsistent and didn't look right. So I ran the test again. This time it returned +14, with +17 as the worst result!



Then I looked at the +14 files. They're all over the map:









I don't know whether this is a camera problem, a lens problem, or both.

What do you think?

Okay, wait, I think I take it back!

I just did some shooting at +14 AFMA, and it is much, much better than it was before.

For example:





I still may send the 5D3 to Canon, just for peace of mind.

What about the Sigma 35? Is a +14 within the realm of reasonable AFMA? In other words, if you bought a lens that required this significant an MA, would you keep it?

Whoa, that's sharp! Are you getting those results a close and far distances? Mine is noticeably sharper at close distances.

If I liked how the lens behaved when focus was good, sure, I'd keep it if it needed an adjustment of 14; why not?

These are both close distance (which is what I tend to work from). I'll try some at longer working distances later.

One of the issues I've had with this lens is that the AF varies depending on focal distance. So if it's adjusted at 25x focal length, that doesn't mean it will focus accurately at 50x. That's my next test.

Op is the lens not focusing? Or is it focusing somewhere else? On your original 3, #2&3 are perhaps prone to focus error. Light and dark (skin&Burgundy jacket) line. Af tends to grab lines.

Are you using multiple af points?

Just calibrated my new Sigma 85 and it is returning a +15 AFMA.

Makes me wonder...

I answered a thread on here, the guy put a lens box in front of white blinds, I think it was dark out, so there were black and white lines on the blinds several feet behind. The af grabbed the blinds, of course, and the guy sent the lens back because he mistakenly thought it was front focusing! Af loves contrast.

badlydrawnboy wrote:
Ernie Aubert wrote:
With the lens that you've now sent back, did you try ignoring what Focal said and attempt the manual, trial-and-error method of accomplishing MA? Someone else reported (in this thread? I don't remember) bogus behavior from Focal. I'd be interested to know how you fare without employing Focal...

Yes, I dialed a +20 in and that helped. But, the front-focus was too extreme to be corrected with even +20, and second, it didn't seem to front-focus much at all at longer distances, so a +20 actually made it back-focus in those situations. That's probably why FoCal was returning a -1. I should have tried calibrating with FoCal at 3 feet to see if it picked up on the front-focus, but since I already knew I was returning it I figured there wan't much point.

This describes my Sigma 35 1.4 as well, to a tee. +20 isn't enough, and I'm having a hassle getting it swapped out from the seller.

I just got mine. It front-focuses on my 5DM3 but a +10 AF adjustment made it perfect at all focusing distances - FYI this is the first lens I've had to dial in any adjustment on my 5DM3. On my 5Dc it focuses just about perfect, which works out well since the 5Dc doesn't have focus adjust. I wonder if Sigma is tuning the lens this way on purpose.

Interesting. I sent my 5D3 to Canon today. It's certainly possible there's nothing wrong with it, but since it's still under warranty I want to be absolutely sure.

My Sigma 35 required a +14 AFMA and my Sigma 85/1.4 required a +15 AFMA. I didn't have a chance to shoot enough with either of them before I sent the 5D3 in to Canon today to verify whether focus is accurate with those adjustments, but it did seem so in the little bit of shooting I did with each.

My Canon lenses (50/1.4, 40/2.8 and 24-70 II) are all no more than 2 points of AFMA in either direction. The 24-70 is zero at the wide end and +1 at the tele; the 50/1.4 is -2 and the 40/2.8 is zero.

badlydrawnboy wrote:
My Canon lenses (50/1.4, 40/2.8 and 24-70 II) are all no more than 2 points of AFMA in either direction. The 24-70 is zero at the wide end and +1 at the tele; the 50/1.4 is -2 and the 40/2.8 is zero.

Your 5D3 is fine; Otherwise, all yours (50/1.4, 40/2.8, 24-70 II and 5d3) are out of calibration.