5D3: AFMA variations between different AF points

Charlie wrote in post #15251305
after reading this thread, maybe people have unrealistic expectations about their gear. Hell, I shoot a 5D2 and outer points the majority of the time, but when it's dark, center point and even manual focus when needed. I figure the 5D3 outer points can only be better than the 5D2 a certain number of EV's, so in a lot of low light situations, it's a no go.

That's a good thing to keep in mind and sound advice on how to cope in low light situations--the center point is still more sensitive than even the cross-type outer points on the 5DIII as well, although now there are 2 AF points above and 2 below it that are as sensitive per the AF literature--but I don't think that's what's in view here. People are seeing this problem in all kinds of light levels, and conversely people without the problem have reported that the outer points work reliably even in low light situations although of course not in as low light as the center point does.

Yohan Pamudji wrote in post #15252521
People are seeing this problem in all kinds of light levels, and conversely people without the problem have reported that the outer points work reliably even in low light situations although of course not in as low light as the center point does.

I think that the points are just not working as advertised. Canon calls the short stacks of cross type side points "high precision" at f/4 AND they are rated to -2 EV. Why should they not work 'reasonably' in low light if they work as advertised?

The problem seems to be more of some AF points being patently miscalibrated rather than unreliable. With my first camera, I could calibrate the off side points because they were reliably and repeatably off from the center point.

That is not at all the same problem as being 'unreliable in low light'.

My second copy of 5D3 is much better behaved, although I have not shot an event yet using my 35L.

Just sayin'...

michelle blossom tree wrote in post #15251020
me too...

I'm unsure what to do in regards to their reply ""Yes, the centre focus point is strongest AF point with guarantee accuracy at f2.8, the other points around it are only guarantee accurate if you set F stop to 5.6"

so is it just tough luck that I cant use the outer points because they are "only guarantee accurate if you set F stop to 5.6" - I dont want to set to 5.6 - i want to get beautiful bokeh! Is there adjusting that can be done? should I send it somewhere?

Michelle, I have been reading Canon's white papers on the 5D3 and 1Dx AF system and I think the tech gave you wrong information.

The side points are high precision at f/4. This does not mean they are ONLY reliable at f/4. It means that if you have a f/4 (or faster) lens mounted, they will function as high precision cross points. If you have a slower lens than that, they are not high precision.

If your side points aren't working, send it into Canon for an adjustment. It absolutely should be covered in warranty.

Without reading the whole thread what's the latest on this, anyone send their gear in?

I'm getting this on my 5d3 and 50 1.2

Just sent in both of my 5D3s yesterday along with the 24-105. We'll see what comes of it.

mco_970 wrote in post #15253040
I think that the points are just not working as advertised. Canon calls the short stacks of cross type side points "high precision" at f/4 AND they are rated to -2 EV. Why should they not work 'reasonably' in low light if they work as advertised?

The problem seems to be more of some AF points being patently miscalibrated rather than unreliable. With my first camera, I could calibrate the off side points because they were reliably and repeatably off from the center point.

That is not at all the same problem as being 'unreliable in low light'.

My second copy of 5D3 is much better behaved, although I have not shot an event yet using my 35L.

Just sayin'...

I agree with you 100%. That's exactly what I was trying to say

Thanks for the info guys and keep us posted!

Hey at least it's not anywhere near as bad as the left side AF issue of the D800.

s2kdriver80 wrote in post #15253269
Just sent in both of my 5D3s yesterday along with the 24-105. We'll see what comes of it.

I am glad that someone is sending gear to Canon to see what they can do. I am sorry you have to send it in, though. I hope that discombobulated mess of thought makes sense.

Thanks again Paul for helping me find the issue with my 5D3 while it could still be exchanged. I am truly interested to see how this situation is handled by Canon. Hopefully they do a great job for you.

Yohan Pamudji wrote in post #15253444
I agree with you 100%. That's exactly what I was trying to say

Yeah, sorry, I was trying to add info to your statement and should have quoted Charlie, too. I pretty much disagree with him about 5D3 being that similar to 5D2 in terms of AF. It's not a simple matter of EV sensitivity improvement.

FYI, thought I would share my AFMA results using 85 f1.8...some side points better than center. Did find that trying to do this too close to target generates spurious results.

EDIT: Pasted picture doesn't show??

Second try...AFMA results with 85 1.8.

Now that's more like it. Could you share some of the testing details? Version of FoCal? Did you print the multipoint target on just 1 sheet? What kind of light? Brightness in EV? Distance from the target? Percentage value you plugged in? The reason the sides did better was probably because the AFMA wasn't optimized for the center point. If you use a value tuned for the center, the numbers would probably be reversed.

s2kdriver80 wrote in post #15254603
Now that's more like it. Could you share some of the testing details? What kind of light? Brightness? Distance from the target? Percentage value you plugged in? The reason the sides did better was probably because the AFMA wasn't optimized for the center point. If you use a value tuned for the center, the numbers would probably be reversed.

Sure...

1000W of halogens (basic Home Depot variety), producing about 10/11 EV from 3 meters or so. Target of 3 pages wide by 2 tall. AF tolerance was 2%. AFMA was tuned for center point at +3, hence my confusion...ran that test yesterday, same lighting but I ran it on the Windows 1.6 version, while today's test was on the OSX 1.7 beta, if that should matter.

Was pleased to see all green but guess I need to revisit the base AFMA calibration...

mco_970 wrote in post #15253055
Michelle, I have been reading Canon's white papers on the 5D3 and 1Dx AF system and I think the tech gave you wrong information.

The side points are high precision at f/4. This does not mean they are ONLY reliable at f/4. It means that if you have a f/4 (or faster) lens mounted, they will function as high precision cross points. If you have a slower lens than that, they are not high precision.

If your side points aren't working, send it into Canon for an adjustment. It absolutely should be covered in warranty.

Thanks for thinking of me and for the info there! when i emailed the warranty place I referred more to the lens (50mm 1.4) having problems and backfocusing when using the outer points as I thought the problem lied with the lens and that was their reply, so maybe the numbers he gave me were for the 50mm 1.4 not the mkiii?? not sure, I'm so confused about this issue!

I bought the pro version of foCal yesterday to try and work it out... now i just have to work out how to use the software properly!! as soon as i do I'll check back in with my results to see what you guys think

Just going to tag along here.