Focus Testing - 5D, Mark III, Focus Checker Chart

Well, I've had my Mark III for a few weeks, and in the field I've been feeling it isn't as sharp as my 5D. Was thinking it was me just now knowing the AF system.

So today I did some focus tests with Tim's focus test chart with my main two lenses, a 24-70mm and a 70-200mm. It seems the Mark III isn't as spot on as the 5D with the exact same lens.

My question is should I correct these in camera with the lens adjustment? Or should I send it in for service? Did I do the test wrong maybe? Thoughts?

Full Smugmug Gallery with exif info, in critique view (with their loupe thing accessible)
http://photography.sed​ura.com/gallery/366032​1#209050237

5D, 24-70mm:

Ah - BTW these are all single shot mode focus, with only the center point being used.

They both look good to me. However, it looks like you shot at too shallow an angle in relation to the chart. As I recall, you should be at a 45 degree angle to the chart. You would probably get better results with both cameras at that angle.

mebailey wrote in post #4134901
They both look good to me. However, it looks like you shot at too shallow an angle in relation to the chart. As I recall, you should be at a 45 degree angle to the chart. You would probably get better results with both cameras at that angle.

Certainly willing to retest and/or admit I suck at the test, however I'm pretty sure it was close to 45 degrees. However, regardless if they are at the same angle shouldn't they be pretty close to the same?

I think my results on the 70-200mm aren't very solid, however the 24-70mm I've tried over and over and can't get it to be as "spot on" as the 5D.

JasonSTL739 wrote in post #4134942
Certainly willing to retest and/or admit I suck at the test, however I'm pretty sure it was close to 45 degrees. However, regardless if they are at the same angle shouldn't they be pretty close to the same?

I think my results on the 70-200mm aren't very solid, however the 24-70mm I've tried over and over and can't get it to be as "spot on" as the 5D.

Not saying that you suck at all. To me both target areas are in focus and they are close to the same. There is perhaps a small front focusing bias on the mkIII 24-70 if this is a consistant finding in your testing.
I used this chart alot with my 50L/5D and never could get as good a results as you got with either of your bodies. I suspect if you sent the mkIII in they would say it was within specs.

mebailey wrote in post #4134982
Not saying that you suck at all. To me both target areas are in focus and they are close to the same. There is perhaps a small front focusing bias on the mkIII 24-70 if this is a consistant finding in your testing.
I used this chart alot with my 50L/5D and never could get as good a results as you got with either of your bodies. I suspect if you sent the mkIII in they would say it was within specs.

No worries.

Retested, and this time measured the angle to try and be 100%. Images added to same gallery linked above for original samples as well as below.

I know this is slight and maybe I'm picky, but dang does it seem to make a difference with faster lenses. The lens doesn't seem to matter, on the Mark III they all are front focusing a little. Back to main question - is this exactly why the lens correction feature exists? Or should I just send it in for service?

5D:

I would try the lens correction feature. When the problem is this mild you are likely to get the camera back worse after service, imo.

mebailey wrote in post #4135049
I would try the lens correction feature. When the problem is this mild you are likely to get the camera back worse after service, imo.

Perhaps you're correct. Here are the results of a focus test with a full +18 of backwards lens correction. Seems like alot however, as the scale of correction only goes to 20. Plus that feels significant out of the box. Enough to where I'm going to test every friggin' lens I have with this thing, and don't think I should have to.

Mark III 24-70mm w/18 of backwards focus correction:

Remember the rules!

Depth of field should show focus 1/3 in front of and 2/3 behind the actual point of focus!

While I'm at it... Post +18 of focus correction on the cam with some other lenses. Appears to fix it - and also explains why the 70-200mm was sucking so badly on my camera. At 200mm, it was WAY off as the DoF is pretty narrow as it is.

Mark III, 70-200mm @ 200mm 2.8; +18 of backwards correction:

Go Go wrote in post #4135145
Remember the rules!

Depth of field should show focus 1/3 in front of and 2/3 behind the actual point of focus!

In that case the original images are way off, correct? Lots more in front. That was my understanding also - a slight amount to the back would be OK; but front like this and that's bad.

BTW here is a non-perfect real-world before and after the correction with the lens adjustment. Took about 10 shots of the end of my boom with the camera with and 10 without the adjustment. All are like these two compared...

Now I know why my shots have been sucking ass compared to the 5D's.

Hi Jason,

All of my lenses calibrated to single digit correction. Just lucky I guess?

The micro adjustment is just that a fine tuning devise for the camera and I personally would send a lens in for calibration if it needed anything close to maximum adjustment to achieve correct focus. But that is just me. YMMV.

The 1/3 in front and 2/3 behind rule needs to be adhered to when adjusting the micro tool.

Cool use of the smugmug site BTW!

Go Go wrote in post #4135217
Hi Jason,

All of my lenses calibrated to single digit correction. Just lucky I guess?

The micro adjustment is just that a fine tuning devise for the camera and I personally would send a lens in for calibration if it needed anything close to maximum adjustment to achieve correct focus. But that is just me. YMMV.

The 1/3 in front and 2/3 behind rule needs to be adhered to when adjusting the micro tool.

Cool use of the smugmug site BTW!

So you also have an overall correction in use on your Mark III? How much forward/back?

I've got it set at +18 backward for all lenses and they are right near 1/2 and 1/2 with the focus chart. Only goes to 20... so if 1/3 and 2/3 is to be it would still be not up to par even at max correction.

Thanks RE: smugmug!

Hey,

I calibrated all lenses individually and suggest you do the same.
Then I do some real world shooting to test the correction and go back and fine tune the correction. It takes some time but it is worth it.

Go Go wrote in post #4135145
Remember the rules!

Depth of field should show focus 1/3 in front of and 2/3 behind the actual point of focus!

Up close it is more like 50%-50%.

http://www.dofmaster.c​om/dofjs.html