Is this considered an AF problem?

Sometimes I get AF that is right on the money and sometimes it back or front focuses slightly but it's still sharp enough to be considered a keeper.

I did the battery test with 3 batteries. Front battery is 1/4" in front of the center battery. Back battery is 1/4" in back of the center battery. I took 10 shots, 50% were right on the money, 50% back and front focusing very slightly.

However if I increase the separation of the front and back batteries from the center battery from 1/4" to 1/2", there is enough distance to make a great separation of DOF between them, the center battery looks focused 100% of the time.

here is my illustration

Are you using one specific/the center focus point or "all focus points? If you do this test with more than one focus point active you are bound to get varying results.

vk2gwk wrote in post #9045090
Are you using one specific/the center focus point or "all focus points? If you do this test with more than one focus point active you are bound to get varying results.

center AF point. i know setting all of the focus points active is not a valid AF test.

Your own findings have given you the answer. The actual AF area is much larger than the box shown in the viewfinder and with the batteries as close as shown in the first diagram it's very likely that the focus will lock onto the 'wrong' battery.

Well .. are you doing the test for your Tamron? For some reason, I got similar results, however, if I use it to take picture normally, it focus fine. So, I don't know if I should trust the testing anymore. For more real life test, I take a close up shot of someone's eye, because it is easier to show if it is sharp with the eye lashes, and you need to know the Tamron is softer wide open at 50mm.

Back to the test. I am thinking, one of the reason can be if the batteries are farther apart, the back one is more out of focus, so, it is more obvious the center one is "more" focus, and relatively sharper. I think you pictures is the top view, right?

gofer wrote in post #9045216
Your own findings have given you the answer. The actual AF area is much larger than the box shown in the viewfinder and with the batteries as close as shown in the first diagram it's very likely that the focus will lock onto the 'wrong' battery.

gte357s wrote in post #9045291
Well .. are you doing the test for your Tamron? For some reason, I got similar results, however, if I use it to take picture normally, it focus fine. So, I don't know if I should trust the testing anymore. For more real life test, I take a close up shot of someone's eye, because it is easier to show if it is sharp with the eye lashes, and you need to know the Tamron is softer wide open at 50mm.

Back to the test. I am thinking, one of the reason can be if the batteries are farther apart, the back one is more out of focus, so, it is more obvious the center one is "more" focus, and relatively sharper. I think you pictures is the top view, right?

I separated the batteries far apart so the AF point won't lock in to the adjacent batteries. My problem is the distance between forward and backward. Maybe I'm asking too much because there is a variance in the AF so the the 1/4" forward and backward shows this where as the 1/2" doesn't show it?

AF is not mm perfect. With a lens slower than f/2.8, or using a focus point that is not an f/2.8 high precision point, the AF is only designed to land you somewhere within the DOF, which could be towards the front edge of the DOF region, or the back edge, or sometimes bang on target, or anywhere in between.

With a lens of f/2.8 or faster, and using a high precision focus point, the accuracy should improve to within 1/3 DOF, but that still leaves room for the exact focus to end up a little in front of or behind the intended subject.

Either way, being somewhere within the DOF zone, your intended subject should still appear sharp, but that does not mean that something else just a little closer or further away won't be even sharper. This is why Chuck Westfall, in his article on performaing AF microadjustment, recommends refocusing at least three times for each microadjustment value, just to try to imrove the statistical accuracy of the measurements. Scroll down this page a bit for his recommended procedure - http://www.digitaljour​nalist.org/issue0812/t​ech-tips.html.

tdodd wrote in post #9045374
AF is not mm perfect. With a lens slower than f/2.8, or using a focus point that is not an f/2.8 high precision point, the AF is only designed to land you somewhere within the DOF, which could be towards the front edge of the DOF region, or the back edge, or sometimes bang on target, or anywhere in between.

With a lens of f/2.8 or faster, and using a high precision focus point, the accuracy should improve to within 1/3 DOF, but that still leaves room for the exact focus to end up a little in front of or behind the intended subject.

Either way, being somewhere within the DOF zone, your intended subject should still appear sharp, but that does not mean that something else just a little closer or further away won't be even sharper.

I agree that AF is not made perfect, however, for my test, my Canon 50mm F1.4 focus "perfectly" with the same setting, while my Tamron cannot. However, I also concluded that my Tamron focus properly from real life photos. So, I concluded that Canon lens AF somehow works better than 3rd party lens, and I will stick with Canon lens whenever possible. So far, all my Canon lens AF are dead on.

point the cam, lens wide open, right at someone's eye, take picture, look on computer. Stop taking pictures of batteries

garycoleman wrote in post #9045337
I separated the batteries far apart so the AF point won't lock in to the adjacent batteries. My problem is the distance between forward and backward. Maybe I'm asking too much because there is a variance in the AF so the the 1/4" forward and backward shows this where as the 1/2" doesn't show it?

tdodd wrote in post #9045374
AF is not mm perfect. With a lens slower than f/2.8, or using a focus point that is not an f/2.8 high precision point, the AF is only designed to land you somewhere within the DOF, which could be towards the front edge of the DOF region, or the back edge, or sometimes bang on target, or anywhere in between.

With a lens of f/2.8 or faster, and using a high precision focus point, the accuracy should improve to within 1/3 DOF, but that still leaves room for the exact focus to end up a little in front of or behind the intended subject.

Either way, being somewhere within the DOF zone, your intended subject should still appear sharp, but that does not mean that something else just a little closer or further away won't be even sharper. This is why Chuck Westfall, in his article on performaing AF microadjustment, recommends refocusing at least three times for each microadjustment value, just to try to imrove the statistical accuracy of the measurements. Scroll down this page a bit for his recommended procedure - http://www.digitaljour​nalist.org/issue0812/t​ech-tips.html.

I think I may have found the answer in your link?

• Expect some minor variations in focusing accuracy within each set of three test images, even though they were all taken at the same microadjustment setting. This is completely normal, and is due to the tolerances of the camera's AF system.

elader wrote in post #9045438
point the cam, lens wide open, right at someone's eye, take picture, look on computer. Stop taking pictures of batteries

lol that's what brought me to doing the battery test. at f/2.8 sometimes the eye come out tack sharp and sometimes not, even with a fast enough shutter speed

garycoleman wrote in post #9045476
I think I may have found the answer in your link?

• Expect some minor variations in focusing accuracy within each set of three test images, even though they were all taken at the same microadjustment setting. This is completely normal, and is due to the tolerances of the camera's AF system.

Yes. That's exactly the point I was trying to highlight - there are variances from shot to shot and it is chance whether you land bang on target each time, or maybe a little in front or a little behind. That is what I meant by not being mm perfect. The thing is, is it perfect enough to produce acceptably sharp focus at the point on which you focused? If so, job done. If not, you have a problem.

This is why the best target is a flat object, facing the camera directly, and not angled at all. Then, just beside the focus target, you have an angled scale so that you can determine exactly where the sharpest point is each time. Battery tests are all very well, but curved targets are not ideal, and having gaps in the distance front to back between batteries does not easily let you figure precisely where focus ended up each time.

The LensAlign tool is pricey, but a good example of a well thought out AF test target and scale. You could construct something almost as good with a DVD case or a box as your target and a regular ruler as your scale. Far better than using a few batteries.

Here's something I set up for calibrating my 100-400 lens in the garden. The box makes a completely unambiguous target, and the scale is quite separate from the subject to be focused upon. The angle of the ruler can be adjusted to alter its "sensitivity" as a way to judge where the DOF is....

HTTP response: 404 | MIME changed to 'image/png' | Byte size: ZERO

Some lenses vary more in AF position than others. You need to do several shots to look for the trend. It can be quite hard to do this just by eye in some cases.

Try my focusing chart out and compare results to your battery test maybe?

https://photography-on-the.net …&highlight=free​+alignment

I also offer a more permanent solid version of this chart for next to nothing if you are interested. I think those that have purchased it have been happy.

garycoleman wrote in post #9045337
I separated the batteries far apart so the AF point won't lock in to the adjacent batteries. My problem is the distance between forward and backward. Maybe I'm asking too much because there is a variance in the AF so the the 1/4" forward and backward shows this where as the 1/2" doesn't show it?

Leave yourself plenty of room between those batteries (if you didn't already) - the AF sensors on the 7D are likely much wider than the spots indicate, as shown in this old 5D image that I scanned from a copy of Popular Photography magazine a few years ago...