Soft Pics:(

Before I go home and Make myself insane. How do you all go about testing this.. I took a few shots tonight at work and was not happy with the results..

I Put a piece of paper with lines and letters down, Held camera at angle (Roughly 45*) and put the focus on a particular letter and it was not too clean.. Above and below looked fine.. Now keep in mind this was looking at the LCD.. I will load on pc when I get home and post results...

What camera?

If above *and* below looked fine, then obviously focus isn't front or back.
For a focus test to have any significance, it *has* to be done in controlled circumstances (tripod, MLU, decent light, good test chart)

Very odd if the 'areas above and below' the point of focus were sharp but the actual point of focus was out.

As Rene says try again under more controled conditions and do post the results if you wish for feedback???

Might try using a focus chart as well and repeating the test. Here is a good chart to use.

http://focustestchart.​com/chart.html

Thanks for the info.. i will use the Focus chart for reference..

ItsMike wrote in post #4110638
Thanks for the info.. i will use the Focus chart for reference..

You should also try to use a tripod for a focus test. You may not notice it, but it is possible to sway a bit and move the camera slightly towards or away from the subject in between focus lock and releasing the shutter.

ItsMike,
I also suggest you use a tripod, as this will help remove possible movement errors by user and stick to the cameras AF accuracy.

Bob_A,
We must have been thinking the same thing at the same time

If above and below showed equivalent drop-offs in sharpness, you might start thinking about the sharpness of your lens when wide open.

This test chart is great. I used when I first got my XTI and used again yesterday to test my new 40D

here is the result

Byte size: ZERO | Content warning: NOT AN IMAGE

I didnt use a tripod, I am at work and this is the best I can get for now..
Both shots with 24-105
Exif is intact..
Center focus selected, Put on the circled "P"

1.

So what are you unhappy about?

The only thing I can see is that focus is not on the encircled 'P', but considering the target you are using for your test, the AF might just as well have locked onto the two parallel lines just above. In addition, You are not using tripod so even if AF did lock onto the P, you may have moved forward just a bit (we're talking 0.5 cm or so here) before the shutter opened.

These kinds of pixelpeeper tests must be performed carefully if you want to draw any certain conclusions from them. (and not handheld at 1/30 sec with a 105mm lens as in the first pic)

/lk

This is not a very scientific test. For 1/30 shutter speed @ 105mm, I'd say it looks as good as it possibly can. As lederK said, I would not draw any conclusions from this test--there are too many variables involved that could adversely affect the end result.

Judging from the sample images, I would say the lens or camera is backfocusing. If the lens was focusing spot on, then you would expect an equal size area in front of and behind the P to be in focus. The sample images show that there is very little in front of the circled P that is in focus, and a lot behind it. Looks like something needs to be calibrated.

Just to verify, you did use the center focus point when doing these tests, correct?

BigBlueDodge wrote in post #4126867
Judging from the sample images, I would say the lens or camera is backfocusing. If the lens was focusing spot on, then you would expect an equal size area in front of and behind the P to be in focus. The sample images show that there is very little in front of the circled P that is in focus, and a lot behind it. Looks like something needs to be calibrated.

Just to verify, you did use the center focus point when doing these tests, correct?

Not really.
That’s called Depth of Field distribution and in most cases there is more focused area behind the point of focus, than in front of it.

It could be precisely calculated for each scenario, but roughly, it goes from 2/3 back, 1/3 front with focus at hyperfocal distance, all the way to almost 1:1 for macro shots.