[MrCynical]

Not my best ever thread title, but here goes... I got my K20D yesterday, have recently done the obligatory 'focus test chart', and the results look fine to me. The catch is that I know absolutely nothing about photography and have an eye condition to boot

Anyway, could some more knowledgeable folks take a look at the image and verify that it is actually OK? Cheers

IMGP0037.jpg picture by Carzor_Stelatis - Photobucket

Had to upload it to Photobucket because I know that resizing images can sometimes cause issues that aren't there, or hide ones that do exist.

PS: Apologies for the comparatively shocking lighting, and also the glitch on the right side 10mm text is my printer not the camera :P

[Bart]

If I'm not mistaken, Focus should be about 1/3 before the focus point and 2/3 behind.
Yours seems to be the other way around but at least the 'This text should be perfectly in focus' is.

So, I don't think you will experience any problems in real life photos.

[Marc Sabatella]

If I'm not mistaken, Focus should be about 1/3 before the focus point and 2/3 behind.

My understanding is that this is something of a myth - the actual "correct" distribution of DOF depends on focal length, but except for extremely wide angle lenses is normally much closer to 1/2 and 1/2 than 2/3 and 1/3.

In any case, aside from being pretty badly underexposed, the test chart shot looks fine to me. But I can't find EXIF data, so it's hard to say if you shot it wide open as you should, or if anything else was "wrong".

[MrCynical]

In any case, aside from being pretty badly underexposed, the test chart shot looks fine to me. But I can't find EXIF data, so it's hard to say if you shot it wide open as you should, or if anything else was "wrong".

Haha tell me about it. Fairly sure that I shot it wide open - the EXIF thing was an unintentional Lightroom screwup on my part. I'll have a look at (and post if necessary) the original data in LR to make sure it was wide open. The same can be said for the underexposure - I set everything up and then failed to notice that I was casting a shadow on the test sheet. No idiot-proof testing chart can withstand the efforts of a sufficiently capable idiot :P

[flyer]

I set everything up and then failed to notice that I was casting a shadow on the test sheet. No idiot-proof testing chart can withstand the efforts of a sufficiently capable idiot :P

You can't learn without doing mistakes, but you can't learn doing always the same mistake.

[Squier]

Since this thread is of interest, hope the OP doesnt mind me posting my chart test.

I'm wondering about this 1/2 and 1/2 thing Marc. I dont think i achieved that here. It looks mostly behind. Not even 1/3rd - 2/3rd

Panagor 90mm @ f2.8 + essentials required for the test

[Marc Sabatella]

One question would be what kind of light did you use. The camer will focus a little further in front with tungsten light than daylight. Not sure if fluorescent makes it focus behind daylight or not. Also realize that actual reults may vary from shot to to shot depending on distance to target, where the focus ring was previously, etc.

Anyhow, whatever the "shoulds" are in the matter, I think the real story is that tolerances in an AF are not necessarily as exact as we'd like. If the target is in focus at all, I'd personally leave well enough alone. Especially if the above was with cooler light and the results are more forward with tungsten but still showed the target in focus. That's actually the result most would probably prefer (as opposed to being perfectly centered in daylight but completely off under tungtsen).

[Squier]

yeah i maybe worrying over something minor, but just for the record, it was daylight and very sunny outside, but taken indoors in shade.

Thing is, you may remember giving advice on screens in other threads, which i said i was going to try. Well this shot is with the FocusingScreen split image, and i took previous shots, and thought the BF was still there, so put the shim inside that they supplied and it didnt make any difference. The shot you see above was with the shim in place.

I may as well remove this as its not changed anything

[Marc Sabatella]

The focus screen has nothing whatsoever to do with AF. The AF sensors are on the *bottom* of the camera, nowhere near the focus screen.

Or are you implying the shot above was using MF? If so, I'd consider the possibility that the issue is with your technique, and getting used to the fact that the screen might be showing you different DOF than the actual photo.

[Squier]

yes, completely manual focus, with tripod , remote, 2 sec mirror lock up, 45 degree angle etc. Approx 18" from front lens to focus point

I dont use and never have used AF.

I'm quite willing to believe technique is the issue, but what do you suggest ?

[Marc Sabatella]

Practice using the *sides* of the focus chart. Set it up on an angle as usual, look through the fviewfinde, try to focus on the center line, but then really pay attention to the lines above and below to see how far in front and behind the center line the DOF appears to extend. Actually make a note of what the further and closest line are that appear to be decently in focus in the viewfinder. Then take the picture and compare the results there with what you perceived in the viewfinder. With the stock screen and a sufficiently "fast" lens, the viewfinder will show much more in focus than the picture does. Not sure about your screen and your lens, but I'm guess they won't match exactly, either.

Anyhow, once you kow what you're up against, then you can start working on focusing by not just paying attention to the line you are focusing on, but by looking at the entire focus zone, and trying to place your target line appropriately within that zone.

[OregonJim]

Or, measure the distance from focal plane to target, and set focus using the distance scale on the lens.

[Squier]

Howdy again - Marc , used Chart side this time, plus FD was now 1ft instead of 1.5ft.

As you say, area of in-focus was wider in VF than the photo, but looking at this new chart test, it looks pretty good to me. So other than improve technique, im sure there isnt much left to do.

The Split image screen has made life a bit easier, especially for someone who wears Bi-focals, but i have since found that i get better pictures when i remove my glasses, and change the diopter for viewing / taking snaps without them. I can see all the metering sharply too.

I only have to put my glasses back on for reading the menu which is much smaller text.

[Squier]

Nearly forgot to mention - i used the Debug mode software to adjust for BF and increased by +30

[Marc Sabatella]

Now, that image looks great. But again, the focus adjustment feature has *nothing whatsoever* to do with MF. It *might* (although i don't know that for sure) affect when the focus confirmation hexagon lights up. but assuming you focus by actually looking at the image, the settigns in the focus adjustment have no more affect on focus than the brand of car you drive does.