Kr - AF Fine Adjustment

Pentax was the first to bring it to this class but the real high end from canon and Nikon already had this.

The Canon 1D MKIII and Nikon D3 had it in 2007, Pentax K20D is from 2008.

Well it's more of a problem for digital then it was for film because film had some thickness to the sensative layer but the digital is very thin.

Seems we are in a dynamic range war now, i don't complain though and they are keeping the pixel at quite a sain level.

I definitely had a focussing issue on my K-X with DA L 55-300mm lens, so much so that I was on the point of returning the lens as I thought it may be faulty (very soft). Since adjusting the focus correction (hidden menu) to +50 this issue has gone away. Manual focus was never a problem, but to manually focus is sometimes tricky with a long zoom. As per previous post, I wish Pentax would make the focus adjustment menu available by default on the K-X (maybe via a firmware update). Clearly they have done this on the K-R.

Regards
David

P.S. The focussing on my son's K-M although it is only 5 focus points and rather slow, is definitely more accurate than my K-X in spot center focus mode.

I think focus issues, have become more pronounced since digital, especially for a lot of amateurs. In the days of film, you didnt take as many photos, and it took time from the end of the roll to the finished picture. It was accepted that some didnt "turn out". Nowadays, your finished product flashes before your eyes in a matter of seconds.

I had a rangefinder when I was a teenager, and dammed if I could get a sharp focus out of it. My Dad who was a professional photographer, tested it for me and said it didnt look like there were any problems, he just said squeeze dont push that button. You didnt argue with my father. But I still thought even his photos were a little soft.

Funny how disappointing results, put you off photography as a whole. Dont want to be in that situation now where less than satisfactory results, make you less than satisfied with your gear, and the art of photography.

Standard focussing screens can't resolve DOFs narrower than what you'll get at f/2.0

Therefore it's nigh on impossible to accurately manual focus a lens at, for example F/1.4, unless of course you use live view on magnify and take advantage of the accuracy of contrast AF


If the AF is 'out of wack' then surely it's sadder to not correct it if the means to do so is there?

Actually manufacturers have added this feature because due to lack of perfection in manufacture of lenses and bodies there needs to be some adjustment to correct AF errors on individual body lens combinations.

I have 7 lenses I use on a regular basis and have checked the AF with each lens on both my K-5 bodies. Out of those 7 lenses TWO required NO adjustment. The other 5 required adjustments. FWIW the same lenses required different adjustments on both my K1-D and K20D when I had them.

Calibrating the AF for a lens is no more complicated that adjusting the dioptre of your viewfinder. Nothing whatsoever to do with messing with the innards.

It's nothing to do with obsession, it's to do with calibrating the AF on your camera. It's dead easy to do. And dead easy to correct with cameras that allow different adjustments for different lenses.

How thick is this sensitive layer of film and is that ever thicker than the DOF thickness at the focal plane


What has dynamic range got to do with AF accuracy?

Film is build up with 3 layers much like a feavon sensor and the grains have some thickness too, no idea about what kind of thickness you're talking about but it's much thicker then digital.
Don't get me wrong it's also an issue with film but less so.

Just to give you an idea how precise it is, the numbers shown when doing the AF adjustments are in 1/100 of a millimeter.

Nothing John started talking about the pixel and ISO race and i believe now it's all about DR.

Must say I agree with Smeggy here, and disagree with those who say AF adjustment isn't necessary. I recently got a Sigma 30mm f1/4 and from the very first shots I could see that the AF was off. All I did was casually shot a few frames of a magazine that was lying on the table, focussing on a drop cap 'A'. Zooming in on the LCD made it perfectly clear that the lens was consistently back focussing by 1 cm or so at the minimum focus distance. A bit of trial and error resulted in a -3 adjustment, and now it seems perfect. Took all of 5 minutes.

Stefan that's an interesting tidbit about the 1/100mm increments. One of your better ones

Smeggy have you been following that thread on the other forum about cutting down one of those specialised short dof Canon screens that can resolve f1.4? It looks very interesting.

I forgot to say that 1/100 of a millimeter is the difference it makes at the sensor, quite stunning how much impact it can have and how precise everything must be.

I've looked the numbers up when talking about medium format digital backs and arguing that the Pentax 645D can be make more precise then the separate backs, and this one of the reasons.
ALPA make cameras where you can fit a digital back onto and they have shims so you can adjust the focus perfectly.

These AF adjustments are just a bit of care for your kit, some have invested so much into this so it's just silly not to get the lenses working at their best.

Indeed. It's simple. No need for men in white coats with a small and diverse selection of pens in the top pocket

Yes I was wondering what one movement of the K20D/K-5's AF adjustment equated to in terms of actual distance.

The K10D debug mode adjustment is in μm, so 10μm for a user adjustment then. When I was adjusting my K10D's AF in debug mode I had to adjust to around 210μm for one lens. That would equate to +23 on the K20D/K5

I should check it out. I've been currently using live view to focus my 1.2 and 1.4 MF lenses, but that's not always a comfortable

I presuming you need to shim the screen to calibrate it though and I'm not 100% sure how to get that accurate.

I have previously shimmed a focus screen with slices of the sticky part of a post it note around the edges that are underneath the metal frame. After some trial and error I ended up using 3 layers IIRC, seemed to work okay

Here's the link to the Canon ee-S screen thread, btw: link

Apologies for jumping into an old thread (I claim New Boy Bad Form rights), but I'd been having some sharpness issues with my Sigma 70-300 on a K-r, and found this discussion. As a result, I spent an hour playing around with the fine-AF settings @300mm, and noticed a big improvement when setting the AF at about -5 (I haven't tested shorter focal lengths yet, as I've got a Pentax 50-200mm as well, so am not likely to use the shorter FLs much with the '300).

The consequence of this is that I've potentially saved myself £250, as I had a big downer on the lens and assumed I'd have to go for the Pentax 55-300 to get the sharpness I wanted. Now that I'm picking out the eyes of pigeons roosting on trees 40 yards away, I think the Sigma will do me just fine for the foreseeable.