K10D Stop down metering exposure problem

I can't argue with that. Use what works best. But at the moment the jury is still out on whether it's some K10Ds that have the alleged problem, or certain users.

I'm prepared to believe that there may be a problem with some cameras. My own experience, however, with the two pre-A lenses that I use often, is that the camera works fine.

G

No pre-judging involved! You whinged, and even confessed to it. ("Sorry. Do I sound like I'm ranting? "

But I am sorry to hear about your problem, and from what you say, you don't appear to be doing anything silly, despite my suggestion to that effect.

However, I do urge you to try a controlled experiment with an A or later lens of the same focal length, and with different combinations of metering method. It would then be interesting to see other users try the same combinations with their cameras.

G

Hey, in the interests of scientific accuracy, I was ranting, not whinging.

If only I were...

Well I have tried all the available metering modes.

But I don't see the point of the experiment you suggest with different lenses of the same focal length - as you say, lenses have their differences. Please tell me what I'm missing here..

My point is that for a given lens (I have only two M series lenses, but I've done this with a couple of As in non-A mode too), ordinary reciprocity does not seem to hold. Stopping the lens down a stop at a time - and then pressing the green button for it to take a peek while stopped down - does not seem to result in the camera picking anything like the exposure that you'd expect.

And specifically with the A lenses, it is possible to compare the exposure the camera picks (in manual mode and stopped down) in non-A mode, with the exposure it (still in Manual) picks in A mode, with the same aperture selected on the camera. In A mode, everything is fine and what you'd expect in the way of reciprocity.

If you really want incident light readings, go get an Expodisc. Works a treat... but then the metering + RAW is really good enough

Matt

I think we might find that different camera bodies give different results, showing that this is not an across-the-board design fault, but a manufacturing defect in some samples.

G

I performed some careful tests with a K 45-125. The results were very inconsistent once you start stopping down the lens.

I also tried with an M 50/1.4,with similar results.

If it's particular to some camera bodies, then it's a defect and the cameras should be exchanged.

yes, but if we knew that some K10D's exhibit this fault and some don't then we COULD exchange them

I did some tests today to try to see if there was any recogiseable pattern to the exposure problems people are having with using older lenses on the K10D. It was a beautiful morning with a clear blue sky and sunshine so the light was consistent during the time taken to do the tests.
Camera settings :- (in Custom) Using aperture ring = permitted; Green button in TAV & M = TV Shift. Exposure Mode = M, Manual Focus.
Lens used Tokina 50-200mm set to 50mm and ISO 200. Cable Release used with camera on tripod. I focussed on the house across the road I then did a series of exposures from f3.5 down to f16 in half stops, pressing the green button before each shot. These are the recorded exposures at each setting :- F3.5 =1/4000, F4 =1/2000, F5.6 = 1/750, F6.3 = 1/500, F8 = 1/250, f8+ = 1/180, F11 = 1/125, f11+ = 1/90, f16 = 1/60.
At F3.5 slight under exposure, F4 about right then getting progressively more and more over exposed down to F16. It would appear to be getting the recorded exposure value about correct but the actual exposure wrong. What is going on.? I don't know if people think my method was sensible enough to give any conclusive results, if not can anyone suggest a better method.? I would like to try the experiment again with a fixed focal length prime lens but unfortunately I do no possess one. What do people think.

As per my post on another question, has anyone tried this after upgrading to firmware 1.11?

I'm planning to repeat my tests over the weekend with it; I'll report back.

I was just thinking the same thing. Ben seems to think it's altered the metering but I don't think he's tried it with non-A lenses

Yes, that's the first thing I did after installing the new firmware. Better, but still off. Very disappointing!!

I wouldn't mind so much if it would at least meter correctly wide open so you can guestimate the rest for yourself.

My K10D is accurate at f4, any focal length, so if I want to use the aperture ring I need to dial from there.

I also want to inform you guys that Pentax Sweden has updated the K and M lens info on their lens compability page with "metering errors may accour" (for those who read Swedish);

http://www.pentax.se/index.asp?url=http://www.pentax.se/default.asp?cat_id=416

So it seems like they are not updated/aware that this is NOT related to just K and M lenses, but has to do with the fact that as soon you are using the aperture ring, the camera makes the error. Doesn´t matter if you use the aperture ring on an A or FA lens. Or stop down a M42 lens.

// Mats

Stopping down the M42 lens manually could be different, because as soon as you do that whatever aperture is set is effectively the wide open aperture of the lens. The camera cannot know differently as there is no connection or stop down mechanism at all.

I haven't got a screw thread lens handy to try it, but if someone has perhaps they could check it out?