Metering with M-lenses on the K-7

[blende8]

Exposure measurement with M-lenses is a problem on the K10 and K20.
So I tested this on the K-7 yesterday.
Lens was the M28/2.

Result:
The metering is more consistent and sufficient to use.
The images from the K20 were overexposed by more than 2 stops using apertures from about 5.6-16.
But still, it is not perfect on the K-7 either.

Thanks to TeKaDe for the opportunity to test the camera.


The scene:



RGB histograms of the individual images:

[jeffkrol]

Too bad. K-7 does have interchangeable screens, correct?

[Ari]

Wow! You're so lucky to test the k-7! Can I make a prediction? Our friendly forum is going to be inundated with new dslr shooters, realizing the quality, build and backwards compatibility of Pentax makes more sense than the marketing prowess of Canon/Nikon. And I welcome all of them! As I slowly move in to the world of pro-photography (and it is a very slooooooooow move) I take much pride in showing off my k-20 to people that have never heard of the brand and explain the value and benefits of shooting with some of the best glass out there.

[blende8]

Originally Posted by jeffkrol

K-& does have interchangeable screens, correct?

Yes.
But I was told that they are of a different size than those of the K10/K20.

[stevebrot]

Wieland,
Thanks for the K-7 update! I have been wondering about non-A lens metering and am glad to see that their has been some improvement. For many of us, that alone may justify the purchase

Steve

[Ole]

Thanks for taking the time to do the test, and in particular for the extra effort of posting the histograms! They make the differences so much easier to see!

[borno]

Nice job Wieland
Good to know it is a bit better especially if the LL-60 won't fit

[d.bradley]

what metering mode did you use, and/or can you tell us if the results were consistant with different modes?

[blende8]

Originally Posted by d.bradley

what metering mode did you use, and/or can you tell us if the results were consistant with different modes?

M-lenses do not allow Multi, therefore I used center-weighted metering.

[d.bradley]

Originally Posted by blende8

M-lenses do not allow Multi, therefore I used center-weighted metering.

Ah right...my mistake. Thanks for the clarification.

[Marc Sabatella]

Thanks for doing this! I agree - definitely looks better. Acceptable if not great. When I tried the same with my M135/3.5 on my K200D, all exposures were within a quarter stop of each other and looked more or less identical on histogram.

Thanks for letting us know, but your image looks pretty durn soft.

[Marc Sabatella]

Not sure if you were thinking the alleged softness of the image (how can you relly tell at the post size?) is an indictment of the K-7, but according to the EXIF, it was taken with the K20D. And since the test was of exposure, not sharpness, who knows if he even bothered to focus.

[arpaagent]

Wieland,

Any idea about metering with M42 lenses? I know on my K200D, if I have it set to Av mode, then it has funky metering where you need to put in about +2 Ev to compensate (or alternatively short out the lens contacts so it thinks it's a K mount lens...). I am wondering if they will allow for regular, correct Av metering (as per a wide-open K-mount lens) for these M42 lenses? I think that would be really great, rather than relying on some weird Ev compensation.

Thanks,
-Jim

[blende8]

Originally Posted by Marc Sabatella

who knows if he even bothered to focus.

Cannot remember, actually, what I did.
Fortunately for our problem it doesn't matter.

Originally Posted by arpaagent

Any idea about metering with M42 lenses?

No, I didn't test M42 lenses.


What I can imagine for the future is that you can program your individual lens line for all apertures.
Also, it would be nice if Pentax could release some white paper about how they did calibrate the metering.

It's really a pity that Pentax isn't interested in this issue.