[LittleSkink]

Picked up a 50-300 over the weekend and seem to have a curious issue

at 300mm with matrix or centre weighted metering it under exposes, at 55mm not so much

But spot meter seems fine - so a scene requiring 1/15 gets that with spot meter but gets 1/100 if i switch to other metering modes

---------- Post added 03-22-2021 at 08:49 AM ----------

done some more quick experiments with grey walls and clear skies etc, spot meter is consistently better and consistently 2 to 3 stops more exposure for any given situation, compared to matrix and c/w (which mostly match)

so I am wondering what the camera is up to, all my other lenses matrix meter is fine (with the usual Pentax bias to the left of the histogram)

---

Last edited by LittleSkink; 03-22-2021 at 02:40 AM. Reason: edit title

[Ole]

Which lens variant do you have, the PLM version or the non-PLM version?
The PLM version is not fully compatible with the K10D.

[LittleSkink]

fwiw its a plain old HD, not PLM HD

have tried to replicate behaviour with other lenses, with a uniform scene (sky, wall etc) there is no exposure difference calculated by the camera when swapping between spot, c/w and matrix - only with the 55-300

also, in case folks think it might be important, the K10d has a Katzeye split screen (which hasnt been an issue before)

[Serkevan]

fwiw its a plain old HD, not PLM HD

have tried to replicate behaviour with other lenses, with a uniform scene (sky, wall etc) there is no exposure difference calculated by the camera when swapping between spot, c/w and matrix - only with the 55-300

also, in case folks think it might be important, the K10d has a Katzeye split screen (which hasnt been an issue before)

Have you checked with other similarly slow lenses maybe? Since you mention it's worse at longer lengths, the split prism might be blacking out part of the image and throwing off the metering system.

[LittleSkink]

tend to use short primes, my slowest lens is 20-40 f2.8-4, not seen this issue before - the 55-300 was a bit of an impulse buy

Am starting to suspect slow aperture kit zoom and katzeye dont play nice

[AfterPentax Mark II]

I think that Stevebrot warned for this: You cannot use spot-metering with the splitscreen.

[DonV]

I think that Stevebrot warned for this: You cannot use spot-metering with the splitscreen.

Littleskink seems to have the opposite situation- correct exposure with spot metering.

[AfterPentax Mark II]

Littleskink seems to have the opposite situation- correct exposure with spot metering.

This is what SteveBrot wrote here: Aftermarket focusing screen - PentaxForums.com about it.

[Dartmoor Dave]

K10D under exposing? Huge difffence between spot and c/w or matrix metering

Is the correct aperture showing in the exif for the shots at 300mm? If the lens wasn't communicating the variable maximum aperture correctly, that would cause progressive underexposure at longer lengths -- with the camera thinking it's metering for f/4 at 300mm rather than f/5.8.

In such a case, the darkening of the split prism as you zoom to 300mm might entirely coincidentally give you enough overexposure with spot metering to make it appear correct, while matrix and centre weighted would be underexposed because of the metering working off f/4 rather than f/5.8.

I know what I'm describing isn't the easiest thing to get your head around, but it's worth checking if the lens is communicating the variable aperture to the camera correctly, even if only to eliminate it as the possible cause. As DonV has noted, if the problem was only with the split prism screen then you should get the opposite behaviour of what you're experiencing.

[j0n4hpk]

Here is where exposure compensation and even auto bracketing can be put to good use. No all of us can get a Katz eye focusing screen. So, it's good to try both techniques before hand and select the best one for the situation.

[stevebrot]

also, in case folks think it might be important, the K10d has a Katzeye split screen (which hasnt been an issue before)

Katzeye + spot metering = potential metering issues when the maximum aperture starts to drop with resulting overexposure. Maximum aperture for your lens at 300mm is only f/5.8. I will have to look up their reference again, but I think there is your smoking gun.

Edit: Here is what the archived katzeyeoptics page for the K10D screen says (LINK):

Spot Metering Mode: Our testing showed spot metering mode to be within factory tolerance for all AF lenses having a maximum aperture of f2.0 or larger (lower f number). Lenses having a maximum aperture smaller than f2.0 (lager f number) will show a significant shift toward increased exposure, with the shift increasing for lenses with smaller maximum apertures. Spot metering mode is NOT RECOMMENDED when using a split prism focusing screen with lenses slower than f2.0. However, if spot mode must be used, it is strongly advised to first take a test shot, check the histogram, and adjust exposure compensation as required.

Please note that KatzEye testing was performed on the K10D and while it is expected that the K20D results will be similar, K20D users are strongly advised to confirm metering results with a test photo prior to beginning critical work.

Steve

[Dartmoor Dave]

Katzeye + spot metering = potential metering issues when the maximum aperture starts to drop with resulting overexposure. Maximum aperture for your lens at 300mm is only f/5.8. I will have to look up their reference again, but I think there is your smoking gun.

But the OP is reporting correct exposure with spot metering and underexposure with matrix and centre weighted, not overexposure. While, as you rightly state, the darkening of the split prism with slower lenses should tend to cause overexposure only. I just can't see how the split prism alone can be causing the symptoms described.

Please note: I'm not saying it isn't the cause of his problem, I'm just saying that it's very strange for it to be causing exactly the opposite effect of what it should be doing.

[stevebrot]

But the OP is reporting correct exposure with spot metering and underexposure with matrix and centre weighted

With a blank wall target all three should agree.

But spot meter seems fine - so a scene requiring 1/15 gets that with spot meter but gets 1/100 if i switch to other metering modes

I read this as spot metering calling for more light.


Steve

[stevebrot]

But spot meter seems fine

May I ask how you are using the spot metering feature? It is best used to manually place exposure and is not very useful for general photography.


Steve

[Dartmoor Dave]

Hopefully the OP will return to the thread and bring some clarity to the actual problem they are having.


Here they say that centre weighted and matrix are underexposing, which I take to mean that they are underexposing relative to known good metering such as another reliable lens on the same camera.

at 300mm with matrix or centre weighted metering it under exposes, at 55mm not so much

But here they say merely that spot metering is resulting in more exposure relative to centre weighted and matrix, without any reference to any known good meter.

done some more quick experiments with grey walls and clear skies etc, spot meter is consistently better and consistently 2 to 3 stops more exposure for any given situation, compared to matrix and c/w (which mostly match)

Perhaps the OP could post some sample shots illustrating the problem. I'm taking "spot metering is consistently better" to mean that the major problem is the underexposure he's getting with centre weighted and matrix, when aiming at a neutral target that shoud produce the same metering in all three modes.

The K10D shouldn't have any problem metering with an f/5.8 lens -- after all, it meters completely accurately with the 18-55mm f/3.5-5.6 kit lens. I suppose it's possible that the main ground glass area of the Katzeye screen is so optimised for a bright viewfinder image with fast lenses that it causes two or three stops of underexposure when used with an f/5.8 widest aperture. In which case the overexposure caused by spot metering with the split prism would compensate for that general underexposure and entirely coincidentally end up with approximately the correct exposure. But surely someone else would have reported behaviour like that by now?


Edit:
I've just remembered that in a later post the OP makes it very clear that the actual problem is underexposure with centre weighted and matrix metering relative to a known good comparison, not overexposure with spot metering:
". . . have tried to replicate behaviour with other lenses, with a uniform scene (sky, wall etc) there is no exposure difference calculated by the camera when swapping between spot, c/w and matrix - only with the 55-300."