Jump to content

Tokina AT-X Pro 28-80/K10D Underexposure?


sleahy73

Recommended Posts

Hi all,

 

I picked up a AT-X Pro 28-80 last December and it was working great. Lately,

though I've been noticing a problem with metering with it on my K10D. It's been

pretty consistently underexposing by ab out 1.5 stops (sometimes more). I've

been using Multi-segmneted metering. It seems to have started about a month

ago. I haven't been shooting anything so tricky to require special metering.

It's shows up in both AV and metered Manual modes (the ones I use the most).

Has anybody else had this kind of experience, or, better still, have any

suggestions for why this might be happening?

 

I don't have an example to post handy at this momnt; I'll try to get one in

later this evening.

 

Thanks,

 

Sean

Link to comment
Share on other sites

I did check the bladdes; they seem fine.

 

I haven't tried stopping it all the ay down, but from f/11, and every point I've tried openning up from there it's been doing it. I suppose I should take it out for a more thorough test, but I'm not sure what to look for jst yet, so all I ould do is keep proving that it is, indeed, underexposing.

 

Any suggestions on how to test it? AV mode through all of the f stops?

 

Thanks,

 

Sean

Link to comment
Share on other sites

I suppose that could be... When I got home, I did some "test" shots on a medium grey shirt through the appertures in Tungsten light. The exposures looked pretty close, if only slightly under. However, on the underexposed example above, the histogram in almost all on the left side, and nothing is anywhere near the right edge. So if the camera meter was trying to hold the highlights, it way overshot...

 

Also, I brought my new 77mm ltd with me and had no problems with the metering.

 

A couple of weeks ago I was in the same location and having the same problems. I thought it had more to do with my cirular polarizer. Next time I get the chance I'll have to go back and try different metering and see if it clears it up.

 

Thanks,

 

Sean

Link to comment
Share on other sites

The camera is fooled by the bright sky, bright sand.

 

Same thing will happen on snow. Where it evaluates the brightness levels of the 16 zones and then figures out a average midtone exposure. The more white in the scene the more underexposed it will all be.

 

Actually on a lot of point and shoots in the old days it was a single "Sand/Snow" mode. Now with 200 vari programs they have a Sand and a Snow mode.

 

What I'm saying, you need to sometimes overide the meter. All those vari program snow/beach modes do is increase exposure by 1-2 stops.

 

Of course I see you are saying the 77mm didn't do that. In that case, it could be the lens. But does it happen everytime whether at the beach or not?

 

Just a note though, I don't see why the camera would try to hold the highlights. I don't think Pentax has scene dedicated algorithms for metrix metering like Nikon does. It might, but I never heard it mentioned. So if it was trying to hold the highlights than it should essentially underexpose every shot you take.

 

all scenes are just percentages of grey for the meter, whether it's a beach or a forest.

Link to comment
Share on other sites

Yeah, I expected to have to push the EV up a little. With the 77mm on the camera, I had to go between +0.3 and +0.7 for an acurate exposure. I did push it as far as +1.3 to "expose to the right". Anything beyond that using the 77mm lens and I blew the highlights. With the Tokina I had to start at +1.3 and that was usually under a bit.

 

On the other hand, when I got home I did some testing of the Tokina through the f/stops on a more or less medium grey shirt in tungsten light (best I could do at the time) and it turned out more or less correct (maybe 0.3 under). So I don't know. I guess I'm going to have to go back and see if I can figure what's bugging the meter.

 

Thanks,

 

Sean

Link to comment
Share on other sites

Page 142 in the K10D manual:

 

"The center-weighted metering mode is automatically set even if you select the mult-segment metering mode when using a lens other than a DA, D FA, FA J, FA, F or A lens"

 

So if you selected multi-segment you got that with the 77/1.8, but got center-weighted with the Tokina.

Link to comment
Share on other sites

Hummm... I was under the impression that it was only PENTAX A-series lenses that permitted multi-segment metering, not any lens with an "A" setting. Of course, that would mean that the K10D can somehow tell when a non-Pentax A lens is mounted! Interestingly, I was told at one point that the K10D uses a "look-up table" in the software to compute the exposure reading... which understandably would only have values for Pentax lenses. The flaw in this idea would be how new Pentax lenses would be incorporated, unless the chip in the lens is the place where this data is stored. I believe that MTF data specific to each lens is stored inside all Pentax autofocus lenses (Post A-Series), which is used by the MTF Priority mode.

 

Having no hard documentation to back any of this up, I can only speculate!!! However, I do have a Tokina AT-X 35-70/2.8, so I can go off and do some expirimentation...

 

Curt

Link to comment
Share on other sites

Curt, you may be right after all! I hadn't thought about the non-Pentax A's behaving differently than the Pentax ones; do let us know if it's true.

 

Hey, I own that very same Tokina, maybe you'd like to share your thoughts about it? Please e-mail me or start another thread so we don't clog this one up with off-topic discussions.

Link to comment
Share on other sites

Glad to hear that all "A" lenses support multi-segment meeting... out of curiosity, how did you confirm that it works? I did a brief check with my Tokina AT-X and didn't see much change in the metered read-out when I switched between center weighted and multi-segment modes, but I didn't do any exhaustive or even well-controlled testing. I did see more of a change when I put on a DA lens... I didn't get so far as to capture a couple of different meter mode images and check their exif data in PhotoMe...
Link to comment
Share on other sites

I took a white sheet of paper and drew a black square in the centre. I then propped it up on the wall and framed a shot such that the black square was covered by the central bracketed square on the focus screen. In spot metering and Av, the camera was calling for 1/8s; in multi-segment metering it read 1/30s. In the first case it was metering off black only, while in the second it was metering off a white sheet with a black square in the middle. I thought there would be more than 2 stops difference in the metering, but in any case, this proves that it doesn't just work with centre-averaged metering.

 

Another method to figure this out is to be Justin, who knows just about everything about all this stuff. But being Justin means you have to be prepared to freeze your arse off in the mountains. I guess there are pros and cons to being Justin :-)

Link to comment
Share on other sites

Hummm again... MM, wouldn't you get similar results if the K10D internally defaulted to center weighted, even though the switch was set to multi-mode? I wouldn't expect it to not meter all all in multi-mode, just switch to center weighted (and not bother to tell you, of course!) if it did not 'like' the mounted lens...
Link to comment
Share on other sites

Matt --

 

I agree that a good EXIF reader will tell you what mode was selected, i.e. the switch position (by the way, I prefer PhotoME to Reveal, for the additional data it displays), but does that say for sure what the microprocessor does with the data?

 

By the way, while poking around in the PhotoME display, I found the neat option to display graphically the LV recorded by each of the metering segments. It looks like all segments are always recorded in the EXIF stream, but they are used in different manners depending on the selected mode... would be neat if this data could be superimposed on the thumbnail, like the selected AF point(S) can be!

 

Curt<div>00ORAw-41748884.jpg.558671ee45c2730e4dc9f10b60fbf6ee.jpg</div>

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...