[foxsp]

Well, it’s been almost four years since I first posted my concerns about the firmware V1.07 for the Pentax K-5iiS. There have been no new firmware updates for that camera so Pentax either has not become aware of the issue or do not consider it worth the time and expense to fix.

When I posted the question 4 years ago, the conversation quickly devolved into the relative merits of whether I should or should not use spot metering for a particular image. The issue is not whether I should be using a particular exposure mode for the particular image I want to capture.

The issue is that the behavior of the camera changed between firmware versions V1.06 and V1.07. All I want to know is whether this behavior is just in my camera or did the firmware V1.07 change the behavior in all K-5iiS cameras.

It’s very easy to find out if your camera also exhibits this change in behavior. Do NOT use flash when executing the following steps.

1. Turn the mode dial to the “P” position.
2. Turn the “Metering Mode” switch to the center weighted (center) position.
3. Point the camera at a well lit blank wall and take an image.
4. Turn the “Metering Mode” switch to the “Spot” (dot) mode.
5. Point the camera at the well-lit wall again at the same spot and take another image.
6. Check both images. With V1.06 firmware both images will show the same exposure. With firmware V1.07, the image taken with the “Spot” setting will be exposed about 3–stops brighter (over exposed).

If you have firmware V1.07 in your camera (K-5iiS) and you see the above behavior, please reply to this thread so I can send it on to Pentax Support. If you don’t see this behavior with V1.07 firmware please also let me know.

Please remember, I am interested in the behavior of V1.07 firmware in the Pentax K-5iiS camera only. But if you also observe this behavior in a K-5ii camera body, please also let me know. I also have a K-5 body running V1.16 firmware and it does NOT exhibit this behavior.

Please let me know. A 5 minute test so I know whether my camera is broken or the firmware is broken.

Thanks

[Just1MoreDave]

I get 1.5 stops brighter. It seems to be consistent but I'd like to try again tomorrow under daylight in case of artificial light trickery.

[stevebrot]

6. Check both images. With V1.06 firmware both images will show the same exposure. With firmware V1.07, the image taken with the “Spot” setting will be exposed about 3–stops brighter (over exposed).

I get 1.5 stops brighter.

Same metered EV?


Steve

[kh1234567890]

On my K-5ii with DA16-45 lens and 1.07 firmware I get exactly the same exposure on both shots.

[Just1MoreDave]

It just occurred to me that I have a Katzeye screen installed, so spot metering is affected by that.

[stevebrot]

It just occurred to me that I have a Katzeye screen installed, so spot metering is affected by that.

I wondered if that were the case and also whether the OP has an aftermarket screen as well. KatzEye actually tested the effect on spot metering for their screens on the K-7,* but even so, I almost never use this feature with my KatzEye-equipped camera.

Pentax K-3 K-5, K-5II, K-7, K-30, K-50 & K-500 KatzEye Focusing Screen

(See Additional Information > Product Details)


Steve

* Essentially no problem at apertures wider than f/2.8

[stevebrot]

* Essentially no problem at apertures wider than f/2.8

I just checked my KatzEye-equipped K-3 and found found a 1.0 stop overexposure bias relative to matrix at f/2.8 and a 2 stop bias at f/3.5.


Steve

[foxsp]

"Just1MoreDave " and "stevebrot" I'm glad you mentioned the KatzEye screen. I too have one in this camera. It is the split image one. That could make sense that it could be causing the problem. If Pentax changed the algorithm for spot metering in V1.07, the split image prisms might be interfering with metering. But when using spot metering with flash, the exposure is correct. I kept the original focus screen so when I locate it I'll put it back in and check this to see if it's the problem.

Like I said, the camera behavior changed between V1.06 and V1.07 firmware.

Thanks everyone.

[c.perkins]

Tested K5IIs with v1.07 = about 2 stops difference
Tested K5II with v1.07 = about 2 stops difference
Tested K7 with v1.13 = about 2 stops difference

I always shoot in "P" mode and thought this was the nature of the beast since all my bodies do it.
Never tried with older versions cause the first thing I do is upgrade to newest so do not know if there was a change from old to new versions, sorry.

Now if you go to "Auto/Green" mode the exposure is the same.
For what it's worth.

Clarence

[stevebrot]

"Just1MoreDave " and "stevebrot" I'm glad you mentioned the KatzEye screen. I too have one in this camera. It is the split image one. That could make sense that it could be causing the problem. If Pentax changed the algorithm for spot metering in V1.07, the split image prisms might be interfering with metering. But when using spot metering with flash, the exposure is correct. I kept the original focus screen so when I locate it I'll put it back in and check this to see if it's the problem.

Like I said, the camera behavior changed between V1.06 and V1.07 firmware.

Thanks everyone.

It is hard to say regarding the firmware version, though I do know that KatzEye has consistently offered a strong caution regarding spot metering since at least 2008 when I first purchased from them. The interference with spotmetering is well known and is one of the reasons (historically) why spot metering has been conspicuously absent from film SLRs having center region focus aides for about as far back as TTL metering was offered.* If there was a K-5IIs firmware version that was capable of detecting and compensating for the issue with a 3rd-party screen, that would be fairly extraordinary and noteworthy. Ditto for P-TTL flash metering with the split-image. FWIW, my K-3 overexposes badly at f/3.5 when spot-metered with the KatzEye. Set aperture on your K-5IIs to f/3.5 with spot meter and see how P-TTL goes.

BTW...what lens are you using for this thread and the example images on the other thread (LINK to post from April 2014).


Steve

* My Mamiya/Sekor 1000 DTL offers dual spot/averaging metering, but the spot meter reads at bottom center of the frame rather that the center microprism.

[Just1MoreDave]

In daylight, I can get three stops overexposure. I tested with the 18-55 because it was on the camera, f5.6. The Katzeye site suggests 2.5 stops for f5.6, pretty close. Spot metering hurts my brain anyway.

[4morse]

Well, it’s been almost four years since I first posted my concerns about the firmware V1.07 for the Pentax K-5iiS. There have been no new firmware updates for that camera so Pentax either has not become aware of the issue or do not consider it worth the time and expense to fix.

When I posted the question 4 years ago, the conversation quickly devolved into the relative merits of whether I should or should not use spot metering for a particular image. The issue is not whether I should be using a particular exposure mode for the particular image I want to capture.

The issue is that the behavior of the camera changed between firmware versions V1.06 and V1.07. All I want to know is whether this behavior is just in my camera or did the firmware V1.07 change the behavior in all K-5iiS cameras
It’s very easy to find out if your camera also exhibits this change in behavior. Do NOT use flash when executing the following steps.

1. Turn the mode dial to the “P” position.
2. Turn the “Metering Mode” switch to the center weighted (center) position.
3. Point the camera at a well lit blank wall and take an image.
4. Turn the “Metering Mode” switch to the “Spot” (dot) mode.
5. Point the camera at the well-lit wall again at the same spot and take another image.
6. Check both images. With V1.06 firmware both images will show the same exposure. With firmware V1.07, the image taken with the “Spot” setting will be exposed about 3–stops brighter (over exposed).

If you have firmware V1.07 in your camera (K-5iiS) and you see the above behavior, please reply to this thread so I can send it on to Pentax Support. If you don’t see this behavior with V1.07 firmware please also let me know.

Please remember, I am interested in the behavior of V1.07 firmware in the Pentax K-5iiS camera only. But if you also observe this behavior in a K-5ii camera body, please also let me know. I also have a K-5 body running V1.16 firmware and it does NOT exhibit this behavior.

Please let me know. A 5 minute test so I know whether my camera is broken or the firmware is broken.

Thanks

Ran your experiment with v1.07, both photos show the same exposure.

[Not a Number]

Would the program line setting have any effect?

[Digitalis]

I only get a 1/3rd stop difference when switching from CWA to spot on my K5IIs with V1.07. Nothing to get tied up in knots about. Program mode is set to MTF. I also have Link AE to AF, from what I understand that only works for Matrix metering modes.

As @stevebrot mentioned: there is a caveat when using AE metering modes with third party focus screens, the microprism and split prism will foul up spot metering by at least two stops on average - possibly more depending upon the cameras metering proclivities. I don't use any third party focusing screens in my cameras because I use spot metering....a lot.

Would the program line setting have any effect?

possibly.