[stevebrot]

Too many variables IMO with TAv and before trying to troubleshoot the CPU, might be best to start with M.

I agree, though TAv is M as far as shutter and aperture are concerned.* The original post was explicit in stating that exposures were longer than that set by the user, particularly in TAv mode.

The worst is TAv mode, in which it insists on extremely slow shutter speeds, as in several seconds, regardless of what I specify.

Steve

* In fact, with the K-30 and certain other bodies (K-50, K-500, and probably others) things get a little more interesting in that M mode with auto-ISO enabled defaults over to TAv. When troubleshooting exposure problems using M mode on those bodies it is good to remind the user to switch to fixed ISO to make sure that the meter and exposure automation system are not in the loop.

---

Last edited by stevebrot; 02-27-2017 at 02:34 PM.

[E-man]

Okay, here are some sample photos. The first, blown-out one was taken a few days ago under bright, sunny conditions. The rest were taken today under overcast conditions:


This first one is egregiously blown out in TAv. Interesting that the EXIF information shows it as manual. Exposure time was recorded as 2.5 seconds.


This one was taken in manual mode this afternoon in overcast conditions


This one was in TAv mode. The camera chose to drop the ISO way down.


This one was in Program mode. The camera chose a wider f-stop (5.6) than I would have.


This one was in "green" mode. Again, the camera chose a darker exposure value than I would have.

[stevebrot]

This first one is egregiously blown out in TAv. Interesting that the EXIF information shows it as manual. Exposure time was recorded as 2.5 seconds.

I managed to access the originals at photobucket (HERE). According to the out-of-camera EXIF:

2.5s and f/16 @ ISO 100
Metering mode = "Spot"
EC = +0.7
Mode = TAv

The set exposure was EV100 6.7 while your description of conditions would be about EV100 15.* The result was about 8 stops overexposure.

Conclusion: I can't say for sure, but I suspect that the spot meter reading was taken from a bright part of one of the flowers and that the camera lowered the ISO to its basement setting. The metered EV, being still out of the camera's range of available settings would have resulted in the viewfinder and rear LCD blinking (indicates error condition). Under this situation, the shutter button would still work, but the results will reflect the user settings at ISO 100 rather than the metered EV.

Is there a problem with the camera? It is not possible to completely rule out some sort of controller failure since you recall setting a longer shutter speed. I would suggest repeating using multi-segment metering (rather than spot) and taking careful note of the actual settings displayed in the viewfinder for comparison to the values recorded by the camera on review. I don't know if it was doing so for this photo, but if the display is flashing, be prepared for poor exposure.

This one was taken in manual mode this afternoon in overcast conditions

This one looks fine. The exposure settings (1/60s, f/8 @ ISO 800) are appropriate for the scene. Again, spot metering was used and the point metered was probably some of the foliage (an appropriate target).

This one was in TAv mode. The camera chose to drop the ISO way down.

Metering (spot again) was on the white flower petal. Given the meter method and target, the exposure (1/160s, f/8 @ ISO 160) is correct, though not appropriate. Again, I would suggest other than spot metering.

This one was in Program mode. The camera chose a wider f-stop (5.6) than I would have.

That happens...the program line is what it is, though with Pentax, you are able to override using the front and rear e-dials.

This one was in "green" mode. Again, the camera chose a darker exposure value than I would have.

Again, this sort of thing happens. The camera is doing its best to avoid "clipping" (overexposing) the white blossoms.

At this point, my opinion is that there is probably not a problem with the camera. The washed out photo above is what I would expect for what was recorded in the EXIF. That being said, I suggest being aware of shutter settings when applied manually and chimp the settings on review to make sure they are the same until you are confident that the camera is following your directions.

My advice is to avoid spot metering. It is a specialized tool and requires a fairly thorough knowledge of what the meter reads and how to place exposure to the part of the frame metered. In both properly exposed images above, the metered area fortuitously fell on the green foliage. That corresponds nicely to the 18% gray value that the meter is calibrated to and results in an appropriate exposure.


Steve

* Exposure Value (EV) is derived from the combination of shutter speed and aperture, qualified by the set ISO. It is the basis for the so-called "sunny 16 rule" and may be used as a way to describe scene luminance. EV100 6.7 would be a typical home interior while EV100 15 would be a bright sunny day. LINK

[E-man]

Steve, thanks for taking the time to review those photos and the accompanying EXIF information. I appreciate your feedback and it certainly makes sense. I will change the metering settings accordingly. I'll let you know how that affects performance.

[stevebrot]

So, in TAv mode you might set for f/5.6 and 1/125s and you would get several second exposure? The same should also happen in M mode. That would indicate a problem with the shutter or its controller.


Steve

Steve, thanks for taking the time to review those photos and the accompanying EXIF information. I appreciate your feedback and it certainly makes sense. I will change the metering settings accordingly. I'll let you know how that affects performance.

The ability to do a forensic evaluation is why there were several requests for examples. I was lucky enough to find the images with EXIF intact on your photobucket account. Often, but not always, there are sufficient clues in the image, the EXIF, and in the photographer's account to find a probable cause of poor performance. I think this may have been one of those cases.

I anticipate that a few others may have some additional insight now that we have some images.


Steve

[SpecialK]

I can't say for sure, but I suspect that the spot meter...

I agree with the conclusion. If you are skilled enough to use spot metering, you don't need to :-)

[E-man]

Taking the camera out of spot metering mode did the trick. Exposures are much more in line with what I expect them to be. I guess that's one of the hazards of buying a second hand camera. You have to remove all or most of the previous user's settings. Thanks for all of your assistance. Now that I'm not as frustrated by this camera's behavior, I can take the time to get more creative with it.