[dtsui]

I understand not everyone experience this problem in the field but anecdotal evidence clearly points to a sizable user population that has this problem which affects four to five different models. If I were Pentax management, that is enough evidence for concern. In fact I believe the problem was not noticed by an even bigger population of users because Pentaxians have got used to the poor PTTL performance in the earlier generations of models, e.g. istDS, K10D, K20D that they have workaround the problem using various means or they just don't use external flash any more.

If Pentax were aware of the problem (they must be) even though it may affect only a certain percentage of the products, they are still deliberating selling defective products to consumers without any intention of fixing it. The fact that the problem has spread from K7 in the beginning to now five models in a period of four years is testament to this. This is very shady business practice.

[Anvh]

@ David, the digital TTL systems of the other brands also have their share of problems sadly enough. Sadly... it's good to see they are struggling as well showing that the problem is quite complex.

[fs999]

I understand not everyone experience this problem in the field but anecdotal evidence clearly points to a sizable user population that has this problem which affects four to five different models. If I were Pentax management, that is enough evidence for concern. In fact I believe the problem was not noticed by an even bigger population of users because Pentaxians have got used to the poor PTTL performance in the earlier generations of models, e.g. istDS, K10D, K20D that they have workaround the problem using various means or they just don't use external flash any more.

I am not sure about that. I think it is the contrary. I am not using the flash any more (except for macro shots) with my K5, due to the high ISO capability.

On the other hand my K10D and K20D had no flash problems at all.
With the K7 they changed some specifications of the PTTL(they don't communicate about this but I think they removed the old film TTL compatibility), and my Dörr macro ring flash didn't work any more. But my Metz continued to work well.

I will make some test with the internal flash.
If there is a problem, then that would prove that Metz has a workaround with its last firmware update of the 48 AF1...

[Noam]

I am not sure about that. I think it is the contrary. I am not using the flash any more (except for macro shots) with my K5, due to the high ISO capability.

On the other hand my K10D and K20D had no flash problems at all.
With the K7 they changed some specifications of the PTTL(they don't communicate about this but I think they removed the old film TTL compatibility), and my Dörr macro ring flash didn't work any more. But my Metz continued to work well.

I will make some test with the internal flash.
If there is a problem, then that would prove that Metz has a workaround with its last firmware update of the 48 AF1...

Indeed, the evidence gathered until now indicates that the problem began with the K-7. The K10/20 were fine in that regard.

fs999, there is little point testing the internal flash -- because it's fine! The problem, as indicated by me here and by many others is with PTTL overexposure when the flash is bounced - and only when it's bounced. this happens regardless of flash manufacturer - be it Pentax, Sigma (mine) or Metz.

EDIT: Of course I forgot to mention that the problem is only with the flash on-shoe. Wireless PTTL bounce works fine! (thx Quicksand.)

dtsui - I agree this is very disappointing.

---

Last edited by Noam; 03-07-2013 at 04:34 AM.

[Anvh]

With the K7 they changed some specifications of the PTTL(they don't communicate about this but I think they removed the old film TTL compatibility)

I believe this actually in the manual of the camera.
They indeed remove it with digital, only the ist models have TTL.
When using TTL on newer cameras it will simply flash at full power.

[Quicksand]

fs999, there is little point testing the internal flash -- because it's fine! The problem, as indicated by me here and by many others is with PTTL overexposure when the flash is bounced - and only when it's bounced. this happens regardless of flash manufacturer - be it Pentax, Sigma (mine) or Metz.

And moreover -- the overexposure only occurs when the external flash is ON-CAMERA. I have found exposure to be excellent, and very reliable, when bouncing an off-camera flash, using wireless optical P-TTL controlled by the internal flash on my K-5.

To me, this points directly toward a firmware bug.

[Anvh]

And moreover -- the overexposure only occurs when the external flash is ON-CAMERA. I have found exposure to be excellent, and very reliable, when bouncing an off-camera flash, using wireless optical P-TTL controlled by the internal flash on my K-5.

To me, this points directly toward a firmware bug.

I can offer you one theorie.

With the flash straight forward, the camera will most likely expect that there will be a flash highlight/reflection in the photo but with the flash bounce this becomes complicated.

Maybe doing some test with a relfective surface in the photo might be able to support this.

[dtsui]

On the other hand my K10D and K20D had no flash problems at all.
With the K7 they changed some specifications of the PTTL(they don't communicate about this but I think they removed the old film TTL compatibility), and my Dörr macro ring flash didn't work any more. But my Metz continued to work well.

In the beginning my K10D suffered bad underexposure with AF540FGZ but then I discovered a fix on this forum, a very serviceable fix indeed. There is a trimpot (trimmer potentiometer) hidden beneath the FIX label just above the hotshoe of the AF540FGZ. By adjusting this trimpot, the sensitivity towards underexposure can be much reduced to almost non-existent. The presence of the trimpot on the flash is an indicator that Pentax is struggling with matching the flash to the camera. For some unknown reason, perhaps Pentax engineers thought they got smart with a new PTTL algorithm, they decided to ditch the function of the trimpot from K7 forward. I have since confirmed the trimpot has absolutely zero effect on the flash exposure for the K7. This is a bad move for me because the last hope of taming this overexposure problem is lost.

[bjdavey]

For those using a K-5 II or IIs, today's firmware update 1.03 is giving me a major improvement on this long standing P-TTL tilt flash overexposure issue. As this is relevant to just a sub-set of users I started a thread here for others to confirm if they are also seeing improvements with this combo:

https://www.pentaxforums.com/forums/pentax-k-5/217693-k-5-iis-firmware-1-03-i...ure-issue.html

The firmware update for K5 II and IIs users is here:

Latest K-5II/K-5IIs Firmware Update : Software Downloads | PENTAX RICOH IMAGING

Firmware notes state "Improved the accuracy of exposure when using AA batteries", so exposure issues are clearly part of the package. I have not yet done definitive tests, but certainly am seeing a big improvement with my IIs and AF-540FGZ flash.

[SteveB]

There's no change in bounce flash overexposure with the K-30's 1.03 firmware by the way.

[JinDesu]

Any other users noticed changes with their k-5II and the issue in this thread post-latest update? If it's solved, that's fantastic news for k-5II users.

[Noam]

Sorry guys, I checked this morning with K-5II 1.03 and the problem status is unchanged.

bjdavey, could you please elaborate on what exactly has improved, and how you tested it?

Here is my methodology (abridged, please see previous posts for more details). I will appreciate it if you try it:
Use mode X in the camera, with the flash set to P-TTL, no compensation. Take one shot of something at a reasonable distance with the flash pointed straight ahead. Take another shot of the same object, this time with the flash slightly tilted (enough to make the flash think it is tilted, showing the appropriate tilt indicator on the UI). Compare the shots.
With my K-5II and Sigma 610 Super, the second shot is consistently much brighter, typically overexposed, than the first shot, which is usually spot on, i.e., well-exposed.

[bjdavey]

Sorry guys, I checked this morning with K-5II 1.03 and the problem status is unchanged.

bjdavey, could you please elaborate on what exactly has improved, and how you tested it?

Here is my methodology (abridged, please see previous posts for more details). I will appreciate it if you try it:
Use mode X in the camera, with the flash set to P-TTL, no compensation. Take one shot of something at a reasonable distance with the flash pointed straight ahead. Take another shot of the same object, this time with the flash slightly tilted (enough to make the flash think it is tilted, showing the appropriate tilt indicator on the UI). Compare the shots.
With my K-5II and Sigma 610 Super, the second shot is consistently much brighter, typically overexposed, than the first shot, which is usually spot on, i.e., well-exposed.

I haven't done a series of controlled tests yet - what I'm noticing in general use after the 1.03 upgrade is consistent properly exposed pics using tilted bounce flash (AF540FGZ) - something I haven't previously experienced with either my old K5, or new K5IIs. I'm on business travel at the moment, but will run some tests as you suggest on my return.

[polur101]

After being an owner of K200D and K20D for 4 years, upgraded to K5, primarily tempted by the deal and for high iso performance. I did a lot of reading regarding the supposed tilted flash over exposure issue. Evidently PTTL is not as user friendly as Nikon flash system. Cannot vouch for that. I don't have a Nikon system. During earlier days of K20D and K200D I got very frustrated about the inconsistent flash exposures. Once I figured out (at least I thought so), how the camera thinks and interprets the flash scenario, I started getting near 95% accurate exposures. So much so now I can guarantee the result.

The learning curve was not that steep for K5. Because of its ability to handle noise in high iso, I tend to push the iso limits. I routinely shoot indoors with iso of 1600-3200. With K20D I didn't dare go past iso 400, lest I incur unacceptable level of shadow noise. In K5, with such high iso while using flash, I find that I need only a whiff of flash. I have -2 flash exp comp in body, stofen diffuser, bounce 60degrees. Over 1000 shots I have had perfect histograms.

I think it comes down to "knowing" your gear.

This was my logic with K20D and k200D. It all depends on how much of preflash hits the sensor. For example, if you use a zoom lens and focus on a very small area (perhaps a bit dark) only a small amount of light hits the sensor. Camera interprets this as darkness and blasts flash. In such a scenario I dial down exposure by -0.7 to -1. When shooting a group of people whith a light color wall behind, larger amount of light hits the sensor and the reverse holds true. I had a simple mnemonic, "If it is less less (as a verb) it, if it is more more (as a verb) it." This worked nearly 100% of the time. Drawing from that experience, I follow similar principle with k5, except I need lot less flash power, so I work in the range of -2 to -0.3 Exp comp. I find it is easier to adjust EC than FEC. We can program the rear dial to directly change the EC without having to press the EC button. It is easy to adjust EC on the fly. Hope this helps. Cheers.

[jamesm007]

I can offer you one theorie.

With the flash straight forward, the camera will most likely expect that there will be a flash highlight/reflection in the photo but with the flash bounce this becomes complicated.

Maybe doing some test with a relfective surface in the photo might be able to support this.

Flash units such as the AF360 or Promasters 7500EDF worked and still work flawless (almost) on bodies such as the K20D and older. In fact when I use flash I will just use my K20D. The K5 has had and after reading this continues to have a huge bug Pentax does not seem to want to touch.

I can bounce the head of my external flash at 30, 60, 90, or turn it around 180 degrees and the flash will fire a near perfect shot. What the OP was saying is Pentax flash system was not as good as Nikons during the K20D days but it got worse with the K5! Much worse, now you can't even bounce the flash because the results are bad and you have to dial in EV comp. You did not have to with the K10D or K20D those bodies and same flash gave good results. I would call that a major problem Pentax should fix. As an owner of the K5 and K20D I can say my Promaster 7500EDF is very good on the K20D as long as it has the power to do the job. K5 works fine as well until I tilt the head up and try bounce flash with the K5 then it under exposes (versus over). I have to dial in about 2EV to compensate. With the K20D and older bodies there was no problem.

I think that's what your saying. That there are technical reasons for the results. Well there are technical reasons. That Pentax needs to fix because with the same flash units there were no problems on the K20D, K10D, K200...