K-7 firmware bug that affects leftmost AF point?

Started Jul 19, 2009 | Discussions
Pete Fang Senior Member • Posts: 1,900
K-7 firmware bug that affects leftmost AF point?

I just read from a Japanese forum that some K-7 users in Japan found an issue with the leftmost (non-cross type) AF point under certain conditions. It goes like this:

  • Turn on automatic level correction

  • Use a shutter speed faster than 1/180s

  • Some say high ISO

I just tested with level correction at 1/250s and ISO400. Indeed when using the leftmost AF point the K-7 will just hunt and will not lock focus at all. Switching to rightmost AF point, no problem. Turn off level correction and use leftmost AF point, no problem either.

Looks like a firmware bug to me but would appreciate it if other K-7 owners test it and see if you can reproduce this as well. I personally almost never use the leftnost or rightmost AF points so this does not bother me that much.

Cheers!

Peter
--
Peter Fang - Pentax user for more than 25 years: K-7 / MZ-S / Z-1 / SFX / LX
My PBase Gallery: http://www.pbase.com/pfang

 Pete Fang's gear list:Pete Fang's gear list
Pentax K-5 Fujifilm X-Pro1 Fujifilm X-T1 Fujifilm X-Pro2 Pentax K-1 +13 more
ManuH
ManuH Veteran Member • Posts: 3,898
Re: K-7 firmware bug that affects leftmost AF point?

Yes, confirmed. 1/180s seems to be the limit. Below, it locks very easily, above it just gives up. And it's only the left that is affected. Very strange bug indeed. It doesn't affect me a lot because I almost never use these outer AF points but I think we will see a firmware update very soon.

But these kind of bugs are not unexpected at all when you see how late was the final firmware before the production started.

ManuH
ManuH Veteran Member • Posts: 3,898
Re: K-7 firmware bug that affects leftmost AF point?

I tested with the DA* 16-50 and the FA 35, no differences. And yes Horizon Correction has to be enabled.

ManuH wrote:

Yes, confirmed. 1/180s seems to be the limit. Below, it locks very easily, above it just gives up. And it's only the left that is affected. Very strange bug indeed. It doesn't affect me a lot because I almost never use these outer AF points but I think we will see a firmware update very soon.

But these kind of bugs are not unexpected at all when you see how late was the final firmware before the production started.

OP Pete Fang Senior Member • Posts: 1,900
Re: K-7 firmware bug that affects leftmost AF point?

Hi,

Many thanks for the test. I tested with the FA31 Limited. I think it is a reproducable firmware bug that's eaily corrected.

Peter

ManuH wrote:

I tested with the DA* 16-50 and the FA 35, no differences. And yes Horizon Correction has to be enabled.

ManuH wrote:

Yes, confirmed. 1/180s seems to be the limit. Below, it locks very easily, above it just gives up. And it's only the left that is affected. Very strange bug indeed. It doesn't affect me a lot because I almost never use these outer AF points but I think we will see a firmware update very soon.

But these kind of bugs are not unexpected at all when you see how late was the final firmware before the production started.

-- hide signature --

Peter Fang - Pentax user for more than 25 years: K-7 / MZ-S / Z-1 / SFX / LX
My PBase Gallery: http://www.pbase.com/pfang

 Pete Fang's gear list:Pete Fang's gear list
Pentax K-5 Fujifilm X-Pro1 Fujifilm X-T1 Fujifilm X-Pro2 Pentax K-1 +13 more
Edvinas Senior Member • Posts: 2,812
Re: K-7 firmware bug that affects leftmost AF point?

ManuH wrote:

But these kind of bugs are not unexpected at all when you see how late was the final firmware before the production started.

Well, these kinds of bugs are hinting that K-7 fiwmware is poor programmed spaghetti code...

-- hide signature --

Edvinas

 Edvinas's gear list:Edvinas's gear list
Nikon D3S Nikon D600 Nikon D700 Pentax smc FA 77mm 1.8 Limited Pentax smc FA 31mm F1.8 AL Limited +5 more
sbc Regular Member • Posts: 217
Re: K-7 firmware bug that affects leftmost AF point?

Man, am I glad you post this!

I don;t have a K-7 but I have a similiar problem with my K20D. All the other 10 AF points are accurate but not extreme left one. It will hunt a lot, even with a good target. Also its focus distance is different from the rest. It's about -5 to -7 on the AF adjustment scale, meaning that it is front focusing.

I did not realise it until now because I seldom use it. Also auto af point does not use the two outer points.

 sbc's gear list:sbc's gear list
Pentax K-5 IIs Pentax smc D-FA 50mm F2.8 Macro Pentax smc D-FA 100mm F2.8 macro Pentax smc DA* 60-250mm F4.0 ED (IF) SDM Pentax smc DA 35mm F2.4 AL +1 more
MightyMike Forum Pro • Posts: 40,698
Re: K-7 firmware bug that affects leftmost AF point?

I had an intermittent problem with AF twitching and non-locking on the K10D and K20D when i set it to auto-select and AF-C... after 2 years i realized that it was likely dust over a portion of the AF module and that a few quick blows onto the module seems to fix the problem, maybe this is not the same issue or maybe its related in some way as the AF twitching also had to do with how the firmware reacted to an AF point or 2 getting the wrong information
--
Mike from Canada

'I like to think so far outside the box that it would require a telephoto lens just to see the box!' ~ 'My Quote :)'

http://www.airliners.net/search/photo.search?sort_order=views%20DESC&first_this_page=0&page_limit=180&&emailsearch=mighty_mike88%40hotmail.com&thumbnails=

ManuH
ManuH Veteran Member • Posts: 3,898
Re: K-7 firmware bug that affects leftmost AF point?

Edvinas wrote:

ManuH wrote:

But these kind of bugs are not unexpected at all when you see how late was the final firmware before the production started.

Well, these kinds of bugs are hinting that K-7 fiwmware is poor programmed spaghetti code...

Maybe but as a programmer myself I would be very curious to know the cause of the bug because the three things (left AF point, shutter speed and horizon correction) seem to have no relationship at all.

OP Pete Fang Senior Member • Posts: 1,900
Re: K-7 firmware bug that affects leftmost AF point?

ManuH wrote:

Edvinas wrote:

ManuH wrote:

But these kind of bugs are not unexpected at all when you see how late was the final firmware before the production started.

Well, these kinds of bugs are hinting that K-7 fiwmware is poor programmed spaghetti code...

Maybe but as a programmer myself I would be very curious to know the cause of the bug because the three things (left AF point, shutter speed and horizon correction) seem to have no relationship at all.

Even if this gets fixed in a future firmware release I guess we'll never know the real cause from official sources. I do think that the K-7 having so many new features actually makes debugging the firmware quite some work, especially when the engineers were trying to meet some kind of deadline for 1.0 code...

Peter

-- hide signature --

Peter Fang - Pentax user for more than 25 years: K-7 / MZ-S / Z-1 / SFX / LX
My PBase Gallery: http://www.pbase.com/pfang

 Pete Fang's gear list:Pete Fang's gear list
Pentax K-5 Fujifilm X-Pro1 Fujifilm X-T1 Fujifilm X-Pro2 Pentax K-1 +13 more
Edvinas Senior Member • Posts: 2,812
Re: K-7 firmware bug that affects leftmost AF point?

ManuH wrote:

Maybe but as a programmer myself I would be very curious to know the cause of the bug because the three things (left AF point, shutter speed and horizon correction) seem to have no relationship at all.

This is why I mentioned spaghetti code...

-- hide signature --

Edvinas

 Edvinas's gear list:Edvinas's gear list
Nikon D3S Nikon D600 Nikon D700 Pentax smc FA 77mm 1.8 Limited Pentax smc FA 31mm F1.8 AL Limited +5 more
duncsuss Contributing Member • Posts: 933
Re: K-7 firmware bug that affects leftmost AF point?

Edvinas wrote:

This is why I mentioned spaghetti code...

Mention it a third time and it will become a fact -- that's how the internet works.

There really is not enough evidence to point at any particular type of coding strategy. I don't know the instruction set of the processor on which the firmware executes -- do you? (Does it have subroutine call and return instructions?)

-- hide signature --

Duncan

Simon97
Simon97 Veteran Member • Posts: 3,173
Re: K-7 firmware bug that affects leftmost AF point?

How would a fairly complex system as a digital camera OS not use sub routines?

I wonder what Pentax CPU derivative their cameras are based on? I think someone said Canon uses x86.

 Simon97's gear list:Simon97's gear list
Canon PowerShot ELPH 360 HS Panasonic Lumix DMC-GX85
_vlad Senior Member • Posts: 2,672
Re: K-7 firmware bug that affects leftmost AF point?

Well as a programmer I can tell only there is no code not a single one in the world without a bug.

Vlad

 _vlad's gear list:_vlad's gear list
Panasonic Lumix DMC-GX85 Panasonic GH5 Panasonic Lumix G Vario 14-140mm F3.5-5.6 O.I.S Panasonic 20mm F1.7 II Panasonic Lumix G 42.5mm F1.7 +5 more
ManuH
ManuH Veteran Member • Posts: 3,898
Re: K-7 firmware bug that affects leftmost AF point?

vlad wrote:

Well as a programmer I can tell only there is no code not a single one in the world without a bug.

Well, "Hello world" is pretty much bug free as far as I know

_vlad Senior Member • Posts: 2,672
Re: K-7 firmware bug that affects leftmost AF point?

OK Manu LOL here, it is better sometimes (or all the times?) to mute down the words. But I hope you know what I meant.

take care

Vlad

 _vlad's gear list:_vlad's gear list
Panasonic Lumix DMC-GX85 Panasonic GH5 Panasonic Lumix G Vario 14-140mm F3.5-5.6 O.I.S Panasonic 20mm F1.7 II Panasonic Lumix G 42.5mm F1.7 +5 more
rareshb Regular Member • Posts: 179
Re: K-7 firmware bug that affects leftmost AF point?

I'm pretty sure I can write a bugless 'Hello World'

_vlad Senior Member • Posts: 2,672
Re: K-7 firmware bug that affects leftmost AF point?

No objections

Vlad

 _vlad's gear list:_vlad's gear list
Panasonic Lumix DMC-GX85 Panasonic GH5 Panasonic Lumix G Vario 14-140mm F3.5-5.6 O.I.S Panasonic 20mm F1.7 II Panasonic Lumix G 42.5mm F1.7 +5 more
duncsuss Contributing Member • Posts: 933
Re: K-7 firmware bug that affects leftmost AF point?

Simon97 wrote:

How would a fairly complex system as a digital camera OS not use sub routines?

There are numerous methods, depending on the instruction set of the processor running the code. Programming languages can implement subroutine functionality even when the hardware upon which the code runs doesn't directly support them.

(Or are you saying that it's unthinkable that anyone would do it?)

My point, though, was that if one doesn't know the capabilities of the hardware, one should not make derogatory comments such as "poor programmed spaghetti code". It might have been the programmers' only option, and it could be exceptionally well programmed spaghetti code.

-- hide signature --

Duncan

networker Regular Member • Posts: 168
Re: K-7 firmware bug that affects leftmost AF point?

Yes,

confirmed

Tried DA17-70, DA* 50-135 and Sigma 18-125. With these lenses i can confirm the bug, even at ISO100. The cam locks easily with 1/160s.

networker

 networker's gear list:networker's gear list
Pentax *ist D Pentax K20D Pentax K-3 II Pentax K-1 Pentax K-1 II +19 more
Simon97
Simon97 Veteran Member • Posts: 3,173
Re: K-7 firmware bug that affects leftmost AF point?

I don't know. I may not understand the definition of "subroutine".

It would seem difficult to code for a device that has several subsystems like a camera without subroutines. Maybe it is done with branch statements?

Think of how busy that CPU is just before and during exposure. Exposure, AF, LCD updates, anti shake and the status of other subsystems I'm not thinking of.

 Simon97's gear list:Simon97's gear list
Canon PowerShot ELPH 360 HS Panasonic Lumix DMC-GX85
Keyboard shortcuts:
FForum MMy threads