[Douglas_of_Sweden]

It may indeed be color related.
Take this:

FA50, f1.4, EV = 1,3
Fokused on the center folder.
The yellow folder is ca. 6 cm in front.
This is reproducible.





NB: The K20 is focusing correctly here!

Very interresting! It may explain why I havn't been able to reproduse this problem on my K-5. I'm thinking now that easiest way to test is would just be to dig up some old colored filters that I used for BW film.

Does anyone know if the Nikon D7000 also use a light wavelength sensor?
If the problem is somehow related to how the cameras make use of that information, they could solve it by simply ignoring the wavelength sensor at low light and use algorithms similar to K-7.

I'm sure Pentax is aware and working on this. And we know from before that they follow discussions on camera forum.
But just contacting usual personel will give us no current info on what their technicians are trying to do. They don't have that sort of info.
If it can be fixed by software changes, they will do that. Pentax has a better track record for upgrading their firmwhare than most.
But it may result in that that once fixed, the K-5 will not focus faster in low light than the K-7. I can live with that. After all, if we are entirely fair, the K-5 with its fantastic high iso performance have an advantage over previous models that it will still deliver good results in low light where older models would just deliver grainy crap.

BTW, loosing face is really different in Japan than from my corner of the planet. I learned that the hard way when I made a Japanese Brittish Airways employee in Tokyo, crying and running away from me back in 2004. When they had lost our luggage (still on Heatrhow in London), all I did was demanding that we should not have to wait for the luggage an extra 24-48 hours because of their shortage of cars to drive the luggage to us in Fuji-Yoshida. Suggesting, eeh, insisting, that it was their mistake and that they should rent a car if needed, was probably outside the frame of the instructions she had from above, and I put her in a corner without anyway to make both us and the boss happy. I'm ashamed for not understanding this then, but I was young.

[dlacouture]

If the problem is somehow related to how the cameras make use of that information, they could solve it by simply ignoring the wavelength sensor at low light and use algorithms similar to K-7.

AFAIK, the K7 also has the wavelength sensor... It was the first Pentax camera to have it.

[Jeff Charles]

Personally this thread has further convinced me to move on from Pentax, fortunately I don't have a huge investment in lenses.

A new Nikon D700 is only a little more $ than this K5 and simply runs rings around it in just about every dept.

When was the D700 brought out?

All the K5 has over it is size and absolutely nothing else.

I recently put a Canon 60d (amazingly it feels very similar to my beloved K10D) through many tests with just 18-135 - flash spot on, focus spot on, fast, silent accurate and the camera just got out the way.

If I had the K5 I would want it to perform in EVERY department with ease and better than the ageing D700 or the "toy" 60D.

Penny wise, pound foolish - the saying goes.

I will be surprised if Pentax does not fix this problem. It is a defect in a core capability of the camera and it has to be hurting sales. I don't own a K-5 - I am planning to buy in June - but if I did, I would be patient for a while longer.

Regarding the D700. It is not just a little bit heavier, and it costs nearly $800 more than the K-5. It also needs lenses . With those, I'd guess that the price gap would grow.

Jeff

[simico]

It may indeed be color related.
Take this:

Did you shoot that under tungsten or other light?

[philbaum]

Okay, i understand that the Japanese have a cultural adversion to not losing face. But i and many potential customers have an adversion to buying something that is broke, especially without any specific aknowledgement of the problem from Pentax (i realize there is third hand information from sources such as Falk, but still, nothing from Pentax officially)

It makes me very uneasy when Pentax doesn't advise us that they even know about the problem, i'd be ok with them not having a solution at this point, but not to even aknowledge the issue sounds like a Toyota program. Comeon Pentax, join the 21st century!

[Ray Pulley]

Very interresting! It may explain why I havn't been able to reproduse this problem on my K-5. I'm thinking now that easiest way to test is would just be to dig up some old colored filters that I used for BW film.

https://www.pentaxforums.com/forums/pentax-k-5-forum/126234-who-cares-bad-aut...ml#post1325432

I am 100% convinced (after testing 3 bodies in many conditions) that the light level is the main issue, but that is not to say that the color sensing sub-system is not the cause of the problem. It could be that this sub-system has difficulty detrmining the color temp in low light and introduces an error in trying to adjust.

Obviously, there is some AF shift with color temp, or Pentax would not have spent the time and money to add the color temp sensor to the K7 and the K5, but the idea of the color sensor is to correct for it.

If one assumes that Pentax was trying to counter-act the K20 type drift to FF under low and tungsten light, then the K5 would AF much better all the way down to the K5 specification of EV -1 There is just no way anyone is going to convince me, short of solid photographic tests, that ANY K5 can focus properly down to EV -1. In my testing, the focus is so far off at this light level that you canot identify the test target.

My K20 shifts to FF progressively as the light dims, bit it does not abruptly change at EV 3-4 like the K5. It stays just barely in the DOF at light levels where the K5 has dropped off the cliff.

While we did not get this volume of complaints about the K7 (which also has the color sensor), go to the other forums and dig up pots by Roland Karlsson about his K7, which behaves almost exactly like the K5 in low lilght.

Ray

[areidjr]

I just spoke to Pentax Customer Service about the FF issue. The Customer Service Rep indicated that there have been calls and a lot of internet chatter about the issue, but that there have been very few, if any, cameras sent in for service.

He suggested doing lens adjustments to correct for the issue. I explained that I had done that, but that the focus is then off in natural light. I told him this was my second one and that both had the problem.

He stated that there has been no word from Pentax Japan confirming that a defect exists (no surprise). He says they need cameras along a description of how to replicate the problem sent in to the service department for troubleshooting.

He also stated that, at my option, I could also return the camera for an exchange. However, B&H has been "out of stock" on the kit all of January and have been out of stock on the Body-Only for most of this week. If I try to exchange it I may, once again end up with old stock.

Any way, I though you might be interested in what Pentax USA had to say.

[Tommot1965]

The Customer Service Rep indicated that there have been calls and a lot of internet chatter about the issue, but that there have been very few, if any, cameras sent in for service..

I think the reason for that is most are waiting for a firmware fix in the next FW release..if its not addressed then, i think pentax will see a lot more returns then they have so far experienced

[mrbrightside]

I don't know why so many people expect Pentax to publicly make a statement about this defect as if it's the norm. Most companies (anywhere in the world) don't make any statements about something wrong with their product until they have a fix in hand -- even then they'll just post as a release note in the update.

I'm disappointed that k-5 having an issue like this though. To me, a newer, more expensive product should not perform worst than an older, cheaper product. I'll just wait for the price to drop a bit and see if Pentax will fix the issues. Of course, it's easier for me to stay calm when I haven't spend $1500 on something that's semi-working :-)

[Ray Pulley]

If one assumes that Pentax was trying to counter-act the K20 type drift to FF under low and tungsten light, then the K5 would AF much better all the way down to the K5 specification of EV -1 There is just no way anyone is going to convince me, short of solid photographic tests, that ANY K5 can focus properly down to EV -1.

2523: Photo by Photographer Paul Stenquist - photo.net

The images in this gallery are not mine, but are the work of a long time PDML member, Paul Stenquist who is now shooting a K5.

You can make the images smaller or larger by clicking on them, but this shot is a good example of the K5 FF issue. Look at the subject and the pool stick, and then look at the corner pocket area in the foreground. The camera missed by a large margin. Unless Paul chose to focus on the corner pocket, nothing in the plane of the subject or his cue and ball are in focus.

Note that I could not find a single comment in the thread (most from long time Pentax users on the list) about the FF in these images. The ISO, speed and aperture listed in the thread put these images at about EV 3-4 or so.

The photographer obviously thinks these images are ok, and he also recieved many favorable comments on them.

This is not in any way meant to disrespect Paul or anyone else on the PDML thread, but the point is that it is this sort of thing that I think is going on when many say that they are getting good AF in low light from the K5.

You can decide for yourselves, but even at the fairly small image size of the photos in this gallery, the FF is easily seen.

Ray

[dave kitson]

How low? ISO 10,000 f2.8 1/40s low. (how many EV is this?)

Using PhotoME, you can see a field in Manufacturer Notes labelled 'Effective LV' which is what people mean when using the term EV in the context of this discussion.

dave

[falconeye]

2523: Photo by Photographer Paul Stenquist - photo.net

Ray,
that photo may be manual focus. At least, the subject asks for an MF technique.

[Smeggypants]

https://www.pentaxforums.com/forums/pentax-k-5-forum/126234-who-cares-bad-aut...ml#post1325432

I am 100% convinced (after testing 3 bodies in many conditions) that the light level is the main issue, but that is not to say that the color sensing sub-system is not the cause of the problem. It could be that this sub-system has difficulty detrmining the color temp in low light and introduces an error in trying to adjust.

Obviously, there is some AF shift with color temp, or Pentax would not have spent the time and money to add the color temp sensor to the K7 and the K5, but the idea of the color sensor is to correct for it.

If one assumes that Pentax was trying to counter-act the K20 type drift to FF under low and tungsten light, then the K5 would AF much better all the way down to the K5 specification of EV -1 There is just no way anyone is going to convince me, short of solid photographic tests, that ANY K5 can focus properly down to EV -1. In my testing, the focus is so far off at this light level that you canot identify the test target.

My K20 shifts to FF progressively as the light dims, bit it does not abruptly change at EV 3-4 like the K5. It stays just barely in the DOF at light levels where the K5 has dropped off the cliff.

While we did not get this volume of complaints about the K7 (which also has the color sensor), go to the other forums and dig up pots by Roland Karlsson about his K7, which behaves almost exactly like the K5 in low lilght.

Ray

My K20 focusses spot on at EV2. And from experience I'm sure it focussing spot on below that too. I will test it out laters

[Ray Pulley]

Ray,
that photo may be manual focus. At least, the subject asks for an MF technique.

Falk,

The image is part of a series. Look at the rtest of them. The subject is generally not sharp. Unless Paul is pretty good at nailing consistent FF manually, it seems to me to be a bit of a stretch to assume a manual focus error.

Clearly Pentax has a pretty big issue with the K5 and Kr cameras relative to AF in low light.

All of my Pentax digital bodies (*istds, K100D, K10, K20, and I owned at least 2 copies of each version) suffered from this to one degree or another, and you can find threads on it going back years on many sites.

However, in the case of the K5 and Kr, it seems that we have a more significant shift that is being reported by many users of both cameras in many parts of the world.

Since Pentax has allegedly upgraded the AF systems since the K10/K20 days, including the addition of a color sensing sub-system to correct for any focus shift in differing light color temperature, one would expect that the camera would focus as least as well as the older models.

If nothing else, it HAS to focus acceptably over the entire published AF range specification of -1 EV to +18 EV and I, for one, have tried 3 bodies that in NO WAY will meet this specification.

How many K5 bodies have you tested for this problem?

So, as a loyal Pentax user since 1978, is it incumbent upon me to keep paying shipping and experiencing lost time with the camera I bought to send it to the Pentax outsourced repair facility to see if they can fix it?

As soon as I send it in, I own it and if it cannot be fixed by the repair folks, what then? I have a little window left to return my 3rd body and the question is fast becoming whether I just move on to another brand?

Ray

[Smeggypants]

Ray,
that photo may be manual focus. At least, the subject asks for an MF technique.

Why? The K-5 low light AF fault aside, shouldn't be a problem using AF on that subject.