[Wheatfield]

Through all this, has anyone bothered to actually quantify at what EV value the AF starts to become inaccurate? I ran a quicky test of my K5 and didn't see any focus errors, so I obviously didn't get the light dim enough (it was still bright enough to see clearly).
It would be nice to know what the cut off point is for when the AF goes wonky, so perhaps one of the people who is bent out of shape about this could actually do something along the lines of a scientific measurement to find out.

[pcarfan]

Through all this, has anyone bothered to actually quantify at what EV value the AF starts to become inaccurate? I ran a quicky test of my K5 and didn't see any focus errors, so I obviously didn't get the light dim enough (it was still bright enough to see clearly).
It would be nice to know what the cut off point is for when the AF goes wonky, so perhaps one of the people who is bent out of shape about this could actually do something along the lines of a scientific measurement to find out.

Well! I don't know whether I'd say I am bent out of shape over this, but I may be able to help, and in turn you can help me confirm your unit to be fine in these lighting condition.

The shot I took was taken at iso 1600, F4 at 1/8th....so, if you could check in similar light and confirm that you don't have any issues that would be great. The test needs to be done with a Screw drive lens AND a relatively fast lens (mine was the 16-45 F4).

P.S: I also noticed, if you do the LV test first and then the VF AF without changing the focus then the AF with the VF doesn't change and they both will look the same....so, you really need to do the VF focus first OR after doing the LV focus, focus to a differnet spot and redo the test with the VF.

[goubejp]

I'm calibrating the focus for my FA*600f4; weather sinks in France right now but here are the first results, all with natural light, target at 7 meters
Dim conditions : iso 100, 1/6 s at f4 : needs -7 correction

Average conditiions : iso100, 1/30 s at f4 : needs +1 correction (2.5 Ev from dim conditions)

Tests made with Tim Jackson focus test chart and K5 firmware 1.1. After focus shot with flash

When the weather will be better I'll try under "normal" conditions, preliminary tests already made gave +2 focus adjustment.

[pcarfan]

I ended up buying a K5 today ... lets just hope firmware 1.02 sorts all this out..

Couldn't resist any longer, hah!

I knew you wanted to know how much better the K-5 high iso performance was over the K-7, and I would agree that the K-5 iso 6400 is not as good as the iso 1600 of the k-7 as you had mentioned here ...

https://www.pentaxforums.com/forums/pentax-k-5-forum/123094-k-7-vs-k-5-a.html#post1271848

Eventhough the the K-5 iso 6400 is not as good as the K-7 iso 1600, the K-5 iso 12800 is better than K-7 iso 3200. It doesn't have that chroma blotchiness that you get with the K-7, especially when exposure is boosted in PP.

Let us know what you think once you get your K-5....hopefully it's something else than calling the K-7 owners too cheap to get the K-5

[Fer]

Couldn't resist any longer, hah!

I knew you wanted to know how much better the K-5 high iso performance was over the K-7, and I would agree that the K-5 iso 6400 is not as good as the iso 1600 of the k-7 as you had mentioned here ...

https://www.pentaxforums.com/forums/pentax-k-5-forum/123094-k-7-vs-k-5-a.html#post1271848

Eventhough the the K-5 iso 6400 is not as good as the K-7 iso 1600, the K-5 iso 12800 is better than K-7 iso 3200. It doesn't have that chroma blotchiness that you get with the K-7, especially when exposure is boosted in PP.

Let us know what you think once you get your K-5....hopefully it's something else than calling the K-7 owners too cheap to get the K-5

I agree. The most impressive thing about high ISO with K5 is how easy it is to clean it up just by adjusting the luminance slider in LR3. There's hardly any chroma noise.
One of the many reasons why I have not returned my K5 despite this AF issue. Bring on fw 1.02!

[Tommot1965]

Couldn't resist any longer, hah!

I knew you wanted to know how much better the K-5 high iso performance was over the K-7, and I would agree that the K-5 iso 6400 is not as good as the iso 1600 of the k-7 as you had mentioned here ...

https://www.pentaxforums.com/forums/pentax-k-5-forum/123094-k-7-vs-k-5-a.html#post1271848

Eventhough the the K-5 iso 6400 is not as good as the K-7 iso 1600, the K-5 iso 12800 is better than K-7 iso 3200. It doesn't have that chroma blotchiness that you get with the K-7, especially when exposure is boosted in PP.

Let us know what you think once you get your K-5....hopefully it's something else than calling the K-7 owners too cheap to get the K-5

I must admit...after just one evening with a K5..ive changed my whole mindset on it

...those 1.2- 2 stops difference..make all the difference ..and the higher DR is noticeable ..and that's just on a few dirty testing captures I've done with it so far ...the af is an improvement { dont mention tungsten} and live view..is now an option...crap AF in LV for me was not a option on the K7..

yep and your right...with all the posts I made voting to stay put with the K7..was just denial on my part...I knew CBA would get me sooner or later..

[Sparkle]

Buyer's Remorse adds to such problem(s) and vice versa.

[pcarfan]

Well! I tested the FA43 and it never showed a problem and with the FA*28-70 I am getting mixed results....I am so confused now.

[Wheatfield]

Well! I don't know whether I'd say I am bent out of shape over this, but I may be able to help, and in turn you can help me confirm your unit to be fine in these lighting condition.

The shot I took was taken at iso 1600, F4 at 1/8th....so, if you could check in similar light and confirm that you don't have any issues that would be great. The test needs to be done with a Screw drive lens AND a relatively fast lens (mine was the 16-45 F4).

P.S: I also noticed, if you do the LV test first and then the VF AF without changing the focus then the AF with the VF doesn't change and they both will look the same....so, you really need to do the VF focus first OR after doing the LV focus, focus to a differnet spot and redo the test with the VF.

I may do some checking after my chores. One thing that I wonder if people are looking at is the published EV sensitivity of the AF system, and then looking at what the actual EV is based on colour temperature.
It's pretty easy to run a system out of range when you only give it a very dim amount of light that is missing a huge % of the spectrum that the EV rating is based on.
For example, the guy who is warped about the AF failure in 1 candle worth of light is probably asking the camera to work well outside of it's operating range.

[apisto]

I may do some checking after my chores. One thing that I wonder if people are looking at is the published EV sensitivity of the AF system, and then looking at what the actual EV is based on colour temperature.
It's pretty easy to run a system out of range when you only give it a very dim amount of light that is missing a huge % of the spectrum that the EV rating is based on.
For example, the guy who is warped about the AF failure in 1 candle worth of light is probably asking the camera to work well outside of it's operating range.

Valid point. What has me holding off buying a K-5 until the next firmware update, however, is reports from multiple sources that the K-5 does not AF in low light as well as the K-7 did. It is reasonable to expect the K-5 to do at least as well as the K-7.

[Fer]

I may do some checking after my chores. One thing that I wonder if people are looking at is the published EV sensitivity of the AF system, and then looking at what the actual EV is based on colour temperature.
It's pretty easy to run a system out of range when you only give it a very dim amount of light that is missing a huge % of the spectrum that the EV rating is based on.
For example, the guy who is warped about the AF failure in 1 candle worth of light is probably asking the camera to work well outside of it's operating range.

Hey, I am "that guy"
I might be ready to concede (although reluctantly for a €1000 DSLR) that my K5 could be unable to focus on a contrasty object placed right next to a candle. However, like I mentioned later in this thread, I have repeated these tests in what I consider to be normal tungsten light, and the AF still doesn't work at all.

But again, this is no longer under discussion, given that Pentax HAS officially acknowledged the issue and said they are working in a firmware fix

[Wheatfield]

Valid point. What has me holding off buying a K-5 until the next firmware update, however, is reports from multiple sources that the K-5 does not AF in low light as well as the K-7 did. It is reasonable to expect the K-5 to do at least as well as the K-7.

Unfortunately, this isn't necessarily the case. Saying this, one would reasonably expect, for example, that the K7 would have had better high ISO performance than the K20, but it didn't.
Every new product is going to have quirks, and sometimes new isn't improved across the board.

[Fer]

Unfortunately, this isn't necessarily the case. Saying this, one would reasonably expect, for example, that the K7 would have had better high ISO performance than the K20, but it didn't.
Every new product is going to have quirks, and sometimes new isn't improved across the board.

Agreed, but this is much more than a quirk. I can't shoot my K5 at, say, a dinner party because the AF simply does not work

[Todd Adamson]

Pentax HAS officially acknowledged the issue and said they are working in a firmware fix

I'm sure I've lost this in the shuffle, but have they really? I see posts about dealers claiming that a firmware update is being worked on, and these are sometimes countered by posts saying the dealers are just doing damage control. I can't recall seeing any official, straight-from-the-horse's-mouth acknowledgment from Pentax. As I said, I've likely just overlooked it, but maybe someone will be kind enough to point it out for me?

[bjan]

Please guys stop doing tests with your K5`s . AF simply does not work under artificial light ! This applies also to your camera. Again the person who has a k5 without this fault can sell it to me for 200% of the new price !
Till today no official prove that PENTAX is working on the issue. I advise you not to shoot any Christmas dinner with your K5 as it will be quit a dissapointment for many shots.