[Rupert]

Why not? Microsoft has been doing it for decades.

Tom G

And we got Windows 7...is that what you are saying?

I have a question, that may have been addressed, but I missed....why can't the K5 be made to use the Assist Lamp on command? I have read where this results in good focus when it operates, and it would certainly work for my needs. Looks like it should be easy enough to program, and other than a few situations, it would cover all my uses.
Regards

[raphtze]

Raphtze, if your K5 is badly focusing in broad daylight, then you have another problem altogether...
Do you have other AF lens? Maybe both your Tokina need huge AF adjust. I understand you tried -6 as AF adjust, but why not lower still, if your pics are still unfocused?

i was afraid of that.....based on what others have said, with their replacement K-5...they had somewhat better focus? i have the kit lens 18-55mm and also a tamron 18-135mm they seem to focus slightly better, but they have smaller aperture to begin with. i'm just saying that when i manually focus with my tokina f/2.8 zooms, i can achieve rather sharp results--it boggles my mind that whatever method pentax uses to achieve autofocus simply doesn't work "out the box". as others have also mentioned, focus is sometimes hit or miss depending on what color, contrast the target is. i like to use single focus point then recompose.

-raph

[raphtze]

raphtze, you can also use LiveView. It should be better than manual.

fortunately, liveview works well and achieves good focus. however, it's cumbersome for me to use--i'm coming over from film SLR and i just like using the viewfinder better. i used to shoot weddings with my film SLRs....and i cannot imagine using LV for shooting tons of frames. well, it's ok.....to be honest, i'm just a casual snapshooter these days so liveview and/or manual focus, i can live with. i also used to shoot CART open wheel racers--and my tactic to achieve a sharp picture would be to pre-focus manually and pan--no need for AF my main film SLR was the PZ-1p, with SAFOX II....while it was slow, it was quite accurate no matter what AF lens i slapped on. for the life of me, i'm not sure how AF could 'regress' in these DSLR's (i'm getting a crash course learning about all the subtle nuances of DSLR photography)--of course, i'm no engineer either, so i don't know what the challenges are between film and digital cameras. if i'm repeating some old points, i apologize

-raph

[Smeggypants]

From Falk`s responses it is clear to me that the AF issue is not af a simple kind to PENTAX.
In case the matter would be a small bug somewhere it would have been solved by now.
I even start to doom think that PENTAX is not able to do any better than they did with the components they used to put the K5 together and they built something beyond their know how. We have seen Canikon struggling with AF under bad light and they have been able to solve it not even to a 100% extend. I start to believe that whilst with Canikon the issue is diminished to a rare issue PENTAX still has a long road to go before they do control the AF issues to the same extend. I don`t want to put petrol on the fire with this posting but this is a very serious option I assume and I would have liked nothing more than being wrong ! It is just this bad bad feeling that grips more and more with every day that passes by. Maybe they should hire FALK for a few months and though he will be quit expensive I am sure we would all be willing to pay 5 Euro more for the camera in order to have it functioning like a Ferrari in witch you can use more than the first 3 gears only. ( I am aware of the fact that I will get criticized by this comparison again but I have got used to it so no problem with that)

Hmmm. Well My K20D focusses in low light with no problem whatsoever. of coruse it doesn't focus as quickly as the K-5 but it does focus.

I've done side by side comparisons with my K20D and K-5.

If the K20 can do it the K-5 should be able to.

The issue shouldn't be impossible to resolve. Even if it's a hardware problem they could at least provide the addition of low light AF offset parameters.

I'm hoping for an eventual firmware fix, but I'd also be happy to input differing AF offsets for each lens at varying EV levels.

Anything to fix the problem eh?

[raphtze]

If the K20 can do it the K-5 should be able to.

i bought my wife a refurb'd K20D off buy.com (fabulous price at 500 bucks with the 18-55 lens) and you know what? it focuses in low light superbly--albeit slow as you mentioned; we shot some photos at a friends wedding and the photos all came out really sharp. using my f/2.8 zooms on it were also nice (but then the K20D had that issue with proper exposure using fast lenses).

[lock2nl]

It is more than a little disconcerting that Pentax is having to depend on its buyers to analyze and find corrections for a camera they built? I had a picture of a well equipped Lab and staff of Engineers that were fully capable of building a highly sophisticated electronic machine, and with all the necessary means to adjust and repair when necessary. Nothing against Falk, he is a very capable and valued Member here, but in my 30+ years of plumbing I can't ever recall having to ask a customer to diagnose a technical problem and recommend a solution for that problem...

I can understand your words, but things are a bit different I'm afraid.
These problems are usually not handled by labs that develop and innovate, but first of all by labs that work according tot strict testing protocols. Suppose the protocols for testing and QC are based on a set of assumptions A, than anything contradicting A wil be spotted as a problem. Nothing else. Of course, this is a highly simplified description but the principles do hold. However, if someone else uses a different set of assumptions B (e.g. derived from practical use), it may be sensitive to totally different things. Sometimes, this thinking 'out of the box' is neccesary to find the origin of the problem, and hence, a solution. This is where people like Falk come in. Basically, this is very common issue with complex products and organisations.

lock

[Rupert]

I can understand your words, but things are a bit different I'm afraid.
These problems are usually not handled by labs that develop and innovate, but first of all by labs that work according tot strict testing protocols. Suppose the protocols for testing and QC are based on a set of assumptions A, than anything contradicting A wil be spotted as a problem. Nothing else. Of course, this is a highly simplified description but the principles do hold. However, if someone else uses a different set of assumptions B (e.g. derived from practical use), it may be sensitive to totally different things. Sometimes, this thinking 'out of the box' is neccesary to find the origin of the problem, and hence, a solution. This is where people like Falk come in. Basically, this is very common issue with complex products and organisations.

lock

That makes some sense, I appreciate the response. Could you answer my question regarding the Focus Assist light? Is there any reason why this wouldn't be workable?
Regards

[Tommot1965]

That makes some sense, I appreciate the response. Could you answer my question regarding the Focus Assist light? Is there any reason why this wouldn't be workable?
Regards

I reckon the AF assist light coming on a bit earlier would be a solution...My K5 is almost in pitch black before it illuminates....but if using LV in the same light, the AF lamp turns on!!.

the only area this would not work IMO, would be using a Longer FL lens where the AF lamp wouldn't reach.....say Bride and groom at the alter , from 5 metres away...but in that case you could set the AF fine adjust for that shot...


Dunno if that was what you were asking Rupert, but thats my take on it

[Rupert]

Yes, that answered my question and would be a solution for my particular needs. I seldom shoot in the low light that is causing the problems and the Assist Lamp is not a problem for me if I could turn it on at my command. Looks like it should be a possible workaround for shooters like me until a better solution is found.
Best Regards

[areidjr]

the only area this would not work IMO, would be using a Longer FL lens where the AF lamp wouldn't reach.....say Bride and groom at the alter , from 5 metres away...but in that case you could set the AF fine adjust for that shot...

But, if you set the fine adjust for that shot, don't you open yourself up the the possibility that the shots with the assist light may suffer from BF?

[Smeggypants]

I reckon the AF assist light coming on a bit earlier would be a solution...My K5 is almost in pitch black before it illuminates....but if using LV in the same light, the AF lamp turns on!!.

the only area this would not work IMO, would be using a Longer FL lens where the AF lamp wouldn't reach.....say Bride and groom at the alter , from 5 metres away...but in that case you could set the AF fine adjust for that shot...


Dunno if that was what you were asking Rupert, but thats my take on it

Not at all IMO. There's loads of situations were the assist light is simply intrusive. The K20D manages to focus fine in these low light levels where the K-5 fails miserably. The assist light should be reserved for extremely low levels where you wouldn't expect the AF to cope on it's own.

[Cannikin]

I just got my new K-5 back from Pentax, serial number 398xxxx. After checking for stains, I went ahead and tested for FF in low light (which I didn't see in my original either).

100mm, f/2.8, near macro focus distance (just over 1 ft from target), center AF point.
~EV 0-1 (5 second exposure, f/2.8, ISO 100). Extremely dark. I struggle to imagine what someone would be shooting in conditions like this.

No FF. Minor backfocus actually (focus tended to lock about 1-2mm behind the target areas). Corrected with +1 in micro AF adjustment.

[falconeye]

Some got an impression from my response that my capabilities exceed that of Pentax' lab. That's not what I meant. But for a very isolated topic like the K-7 shutter-induced blur, I may have been able to dig deeper than possible in a milestone-driven development environment. These poor guys have to release 3 cameras a year. If I would have to do the same I might be unable to look sideways at all

It is as always: Engineers are never too few or too expensive. If you look at the budget of a corporation like Pentax (or their part within Hoya), the salary of development engineers must be neglegible. But if they are too few or not good enough, they'll spoil the manager's party

[Tommot1965]

But, if you set the fine adjust for that shot, don't you open yourself up the the possibility that the shots with the assist light may suffer from BF?

turn the AF assit lamp off when in that situation..turn it back on when your not..

[Wheatfield]

What should be of concern (I can't believe I'm actually contributing to this thread) is that the "cure" for this will be for the AF light to come on at EV4.
Unfortunately, while this would more or less work, it certainly isn't ideal. I would hate to see it as a switchable item in the menu as part of the cure (though it would be fine as an operational improvement) because it is one more thing for the photographer to have to worry about getting right during a shoot.
My K5 front focuses extremely badly below EV4, to the point that the AF is unusable below that light level, since the needed bias is rather more than what is available as micro adjustment (I suspect -18 or so would be about right).
For me, I can manual focus, but it is a problem in the studio if I want to use conventional AF.
Fortunately for me, it's been a rather slow winter, and Live View AF is working well enough, but it is a less than ideal situation.
At the moment, the rep that handles my area is on vacation, but I will be talking to him when he is back in the office to see what, if anything, they feel they can do about the front focusing.