Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
02-23-2011, 07:02 PM   #76
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
Original Poster
QuoteOriginally posted by Christine Tham Quote
Here is an attached photo showing the K-5 misfocusing on an incandescent torch - hopefully you can see on this photo the camera is FF by quite a large amount - surprising since the light source is relatively strong so this is not an issue relating to light level at all.
Christine, thanks for your experiments.

I think it is light brightness AND light color related.

But there may be more to it than I found in my study.
As noted here, having three tungsten bulbs with a color filter can only spawn so much of the spectrum. Maybe not enough.

A comment on my blog makes me think most.

It may be that a strong halogen light triggers the focus jump while a weak LED light does not. I feel tempted to include both as yet another light color into my data. However, I won't find the required time soon. Esp. studio halogen modelling lamps seem to have a high threshold.


OTOH, I may already cover this phenomenon in my study. Fig.4 shows that blueish light "jumps" focus at about -2EV darker light levels than red. That may cover already the difference between tungsten and LED. Maybe not Halogen though.


Another test I couldn't do is a comparison of scene brightness vs. target contrast. In theory, only their product should matter.

Eventually, it is the task of Pentax to figure it all out...


Last edited by falconeye; 02-23-2011 at 07:53 PM.
02-23-2011, 08:02 PM   #77
Veteran Member
Christine Tham's Avatar

Join Date: Nov 2008
Location: Sydney, Australia
Photos: Gallery
Posts: 2,269
QuoteOriginally posted by falconeye Quote
It may be that a strong halogen light triggers the focus jump while a weak LED light does not.
I can definitely confirm this. The camera is able to focus on a weak LED light (first photo), but misfocuses on a relatively strong yellow light, as you can see from my second photo.

What your study seems to confirm is that it's not a consistent misfocus, there are times when the camera does it right, but there's a definite "shift" when it happens. I've now seen this too, on my much less rigorous testing.

What do you think of the theory proposed by dlacouture that the behaviour of the AF assist light may also be lens dependent?
02-23-2011, 10:14 PM   #78
Veteran Member




Join Date: Jan 2007
Location: Hawaii
Posts: 592
1) Does the K7 show this behavior?

2) Does the 645D show this behavior as well since it (presumably) shares the same focusing system?
02-23-2011, 10:24 PM   #79
Banned




Join Date: Jan 2011
Posts: 242
QuoteOriginally posted by HawaiianOnline Quote
2) Does the 645D show this behavior as well since it (presumably) shares the same focusing system?
Maybe the AF system of K5/Kr is like the 645D's, but I'm (almost) convinced is not identical. Since the 645D has not problems (I believe it's so, at a 10000 euros camera), it's logical to suppose that on K5/Kr it can be something like a Safox9 light. The rest is just marketing.


Last edited by ursamajor; 02-23-2011 at 11:13 PM.
02-23-2011, 11:01 PM   #80
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 8,934
QuoteOriginally posted by Christine Tham Quote
So there still may be hope for a firmware fix.
Do you now acknowledge that there is a problem? For a long time you claimed that your K-5 copy was performing above expectations. Did you change your mind and if so, what caused this to happen? Did you stumble across a colour dependency that you somehow managed to avoid in earlier tests?

These are honest questions. I'm interested in finding out where your K-5 may perform better than others and what may have caused you to consider that even your K-5 has a problem too.

QuoteOriginally posted by Christine Tham Quote
I'm starting to think that the FF problem may be related to spherical aberration
This wouldn't explain the dependency on light level and/or light temperature.

QuoteOriginally posted by falconeye Quote
It may be that a strong halogen light triggers the focus jump while a weak LED light does not.
So do you think that scenario isn't covered by the three lightbulbs you used?
Should we not see bright (x-axis) red dots with high defocus and dim blue dots with low defocus in your graph then? As you say, I think Fig. 4 (and others) seem to provide evidence to the contrary.

I understand that different light sources have different spectra and, e.g., why metamers exist, but I wouldn't expect the AF module to be so complex as to depend on certain spectral composition structures.

It is not entirely impossible that the various observations made regarding the K-5's AF behaviour are not caused by one and the same fault. Maybe the colour temperature sensor is defective in a subset of the units. Just a thought if we find that there is no way to explain all observations by a single cause.

QuoteOriginally posted by falconeye Quote
Eventually, it is the task of Pentax to figure it all out...
And they should have a much easier task doing so with less of guesswork about the AF module and access to in-camera values while the AF does its work.

QuoteOriginally posted by HawaiianOnline Quote
2) Does the 645D show this behavior as well since it (presumably) shares the same focusing system?
Maybe someone can saw their 645D in two halves so that we can have a look at its AF module.
02-23-2011, 11:32 PM   #81
Veteran Member
Christine Tham's Avatar

Join Date: Nov 2008
Location: Sydney, Australia
Photos: Gallery
Posts: 2,269
QuoteOriginally posted by Class A Quote
Do you now acknowledge that there is a problem? For a long time you claimed that your K-5 copy was performing above expectations. Did you change your mind and if so, what caused this to happen? Did you stumble across a colour dependency that you somehow managed to avoid in earlier tests?
I am not sure I understand your questions.

1. I have never said my K-5 *didn't* have a problem. Right from the first post, I said I have found evidence of my K-5 to have a tendency to FF in low light conditions, described exactly what those conditions were, and then compared the K-5 performance with two other cameras. I have also posted photos showing my K-5 exhibiting FF in low EV halogen light situations

2. I never said my K-5 was performing "above expectations." Whose expectations? Mine? Yours? A typical purchaser?

3. I did say that my K-5 did not seem exhibit behaviour that I would consider to be worse than a K10D and a Canon EOS-1D. Indeed, it's focusing ability could perhaps be regarded as superior to those cameras in low light situations (provided either Live View or AF assist was used to do the focusing). Perhaps, in that sense, it is "above" expectations - but to be honest I had no expectations before I bought the camera - I had an open mind.

4. I also said that I have never encountered any issues shooting in "real world conditions" as opposed to artificial test scenarios which I don't consider to be typical shooting conditions. By and large, I am pretty happy with all the shots I have taken so far on this camera.

5. Therefore, I stated that with regards to my camera, and my camera only, that the issues with that I have discovered with low light focusing anomalies are not "significant" - they are not in my mind justification for me returning the camera within it's 45 day money back guarantee period, or to send it in for servicing.

6. The latest round of tests I have done have not changed that opinion. They have simply made it clearer what the conditions for the focusing anomalies are. I still don't consider those conditions to be significant in terms of how I would use the camera. If I had known exactly what these conditions were prior to purchase, I would still have bought the camera.

7. I understand other people may feel different about these focusing anomalies. Clearly they are entitled to do so. My advice to them would be if they are not happy with these anomalies, they should return the camera or ask for it to be exchanged/serviced.

8. My unsubstantiated opinion, and my opinion only, are that the anomalies I have confirmed via testing are probably not fixable by firmware, and may require a substantial redesign on the camera. Certainly, if I am proven wrong, I will be pleased and will download and apply this firmware fix.

9. However, I neither demand nor require Pentax to "fix" these anomalies - in my experience all cameras have focusing anomalies in low light - certainly all the cameras I am currently using at the moment have interesting "quirks" in low light (2 Pentax, 2 Canon, not to mention 3 HD camcorders - 2 Sony, 1 Canon).

10. Until then, I will continue to enjoy using my camera. This Sunday is World Pentax Day, and I intend to join other owners of the K-5 in using it, and I hope I will take a photo (or several) that I will be proud of.
02-23-2011, 11:47 PM   #82
Veteran Member
Christine Tham's Avatar

Join Date: Nov 2008
Location: Sydney, Australia
Photos: Gallery
Posts: 2,269
QuoteOriginally posted by Class A Quote
This wouldn't explain the dependency on light level and/or light temperature.
Actually, it does. Light bends at different angles depending on frequency/colour - if you recall your high school physics.

If it is spherical aberration, then it would explain why the problem is more pronounced on some lenses, and less pronounced on others.

It also explains why the problem is more pronounced with yellow light than neutral light.

It doesn't explain dependency on light level, but as you can see from my test photos, the problem is not that dependent on light level (apart from the usual caveat that the lower the light level, the less accurate AF is simply due to signal/noise ratio).

It is possible to see the problem with a strong yellow light, and not see the problem with a very weak LED light (which has a fairly high colour temperature). For example, I am amazed that my K-5 achieved perfect focus with that single battery - because the EV value for that shot is -0.14. The EOS-1D would have just refused to focus.
02-24-2011, 12:03 AM   #83
Site Supporter




Join Date: Dec 2010
Location: Oregon
Posts: 44
At least we can all finally agree that their is a focusing flaw; the severity of the flaw, however, is up for interpretation. For me, and I'm sure many others, its simply extremely disappointing that this camera possess such a documented focusing anomaly. It puts a irritating taint on such a otherwise commendable camera.

Pentax may, or may not solve the issue but ultimately consumers have see it for for themselves; at least we know what to look for now.

After reading all (OK most) of your post, I've decided to go ahead and purchase a K5. Thanks to everyones input its obvious that I need to see the issue for myself. The degree of the issue seems to fall somewhere between: an annoyance and something you have to keep in mind, it depends on how much low light photography you do.

Thanks for hard work.


Last edited by PixleFish; 02-24-2011 at 12:14 AM.
02-24-2011, 12:15 AM   #84
Veteran Member




Join Date: Jul 2008
Location: Var, South of France
Photos: Albums
Posts: 1,071
QuoteOriginally posted by HawaiianOnline Quote
1) Does the K7 show this behavior?
From my one year with it, I'd say no...
I've upgraded from the K7 to the K5 4 months ago, and the FF problem just leaped at me... This was while shooting in the same situations, using the same calibrated lenses that perfectly focused in broad daylight...

This was something I've focused upon (pun intended ) when I got the K7, as I've found my K10 and K20 to be quite prone to FF under tungsten lights...
02-24-2011, 12:22 AM - 1 Like   #85
Pentaxian
jsherman999's Avatar

Join Date: Dec 2007
Photos: Gallery
Posts: 8,228
This situation illustrates something I've noticed in various forums over the years (primarily Nikon/Pentax,) and that is:

When an issue like this pops up, for every person actually experiencing 'user error', there is probably another who actually has a problem with their equipment - but that problem is masked by a lower standard of expected performance.

"My camera/lens has no problem" often really means "I'm OK with it's performance - problem or not."
02-24-2011, 12:25 AM   #86
Veteran Member
Christine Tham's Avatar

Join Date: Nov 2008
Location: Sydney, Australia
Photos: Gallery
Posts: 2,269
QuoteOriginally posted by PixleFish Quote
At least we can all finally agree that their is a focusing flaw; the severity of the flaw, however, is up for interpretation. For me, and I'm sure many others, its simply extremely disappointing that this camera possess such a documented focusing anomaly. It puts a irritating taint on such a otherwise commendable camera.
I think the service Falk has done to the community is provide reliable data and evidence, and a very good description of what the issue is - pertaining to his K-5, and his K-5 only. Prior to this, we had more or less conflicting anecdotal statements made by various posters (including myself), but not rigorous tests.

At this stage, there is no evidence that this problem exists on all cameras to the same degree. My own testing shows that the problems seems to be slightly different in magnitude and characteristic on my camera, but it's hard to tell since I am running different tests and I don't claim my tests are rigorous.

I am only saying that the results do not imply the problem exists on all units because they are posters who claim they do not have a problem, and I would like to give them the benefit of the doubt. However, if my suspicions are correct, then the problem will be present on all units, but perhaps not in the same degree depending on lens used, manufacturing tolerance, etc. etc.

Personally, I am glad that the problem is now described so well. I wish low level focusing anomalies are as well documented on my other cameras - they too have their share of quirks.

The good news is - because the problem is so specific, there are many workarounds so there should be no reason why you can't enjoy your planned K-5 purchase.

When shooting in low light or yellowish light, use Live View, or manual focus (if you have an MF focusing screen), or use a LED light to illuminate the subject prior to shooting. Using any of those workarounds will ensure you have a perfectly focused photo. Or stop down the lens.

Last edited by Christine Tham; 02-24-2011 at 12:38 AM.
02-24-2011, 01:11 AM   #87
Veteran Member
Kenn100D's Avatar

Join Date: Feb 2007
Location: Paranaque City, Philippines
Photos: Gallery
Posts: 645
I think also this is the first time pentax choose phase detection AF and when K-5 was released it was not yet that ready. A lot of tweaking has to be done.

Engineers will have to write how the PDAF works and behave in Pentax system.

Thanks Falk for the work. Hope Pentax will fix the problem related to your study.
02-24-2011, 01:41 AM   #88
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
A slight FF in tungsten light seems to be commonplace with all brands/models. K5's problem gets quite pronounced below a rather sharp threshold. Also, the camera still locks focus quite snappily and indicates it has done so; the implication is that AF is almost working and eliminating that "almost" should/might be possible.

In a very strict sense we cannot say that all K5s have the problem without checking them individually, which is impossible. Even requiring a statiscally significant sample is hardly practical for us, so we are left from extrapolating from smaller numbers. This is rather obvious and as such could be assumed without specific qualification. Actually, if we had a better handle as to the mechanism of the tungsten light FF error (the implementation of SAFOX+, in fact) we might also be able to extrapolate from a limited sample on a quite strict basis. However, assuming that if one K5 has a (mis)feature the other copies have that too is not an unreasonable thing to do, unless there is a specific reason to doubt that IMO.

It is perculiar how we end up looking at a man made thing as if it was a mystery posed by nature herself .
02-24-2011, 02:05 AM   #89
Veteran Member
Christine Tham's Avatar

Join Date: Nov 2008
Location: Sydney, Australia
Photos: Gallery
Posts: 2,269
QuoteOriginally posted by jolepp Quote
A slight FF in tungsten light seems to be commonplace with all brands/models. K5's problem gets quite pronounced below a rather sharp threshold.
The second statement is unfortunately unsubstantiated.

If you had said "*My* K5's problem gets quite pronounced below a rather sharp threshold." that would have been okay. But to imply that it is "some", "most" or "all" K5 requires substantiation.

"My" K-5 varies depending on which lens I use, and light temperature, and even from shot to shot (depending on the where the lens focus position is prior to the shot).

For example, note the shot I posted of a perfectly focused battery, taken at EV below 0.

I don't believe there is a "sharp" threshold, and Falk's test does not necessarily indicate so either. Falks test indicate that when the camera misfocuses, it is a discrete change, not a gradual misfocus. But the conditions under which this occurs is not well defined.
02-24-2011, 02:15 AM   #90
Veteran Member




Join Date: Nov 2008
Location: Umeň, Sweden
Posts: 755
I'm crossposting this from another thread just to confirm (for the nth time) that this is indeed a real issue, acknowledged by official Pentax sources, and that it is specific to the K-5:

I called up the Pentax service center in Sweden today. They said that at the moment, there is nothing they can do about K-5's turned in for artificial light FF. They said it is a known issue, they confirmed that Pentax Japan is working on a fix, and they're just waiting at the moment. He recommended I use Live View focusing as a temporary workaround -- but me sending in my camera for repair two days ago was all for naught. There is nothing they can do until Pentax Japan releases a patch.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
copy, ev, focus, front, issue, k-5, light, pentax, pentax news, pentax rumors, plane, study
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Any tips for low light focus with K-5 designinme_1976 Pentax K-5 5 11-21-2010 08:38 PM
focus hunting in low light sorin Pentax DSLR Discussion 12 07-17-2010 02:20 PM
Low Light auto focus JohnKSA Troubleshooting and Beginner Help 6 03-10-2010 04:19 AM
Pentax Low Light Focus indy1984 Pentax DSLR Discussion 15 02-17-2010 09:42 AM
EV low light focus question tarsus Photographic Technique 3 06-26-2008 08:22 AM



All times are GMT -7. The time now is 10:38 AM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top