LumoLabs: Pentax K-5 low light focus study is published

Started Feb 23, 2011 | Discussions
falconeyes
falconeyes Senior Member • Posts: 1,454
LumoLabs: Pentax K-5 low light focus study is published

LumoLabs: Pentax K-5 low light focus study is published

Fig.1: Accuracy of the Pentax K-5 phase detect AF vs. luminosity in EV. The above chart includes all measurements, i.e. various lenses, light colors, distances and apertures. The accuracy is measured as deviation of the focal plane from the sensor plane, in µm. [from the study linked below]

LumoLabs (this is what I call my lab activities) has decided to have a closer look at the K-5 low light front focus issue. After careful evaluation and many hundred test shots I must say that the issue is real.

A preliminary copy of the paper was provided to Pentax earlier this week and the head of Pentax Europe officially receives a printed copy today. I have been told that Pentax engineering will receive a copy too.

The study

The results are too complex to be presented in the scope of a forum article. Fig.1 above may provide a first idea of the work though.

Please, refer to the full blog article and to the complete paper for our findings:

(PDF version for download and offline reading)

You'll gain a deeper understanding of Fig.1 too

In a nutshell, this is what we find:

1. The K-5 as it presently ships indeed has a flaw in its phase detect autofocus module or software which causes it to front focus in low light below a lens-dependent threshold in EV.

2. If it does, it seems to consistently focus ≈ 255 µm behind the sensor plane (although with a significant ± 75 µm scatter of results which is about twice as large as the normal scatter of result).

3. Faster lenses seem to keep working in lower light but of course, are prone to more blur when the front focus does eventually happen. Slower lenses can already start to front focus at light levels metering as 4 EV or 6 EV even. A fast lens may work down to 0 EV in white light.

4. Light sources other than daylight emphasize this problem as they simply appear darker to the AF module. Moreover, it seems to be moderately color blind for red which further emphasizes the effect in deep tungsten light.

5. The effect is real and can negatively impact the daily work of a photographer. On the other hand, it is possible to run into a low light tungsten situation without the problem.

6. The paper clarifies conditions to hit or avoid the issue. White light (halogen is not white enough though) and a wide lens stopped down help to work around the problem. AF assist light typically doesn't help though. But an LED flash light does.

7. I don't rule out that a µm-valued variable becomes 0xFF (255) and causes a false shift of the focus plane by 255 µm. I call it the +0xFFµm hypothesis.

Please, refer to the full study and accompanying material before discussion. I am available to answer questions. Enjoy the read

 falconeyes's gear list:falconeyes's gear list
Sony RX100 III Pentax *ist DS Nikon D800E Nikon D500 808
13thBagel
13thBagel Senior Member • Posts: 1,246
Clarification?

May I ask if your results lie outside of (or inside) the 'difficult conditions' warning for AF/MF operation identified on page 75 of the K-5 Manual (English)? Or... Are you just trying to identify the low light/temperature AF/MF fail point threshold without regard to the manufacturer's warning?

Just curious... M

 13thBagel's gear list:13thBagel's gear list
Pentax Q7 Pentax K-1 II Pentax smc DA* 300mm F4.0 ED (IF) SDM Pentax 01 Standard Prime Pentax 06 Telephoto 15-45mm +4 more
bkpix
bkpix Contributing Member • Posts: 748
Re: LumoLabs: Pentax K-5 low light focus study is published

Once again, fabulous, precise work.

I wish you could apply your careful methodology to the SDM issue.

Bob

 bkpix's gear list:bkpix's gear list
Pentax K-5 Pentax K-5 IIs Canon EOS 6D Canon EOS 7D Mark II Pentax KP +13 more
dexmus Regular Member • Posts: 176
Re: LumoLabs: Pentax K-5 low light focus study is published

Falk - Thanks for putting time to analyze this issue so brilliantly.

Pentax - Please, fix the issue. Now you have the help of such a clear analysis. Don't do another SDM or k-7 mirror slap on this.

cheers,

Abhi

-- hide signature --
falconeyes
OP falconeyes Senior Member • Posts: 1,454
Re: Clarification?

13thBagel wrote:

May I ask if your results lie outside of (or inside) the 'difficult conditions' warning for AF/MF operation identified on page 75 of the K-5 Manual (English)? Or... Are you just trying to identify the low light/temperature AF/MF fail point threshold without regard to the manufacturer's warning?

Just curious... M

Thank you for the question.

It's actually answered in the study as I clearly documented the focus target. It perfectly avoids warnings a,c,d,e,f. It avoids warning b (low light reflectance) too bit there is a trap.

Because, even my target (very good reflection) doesn't reflect light if there is no light shining on it.

So, manufacturer warnings must be accompanied by a specifications which light level to avoid.

Fortunately, Pentax advertizes the operating range of its AF system to be > = -1EV.

Therefore, the region right of the vertical line passing through -1 (a bit left to the vertical line actually drawn in the chart) is avoiding all warnings and is within specification.

As it turns out, the Pentax specification is conservative actually. It actually locks below -1 EV as well.

The problem is that if it locks, it may be a false focus.

The manufacturer warnings are meant to describe situations where it is difficult to achieve a focus lock at all. That isn't the issue we are concerned about.

 falconeyes's gear list:falconeyes's gear list
Sony RX100 III Pentax *ist DS Nikon D800E Nikon D500 808
John Cafarella Contributing Member • Posts: 986
Re: LumoLabs: Pentax K-5 low light focus study is published

Nice work, appreciated,
--
John Cafarella
Melbourne, Australia

moving_comfort
moving_comfort Veteran Member • Posts: 8,227
Re: LumoLabs: Pentax K-5 low light focus study is published

falconeyes wrote:

...

7. I don't rule out that a µm-valued variable becomes 0xFF (255) and causes a false shift of the focus plane by 255 µm. I call it the +0xFFµm hypothesis.

.

This just seems too odd to be true... (and also would imply an easy firmware fix, correct?)

.

 moving_comfort's gear list:moving_comfort's gear list
Pentax K20D Nikon D800 Nikon AF-S DX Nikkor 35mm F1.8G Nikon AF-S Nikkor 300mm f/4D ED-IF Nikon AF Nikkor 20mm f/2.8D +10 more
moving_comfort
moving_comfort Veteran Member • Posts: 8,227
Confused about contradictory "EV specifications"

Falk, can you elaborate on this statement a bit? (from section 4.9, item 4)

4. EV specifications are typically according to external light meter readings and then the transition would probably appear to be at 2 EV rather than 0.5 EV.

What EV specification frameworks are at play here? It seems you refer to more than one.

(Thanks, btw, huge effort on your part, exceptional work.)

 moving_comfort's gear list:moving_comfort's gear list
Pentax K20D Nikon D800 Nikon AF-S DX Nikkor 35mm F1.8G Nikon AF-S Nikkor 300mm f/4D ED-IF Nikon AF Nikkor 20mm f/2.8D +10 more
moving_comfort
moving_comfort Veteran Member • Posts: 8,227
Re: LumoLabs: Pentax K-5 low light focus study is published

moving_comfort wrote:

falconeyes wrote:

...

7. I don't rule out that a µm-valued variable becomes 0xFF (255) and causes a false shift of the focus plane by 255 µm. I call it the +0xFFµm hypothesis.

.

This just seems too odd to be true... (and also would imply an easy firmware fix, correct?)

.

Never mind, you say as much in the actual paper:

"... and 255 is a good candidate for an “overflow” value (-1 or 0xFF). Of course, it would require a firmware variable to use a µm unit (or close to). Not likely but possible. ...

So, if I had only one vote: then it would go to an unsigned byte-valued variable which becomes 0xFF. Which means that it could be fixed by a firmware update ."

.

 moving_comfort's gear list:moving_comfort's gear list
Pentax K20D Nikon D800 Nikon AF-S DX Nikkor 35mm F1.8G Nikon AF-S Nikkor 300mm f/4D ED-IF Nikon AF Nikkor 20mm f/2.8D +10 more
PRISCILLA TURNER
PRISCILLA TURNER Senior Member • Posts: 1,383
Re: LumoLabs: Pentax K-5 low light focus study is published

"There seems to be no strong dependency on the light color. Except that colored light causes the transition to happen earlier as it means less usable light for the AF module. It seems to have a somewhat low sensitivity esp. for red light."

Does this mean that to use an 81A in incandescent would delay the transition a bit? I already favour this for the sake of richer and truer colour, in spite of the light-loss.

-- hide signature --

'To see, not with, but through the eye.' [William Blake]

http://www.flickr.com/photos/22905474@N06/

 PRISCILLA TURNER's gear list:PRISCILLA TURNER's gear list
Sigma DP1 Pentax K-3 II Sigma 10-20mm F4-5.6 EX DC HSM Pentax smc DA* 16-50mm F2.8 ED AL (IF) SDM Tamron SP AF 90mm F/2.8 Di Macro +3 more
andrew britten Senior Member • Posts: 1,165
Re: LumoLabs: Pentax K-5 low light focus study is published

falconeyes wrote:

AF assist light typically doesn't help though. But an LED flash light does.

That's strange, isn't it? Is it because of the spread of light, intensity or something else?

AB

 andrew britten's gear list:andrew britten's gear list
Pentax K10D Pentax K-7 Pentax K-x Pentax K-5 Pentax K-01 +1 more
HozicEmir Contributing Member • Posts: 739
Re: LumoLabs: Pentax K-5 low light focus study is published

remarkable work is up to pentax now even if falconeye left something now is their job to fix last generation of cameras.
I'm still on hold for buying until this is solved.

awaldram
awaldram Forum Pro • Posts: 13,175
Re: LumoLabs: Pentax K-5 low light focus study is published

Nice work.

I few to many variables from uncorroborated sources (some guy in some shop tested some k5 and said AF was consistent / some reports of AF assist working correctly etc

I don't think using these rumours to back your conclusions is correct for a scientific paper.

But overall solid work.

You conclusions seem over simplified and again rumour creeps in.

"Pentax communicates to work on the issue but with no positive outcome since several months already"

I have seen no such reports,It has been rumoured the Pentax work on a fix for the kr but there is no indication the two faults are related.

And lastly

"So, if I had only one vote: then it would go to an unsigned byte-valued variable which becomes 0xFF. Which means that it could be fixed by a firmware update"

If is is a single byte storage value whether 8 bit or more overflowing and that is the only byte allocated for this value then your conclusion re-firmware would be incorrect.

It could well require complete re-write of code portions, re-address memory mapping to free up two bytes of space, Or even worse it may not be possible at all without hardware changes. (more storage space).

 awaldram's gear list:awaldram's gear list
Pentax K-x Pentax Q Olympus PEN E-PM2 Pentax Q7 Pentax K-3 +17 more
mvb123 Veteran Member • Posts: 3,302
thanks for the good work!

It seems reproducable and with little variation so hopefully it can be solved with firmware or with a calibration.
--
Menno

DLBlack Forum Pro • Posts: 13,286
Good work

Thanks for all the effort in the testing and reporting. I still haven't noticed the FF issue with my K-5 (three weeks old) but I am sure from your testing that I should have this problem.

You did nail it with the "shutter blur" problem and explained why I had way too may blurry photos in the 1/60 to 1/100 shutter speed range. The K-5 doesn't have the shutter blur problem.

Hopefully a firmware update will fix the LL FF problem.

Dave
--

 DLBlack's gear list:DLBlack's gear list
Pentax K-7 Pentax K-5 Olympus OM-D E-M5 Olympus E-M1 Olympus E-M5 II +38 more
mgm2 Regular Member • Posts: 446
Re: LumoLabs: Pentax K-5 low light focus study is published

Very well done. As a layman and from a practical perspective, based on anecdotal evidence I've seen and the professional reviews out there, this is an issue on less than 1% of photos taken by K-5 users?

 mgm2's gear list:mgm2's gear list
Pentax smc DA 50mm F1.8
Rinneh Regular Member • Posts: 210
Re: LumoLabs: Pentax K-5 low light focus study is published

the only thing i hope is that it is fixable by firmware, it could mean something really bad for pentax if all K5's need to be returned for this.

ManuH
ManuH Veteran Member • Posts: 3,898
Re: LumoLabs: Pentax K-5 low light focus study is published

Thank you very much for clarifying this. Since the beginning I noticed this issue on my body and couldn't pinpoint what was really going on because in daylight I had no issue to get tack sharp pictures. The good news is that it's not my particular body. The bad news is that we have to wait for a fix (hopefully a new firmware, I don't want to send again my body for a few weeks as I have done for the sensor recall).

falconeyes
OP falconeyes Senior Member • Posts: 1,454
Re: LumoLabs: Pentax K-5 low light focus study is published

PRISCILLA TURNER wrote:

Does this mean that to use an 81A in incandescent would delay the transition a bit? I already favour this for the sake of richer and truer colour, in spite of the light-loss.

Maybe not. It seems to be the overall amount of light, weighted by the color sensitivities which matters. That wouldn't increase with a filter.

 falconeyes's gear list:falconeyes's gear list
Sony RX100 III Pentax *ist DS Nikon D800E Nikon D500 808
falconeyes
OP falconeyes Senior Member • Posts: 1,454
Re: LumoLabs: Pentax K-5 low light focus study is published

andrew britten wrote:

That's strange, isn't it? Is it because of the spread of light, intensity or something else?

Most LED flashes are brighter. But more importantly, they stay on during the entire focus operation.

 falconeyes's gear list:falconeyes's gear list
Sony RX100 III Pentax *ist DS Nikon D800E Nikon D500 808
Keyboard shortcuts:
FForum MMy threads