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

Graham67 · 03-10-2011, 03:55 PM

Having taken a few dozen test shots around the house in dim light with my Sigma 30mm at f/1.4, I have to say I'm *very* pleased with Firmware V1.03.

Christine Tham · 03-10-2011, 06:02 PM

Originally posted by Class A

You should have clarified (as you did in another post) that you allowed the AF assist to come on in some occasions.

Thanks - I probably should have.

I think the AF assist was engaged for most of the shots (including both those that FF and those that were fine) - hence even using the AF assist the camera still misfocuses.

This is an apparent behaviour change from the factory firmware - the AF assist is much more likely to turn on than the factory firmware.

**Aristophanes** · 03-10-2011, 06:39 PM

Originally posted by glanglois

We need a new forum: Falk Lumo Reports. Saves your valuable time.

And a PayPal button so we can donate a Pentax logo custom seat cover for his Audi.

Erik · 03-11-2011, 01:44 AM

Originally posted by Aristophanes

And a PayPal button so we can donate a Pentax logo custom seat cover for his Audi.

He drives an Audi? Guy's got taste

dlacouture · 03-11-2011, 05:53 AM

Originally posted by Erik

He drives an Audi? Guy's got taste

Guy's is German, IIRC, so Audi, BMW, or VW were pretty much mandatory

Same thing for me : Peugeot and Renault (mmmmh, only this does not sound so cool

)

**NaClH2O** · 03-11-2011, 12:29 PM

Originally posted by dlacouture

Guy's is German, IIRC, so Audi, BMW, or VW were pretty much mandatory

Same thing for me : Peugeot and Renault (mmmmh, only this does not sound so cool

)

you could always go with Citroen, they ain't so bad

NaCl(they do pretty good in the World Rally Championship)H2O

gazonk · 03-11-2011, 04:26 PM

Originally posted by dlacouture

Same thing for me : Peugeot and Renault (mmmmh, only this does not sound so cool

)

Bah, I wouldn't trade our Peugeot for an Audi

dlacouture · 03-12-2011, 02:25 AM

Originally posted by NaClH2O

you could always go with Citroen, they ain't so bad

Man, I'd love a C6, those "hydractive" pneumatic shock absorbers are pure bliss...

falconeye · 03-13-2011, 05:34 PM

Originally posted by Class A

Do you think it might perhaps be good to get some data with the old firmware regarding the potential peak of defocus around EV 8?

I will not evaluate the EV6-8 region in this context. If the problem persists, it may become a separate topic. Currently, I live under the assumption that a mismatch of color at the AF point and globally, may have caused issues around that EV range. I'll include a look at this "color mismatch" in my updated paper, although at lower EV.

I've already done all test shots and now need to run the analysis and write it up.

Class A · 03-13-2011, 06:43 PM

Hi Falk,

Originally posted by falconeye

I will not evaluate the EV6-8 region in this context.

Fair enough, I fully understand. I just thought I mention the problem before you update the firmware as there is probably no straightforward way of reverting to an older version (that doesn't involve hacking the old firmware to pretend it isn't older).

Originally posted by falconeye

I've already done all test shots and now need to run the analysis and write it up.

Great! My gut feeling is that while some improvement has been achieved there is still a residual problem. So probably one, rather than two thumbs up.

Class A · 03-13-2011, 06:46 PM

Originally posted by NaClH2O

you could always go with Citroen, they ain't so bad

Citroëns are funtastic (sic).

mrmentera · 03-17-2011, 02:13 PM

tomtomtomtomtomtomtomtomtomtomtom...
as in Jaws..sort of...eh.

falconeye · 03-17-2011, 05:59 PM

Hi friends,

I am through the analysis of the v1.03 data and I am a bit confused. Maybe, somebody can help before I write my update.

According to my measurements, v1.03 brings no significant progress IF the following conditions are met:

1. The FA 31 lens is used at 1m distance and f/1.8 aperture.
2. The background is bright.
3. The camera white balance is manually set to daylight (maybe doesn't matter).
4. The light color is tungsten (maybe doesn't matter).

Wrt (3.), I see some positive effect if the camera white balance is set to automatic or tungsten. However, I don't have enough data points (I have 21 measurements for this case) to call this effect significant, i.e., these 21 data points all lie within the cloud of 94 data points obtained with daylight white balance. I even see some negative effect of v1.03 if the subject has a red color and the lamp has not.

OTOH, I trust the numerous positive reports and I must assume that I must miss this effect for some reason.

Maybe, the white balance effect is significant. Maybe, the effect is only noticeable with lenses slower than f/1.8. Or maybe, only at higher EV values with darker backgrounds.

Unfortunately, I am far away from my lab and can't do further measurements for the next three weeks.

Is there anybody wanting to check if the positive v1.03 effect disappears with a wide aperture and/or a non-matching camera white balance?

Thanks everybody.

Duplo · 03-17-2011, 08:20 PM

Originally posted by falconeye

Hi friends,

I am through the analysis of the v1.03 data and I am a bit confused. Maybe, somebody can help before I write my update.

According to my measurements, v1.03 brings no significant progress IF the following conditions are met:

1. The FA 31 lens is used at 1m distance and f/1.8 aperture.
2. The background is bright.
3. The camera white balance is manually set to daylight (maybe doesn't matter).
4. The light color is tungsten (maybe doesn't matter).

Wrt (3.), I see some positive effect if the camera white balance is set to automatic or tungsten. However, I don't have enough data points (I have 21 measurements for this case) to call this effect significant, i.e., these 21 data points all lie within the cloud of 94 data points obtained with daylight white balance. I even see some negative effect of v1.03 if the subject has a red color and the lamp has not.

OTOH, I trust the numerous positive reports and I must assume that I must miss this effect for some reason.

Maybe, the white balance effect is significant. Maybe, the effect is only noticeable with lenses slower than f/1.8. Or maybe, only at higher EV values with darker backgrounds.

Unfortunately, I am far away from my lab and can't do further measurements for the next three weeks.

Is there anybody wanting to check if the positive v1.03 effect disappears with a wide aperture and/or a non-matching camera white balance?

Thanks everybody.

first of thanks for all the hard work.

I have got to admit to not having read you paper fully yet.

Talking lenses in the F2.0 and below range, I have tested with the FA*24, FA43ltd and FA77, the 24 and77 were pretty accurate already before 1.0.3, but 1.0.3 has really improved on the 43, so now all of them are spot on, with only an occasional miss.

Is it possible that the reason you are not seeing an improvement could be an issue with your specific lens/body combination?

Class A · 03-17-2011, 08:21 PM

Hi Falks, thanks for the update.

Originally posted by falconeye

According to my measurements, v1.03 brings no significant progress IF the following conditions are met:

Wow, that's bad news. Even in the conditions you list, AFAIC, there shouldn't be such a discontinuous jump in the choice of focus plane. Unless the colorimetric sensor really has such a sharp transition to blindness, a discontinuous behaviour would still point to a firmware bug (or a limitation of a correction table size).

Is your set of criteria a positive list only? In other words, I guess that I cannot conclude from it that there is a significant improvement if the background is not bright, right?

Originally posted by falconeye

Wrt (3.), I see some positive effect if the camera white balance is set to automatic or tungsten.

Oh dear, would be a shame if the K-5 were relegated to K-r tricks...

(I know that this effect is not confirmed but it seems that this would be a reasonable step to take in cases where the colorimetric sensor isn't useful anymore).

Originally posted by falconeye

Maybe, the effect is only noticeable with lenses slower than f/1.8.

Ray Pulley has posted test shots that show a significant change to the focus plane when using an FA 50/1.4 at EV 2 (<- Sekonic L-358 reading at focus target).

To be honest, overall, that's quite disappointing news...

03-17-2011, 05:59 PM	#343
falconeye Veteran Member Join Date: Jan 2008 Location: Munich, Alps, Germany Photos: Gallery Posts: 6,871 Original Poster	Intermediate update for firmware 1.03 Hi friends, I am through the analysis of the v1.03 data and I am a bit confused. Maybe, somebody can help before I write my update. According to my measurements, v1.03 brings no significant progress IF the following conditions are met: 1. The FA 31 lens is used at 1m distance and f/1.8 aperture. 2. The background is bright. 3. The camera white balance is manually set to daylight (maybe doesn't matter). 4. The light color is tungsten (maybe doesn't matter). Wrt (3.), I see some positive effect if the camera white balance is set to automatic or tungsten. However, I don't have enough data points (I have 21 measurements for this case) to call this effect significant, i.e., these 21 data points all lie within the cloud of 94 data points obtained with daylight white balance. I even see some negative effect of v1.03 if the subject has a red color and the lamp has not. OTOH, I trust the numerous positive reports and I must assume that I must miss this effect for some reason. Maybe, the white balance effect is significant. Maybe, the effect is only noticeable with lenses slower than f/1.8. Or maybe, only at higher EV values with darker backgrounds. Unfortunately, I am far away from my lab and can't do further measurements for the next three weeks. Is there anybody wanting to check if the positive v1.03 effect disappears with a wide aperture and/or a non-matching camera white balance? Thanks everybody.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Any tips for low light focus with K-5	designinme_1976	Pentax K-5 & K-5 II	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

03-10-2011, 03:55 PM	#331
Graham67 Inactive Account Join Date: Sep 2008 Location: Leicester, England Photos: Gallery Posts: 89	Having taken a few dozen test shots around the house in dim light with my Sigma 30mm at f/1.4, I have to say I'm very pleased with Firmware V1.03.

03-10-2011, 06:02 PM	#332
Christine Tham Veteran Member Join Date: Nov 2008 Location: Sydney, Australia Photos: Gallery Posts: 2,269	Originally posted by Class A You should have clarified (as you did in another post) that you allowed the AF assist to come on in some occasions. Thanks - I probably should have. I think the AF assist was engaged for most of the shots (including both those that FF and those that were fine) - hence even using the AF assist the camera still misfocuses. This is an apparent behaviour change from the factory firmware - the AF assist is much more likely to turn on than the factory firmware.

03-10-2011, 06:39 PM	#333
Aristophanes Site Supporter Join Date: Jul 2008 Location: Rankin Inlet, Nunavut Photos: Albums Posts: 3,948	Originally posted by glanglois We need a new forum: Falk Lumo Reports. Saves your valuable time. And a PayPal button so we can donate a Pentax logo custom seat cover for his Audi.

03-11-2011, 01:44 AM	#334
Erik Veteran Member Join Date: Nov 2008 Location: Umeå, Sweden Posts: 755	Originally posted by Aristophanes And a PayPal button so we can donate a Pentax logo custom seat cover for his Audi. He drives an Audi? Guy's got taste

03-11-2011, 05:53 AM	#335
dlacouture Veteran Member Join Date: Jul 2008 Location: Var, South of France Photos: Albums Posts: 1,074	Originally posted by Erik He drives an Audi? Guy's got taste Guy's is German, IIRC, so Audi, BMW, or VW were pretty much mandatory Same thing for me : Peugeot and Renault (mmmmh, only this does not sound so cool )

03-11-2011, 12:29 PM	#336
NaClH2O Site Supporter Join Date: Sep 2006 Location: Bronx NY Photos: Gallery \| Albums Posts: 5,631	Originally posted by dlacouture Guy's is German, IIRC, so Audi, BMW, or VW were pretty much mandatory Same thing for me : Peugeot and Renault (mmmmh, only this does not sound so cool ) you could always go with Citroen, they ain't so bad NaCl(they do pretty good in the World Rally Championship)H2O

03-11-2011, 04:26 PM	#337
gazonk Pentaxian Join Date: Feb 2009 Location: Oslo area, Norway Photos: Gallery \| Albums Posts: 3,746	Originally posted by dlacouture Same thing for me : Peugeot and Renault (mmmmh, only this does not sound so cool ) Bah, I wouldn't trade our Peugeot for an Audi

03-12-2011, 02:25 AM	#338
dlacouture Veteran Member Join Date: Jul 2008 Location: Var, South of France Photos: Albums Posts: 1,074	Originally posted by NaClH2O you could always go with Citroen, they ain't so bad Man, I'd love a C6, those "hydractive" pneumatic shock absorbers are pure bliss...

03-13-2011, 05:34 PM	#339
falconeye Veteran Member Join Date: Jan 2008 Location: Munich, Alps, Germany Photos: Gallery Posts: 6,871 Original Poster	Originally posted by Class A Do you think it might perhaps be good to get some data with the old firmware regarding the potential peak of defocus around EV 8? I will not evaluate the EV6-8 region in this context. If the problem persists, it may become a separate topic. Currently, I live under the assumption that a mismatch of color at the AF point and globally, may have caused issues around that EV range. I'll include a look at this "color mismatch" in my updated paper, although at lower EV. I've already done all test shots and now need to run the analysis and write it up.

03-13-2011, 06:43 PM	#340
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Hi Falk, Originally posted by falconeye I will not evaluate the EV6-8 region in this context. Fair enough, I fully understand. I just thought I mention the problem before you update the firmware as there is probably no straightforward way of reverting to an older version (that doesn't involve hacking the old firmware to pretend it isn't older). Originally posted by falconeye I've already done all test shots and now need to run the analysis and write it up. Great! My gut feeling is that while some improvement has been achieved there is still a residual problem. So probably one, rather than two thumbs up.

03-13-2011, 06:46 PM	#341
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by NaClH2O you could always go with Citroen, they ain't so bad Citroëns are funtastic (sic).

03-17-2011, 02:13 PM	#342
mrmentera Forum Member Join Date: Nov 2009 Location: Falun Posts: 90	tomtomtomtomtomtomtomtomtomtomtom... as in Jaws..sort of...eh.

03-17-2011, 08:20 PM	#344
Duplo Veteran Member Join Date: Nov 2006 Location: Copenhagen Photos: Gallery Posts: 924	Originally posted by falconeye Hi friends, I am through the analysis of the v1.03 data and I am a bit confused. Maybe, somebody can help before I write my update. According to my measurements, v1.03 brings no significant progress IF the following conditions are met: 1. The FA 31 lens is used at 1m distance and f/1.8 aperture. 2. The background is bright. 3. The camera white balance is manually set to daylight (maybe doesn't matter). 4. The light color is tungsten (maybe doesn't matter). Wrt (3.), I see some positive effect if the camera white balance is set to automatic or tungsten. However, I don't have enough data points (I have 21 measurements for this case) to call this effect significant, i.e., these 21 data points all lie within the cloud of 94 data points obtained with daylight white balance. I even see some negative effect of v1.03 if the subject has a red color and the lamp has not. OTOH, I trust the numerous positive reports and I must assume that I must miss this effect for some reason. Maybe, the white balance effect is significant. Maybe, the effect is only noticeable with lenses slower than f/1.8. Or maybe, only at higher EV values with darker backgrounds. Unfortunately, I am far away from my lab and can't do further measurements for the next three weeks. Is there anybody wanting to check if the positive v1.03 effect disappears with a wide aperture and/or a non-matching camera white balance? Thanks everybody. first of thanks for all the hard work. I have got to admit to not having read you paper fully yet. Talking lenses in the F2.0 and below range, I have tested with the FA*24, FA43ltd and FA77, the 24 and77 were pretty accurate already before 1.0.3, but 1.0.3 has really improved on the 43, so now all of them are spot on, with only an occasional miss. Is it possible that the reason you are not seeing an improvement could be an issue with your specific lens/body combination?

03-17-2011, 08:21 PM	#345
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Hi Falks, thanks for the update. Originally posted by falconeye According to my measurements, v1.03 brings no significant progress IF the following conditions are met: Wow, that's bad news. Even in the conditions you list, AFAIC, there shouldn't be such a discontinuous jump in the choice of focus plane. Unless the colorimetric sensor really has such a sharp transition to blindness, a discontinuous behaviour would still point to a firmware bug (or a limitation of a correction table size). Is your set of criteria a positive list only? In other words, I guess that I cannot conclude from it that there is a significant improvement if the background is not bright, right? Originally posted by falconeye Wrt (3.), I see some positive effect if the camera white balance is set to automatic or tungsten. Oh dear, would be a shame if the K-5 were relegated to K-r tricks... (I know that this effect is not confirmed but it seems that this would be a reasonable step to take in cases where the colorimetric sensor isn't useful anymore). Originally posted by falconeye Maybe, the effect is only noticeable with lenses slower than f/1.8. Ray Pulley has posted test shots that show a significant change to the focus plane when using an FA 50/1.4 at EV 2 (<- Sekonic L-358 reading at focus target). To be honest, overall, that's quite disappointing news...