Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
02-11-2011, 04:37 AM   #61
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
Original Poster
LED, DFA 100mm 1:2.8 f=2.8

Direct light from 2W Philips LED bulb replacement ('MyAccent') at right to camera, ~45deg angle from lens axis, distance varied to adjust intensity, DFA 100mm 1:2.8, f=2.8, target distance to sensor plane 1m, assuming camera metering (spot) correct the EVs given should be accurate within 1/3 EV

#1 : EV 4 (t=0.5s) : focus OK : camera wb 3104K
#2 : EV 3 (t=1.0s) : focus OK : camera wb 3075K
#3 : EV 2 (t=2.0s) : focus OK : camera wb 3088K


Last edited by jolepp; 03-03-2017 at 02:40 AM.
02-11-2011, 04:55 AM   #62
Veteran Member
Tommot1965's Avatar

Join Date: Sep 2010
Location: Perth Western Australia
Photos: Gallery | Albums
Posts: 1,026
good test results Jolepp...shows Tungsten is a problem .....my lights are halogen...as you know a form of tungsten lighting..
02-11-2011, 05:32 AM   #63
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
Original Poster
QuoteOriginally posted by Tommot1965 Quote
good test results Jolepp...shows Tungsten is a problem .....my lights are halogen...as you know a form of tungsten lighting..
AFAIK they are souped up tungsten where a bit of halogen in the bulb atmosphere helps keep the filament from depositing itself on the inside glass surface despite its higher temperature and consequently higher efficiency in producing visible light. Plenty of IR with that too, though.
02-11-2011, 06:15 AM   #64
Site Supporter
firefly's Avatar

Join Date: Mar 2008
Location: New Brunswick
Photos: Albums
Posts: 200
So maybe you, the users have the fix - replace all your inefficient, IR-laced tungsten bulbs with energy efficient CFL or LED... Pentax just wants you to go green, folks, c'mon!

02-11-2011, 05:50 PM   #65
Junior Member




Join Date: Jan 2009
Location: Berlin, DE
Posts: 30
Which makes me think... will an IR-cut filter solve the FF issue once and forever? I recall Leica having an issue with weak IR filter on the sensor, to a degree they provided buyers with free IR filters for their lenses. Could it be the case with Pentax?
02-12-2011, 05:46 AM   #66
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
Original Poster
QuoteOriginally posted by aoleg Quote
Which makes me think... will an IR-cut filter solve the FF issue once and forever? I recall Leica having an issue with weak IR filter on the sensor, to a degree they provided buyers with free IR filters for their lenses. Could it be the case with Pentax?
Occurred to me to, it took a couple of days to get my mitts on such a filter (see below ).
02-12-2011, 05:50 AM   #67
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
Original Poster
tungsten, DFA 100mm 1:2.8 f=2.8 + IR block fiter

These have been taken in direct light from a 40W tungsten bulb with the DFA 100mm 1:2.8 at EV 3 (f=2.8 t=1s) @1m as above, but without/with a Tiffen "77MM Standard Hot Mirror" (77SHM EAN 0 49383 15325 5); this is supposed to block (most) IR, see: "sell sheet" + attachment #3 from that.

#1 : without filter : camera wb 3266K
#2 : with filter : camera wb 3233K

In comparison #2 is less front focused although not spot-on. Still, I think it is fair to conclude that my K-5's AF is affected by the ratio IR/visible light. Whether this is the sole explanation for the different results in tungsten / CFL / LED light is unclear, though.

Last edited by jolepp; 03-03-2017 at 02:40 AM.
02-12-2011, 06:27 AM   #68
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
Original Poster
A comparison without (#1) and with (#2) the Tiffen hot mirror: DFA 100mm 1:2.8 @1m, daylight (EV 8.6 : f=2.8 t=1/50s) from a window. After having autofocus to operate by a half-press of the shutter button I moved the body selector to MF, took #1, put the filter on and took #2. The point of these is to give an idea of effect of the filter as such: not much. (Also, assuming I managed not to disturb the focus setting this should demonstrate that AF works just fine in these conditions ).


Last edited by jolepp; 03-03-2017 at 02:40 AM.
02-12-2011, 10:33 AM   #69
Veteran Member




Join Date: Jan 2007
Location: Northern California
Posts: 475
Not Quite

QuoteOriginally posted by jolepp Quote
These have been taken in direct light from a 40W tungsten bulb with the DFA 100mm 1:2.8 at EV 3 (f=2.8 t=1s) @1m as above, but without/with a Tiffen "77MM Standard Hot Mirror" (77SHM EAN 0 49383 15325 5); this is supposed to block (most) IR, see: "sell sheet" + attachment #3 from that.

#1 : without filter : camera wb 3266K
#2 : with filter : camera wb 3233K

In comparison #2 is less front focused although not spot-on. Still, I think it is fair to conclude that my K-5's AF is affected by the ratio IR/visible light. Whether this is the sole explanation for the different results in tungsten / CFL / LED light is unclear, though.
This theory has been investigated before, IIRC, going back as far as the K10 when it had focusing issues.

I do believe that the PDAF systems of most cameras have some shift due to color temperature as otherwise why would Pentax and Nikon and Canon all incorporate color sensing elements in the AF system? I also wonder why, then, would they not simply place an IR cut filter in the AF light path just like they do in front of the main sensor?

That being said, this test does not rule out a similar shift caused simply by putting any filter on the lens. I have seen similar focus shifts caused by any number of filters, including simple skylight, polarizers, ND and so on.

Ray
02-20-2011, 03:39 PM   #70
Veteran Member
Kenn100D's Avatar

Join Date: Feb 2007
Location: Paranaque City, Philippines
Photos: Gallery
Posts: 646
Maybe the phase detection AF of Pentax still needs tweaking. Hope they add to firmware for to use contrast AF in TTL just like LV.
02-24-2011, 01:56 PM   #71
Inactive Account




Join Date: May 2007
Location: Central NJ
Posts: 469
QuoteOriginally posted by Kenn100D Quote
Maybe the phase detection AF of Pentax still needs tweaking. Hope they add to firmware for to use contrast AF in TTL just like LV.
How would you use contrast AF with the mirror down? i mean contrast focus with LV IS TTL. THe light path goes through the lens.. to the sensor and onto the rear display. To get it to the viewfinder you need a mirror in front of the sensor, so...
02-24-2011, 06:32 PM   #72
Veteran Member
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,863
QuoteOriginally posted by raz-0 Quote
How would you use contrast AF with the mirror down? i mean contrast focus with LV IS TTL. THe light path goes through the lens.. to the sensor and onto the rear display. To get it to the viewfinder you need a mirror in front of the sensor, so...
Theoretically, you could use a full (albeit smaller) sensor in the bottom of the mirror box (i.e. an AF module) which does contrast AF. It would be more accurate but slower. But would keep working while using the optical VF.

Most think that would be an inferior solution. I don't. It's just that contrast AF algorithms are still at their infancy due to a lack of genius of the engineers in charge.

Of course, that wouldn't be a firmwarw upgrade for sure
03-03-2011, 08:22 PM   #73
Pentaxian




Join Date: Mar 2009
Location: Perth Australia
Photos: Albums
Posts: 1,511
contrast af got a lot faster with the K-5. Its also almost always a superior result to phase. Which is so prone to variation.
03-12-2011, 04:34 AM   #74
Veteran Member
jolepp's Avatar

Join Date: Oct 2010
Location: Finland
Photos: Albums
Posts: 3,196
Original Poster
Fw 1.03

Enclosed is a series with the DFA 100mm 1:2.8 @1m f=2.8 in direct light from a 25W tungsten bulb (right of camera, ~-30-40 deg angle, distance varied to control light intensity), target from post #43 folded for 1" difference in depth from right or left to middle section.

#1 : EV 4 (t=0.5s) : camera wb 3371K
#2 : EV 3 (t=1.0s) : camera wb 3308K
#3 : EV 2 (t=2.0s) : camera wb 3266K
#4 : EV 1 (t=4.0s) : camera wb 3278K

Conclusion: the new firmware was a marked improvement for me, with 1.02 FF set in around EV 4-3, now this seems to happen around EV 2-1. Crucially this means that AF should now work with something like f=2.8 t=1/30s ISO=3200 (EV 3) even if there happens to be a tungsten bulb around .

(Ufraw still seems to corrupt some .DNGs with the vertical stripe on the right, this should not matter for the purpose at hand though).

Last edited by jolepp; 03-03-2017 at 02:40 AM.
03-12-2011, 05:51 AM   #75
Veteran Member
Tommot1965's Avatar

Join Date: Sep 2010
Location: Perth Western Australia
Photos: Gallery | Albums
Posts: 1,026
nice one Jolepp..thanks for doing the test mate..shows that 1.03 is a success for you...great stuff
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!
af, camera, conditions, dslr, k-5, k-5 ii, k-5 iis, k5, pentax k-5, results, test, tests
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
645D lens test results Tim Wilson Pentax Medium Format 20 01-12-2011 08:18 AM
Wireless P-TTL test results adamfogerty Pentax Camera and Field Accessories 3 06-20-2010 05:18 PM
Test results: Pentax SV / SuperTak 55 1.8 ismaelg Pentax Film SLR Discussion 2 07-31-2009 05:19 PM
Pentax 50mm F2 Test results eccs19 Pentax SLR Lens Discussion 6 08-29-2008 12:42 PM
Need help evaluating results from focus test pdmlmember Pentax DSLR Discussion 19 09-11-2007 02:09 AM



All times are GMT -7. The time now is 07:36 AM. | See also: NikonForums.com, CanonForums.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