K-5 1.03 versus K20D AF tests

melander · 03-13-2011, 10:50 PM

I dont understand how your k20d can focus in anything below ev3, my k20d couldnt focus and hunted like crazy.

Smeggypants · 03-14-2011, 09:06 AM

Originally posted by Ray Pulley

Hi Smeggy,

I normally set my K20 AF adjust to a somewhat BF condition in daylight so that it will stay within the DOF in lower tungsten light, which causes a move to FF.

As long as the image stays at an acceptable focus as the light changes, I see nothing wrong with that. In fact, setting the camera to be perfectly centered in one lighting condition or another removes a significant portion of the available range of the AF system from bright daylight to dim tungsten.

It is the only possible work-around for the earlier cameras that slide to FF in low tungsten light other than changing the AF adjust every time you go from inside to outside or vice versa. Otherwise, you have to set it for the conditions you shoot in most often.

I suppose a balance of color temp and light level could be calculated somehow and the AF point set in the center under a balanced condition, but most of us have no easy way to do that.

The trick is to not set the daylight AF calibration point so deep that you get BF in more blue daylight conditions. Field shots to confirm are the best way to tell this, and if your K20 is holding acceptable focus from daylight to dim tungsten in the field, the BF cal is working for you.

On another note, if you have Photoshop, try this technique to easily see where the DOF begins and ends on your test shots (I took the liberty of borrowing one of your test shots to illustrate, I hope that you do not mind):

The steps are:

Open the image

Desaturate to B&W

On the Filter menu select "Stylize" and then "emboss"

Play with the angle to get the best level of embossing (this will depend upon where the main sources of light were when you shot it).

Leave the height at 3 pixels and adjust the amount to make the lines look as clear as possible.

Adjust contrast if the image was a bit soft (and the FA 50 @ f1.4 is always a bit soft!).

EDITED TO ADD: This technique is obviously not at all needed on the Sigma test you posted above

Ray

Cheers Ray.

I don't use Photoshop, I Use LIghtroom+Photopaint, but photopaint does embossing. I'll remember that technique if and when I do some moreaccurate analysis in the future.

I've messaged my dealer now regarding replacing my K-5 and asked how long it would take. I'm planning on buying a 2nd K-5 in the next few weeks and it would be ideal to get that first so I'm not without one.

woof · 03-14-2011, 10:05 AM

Originally posted by Smeggypants

Hi there. Please see this post for info. You may have missed it.

https://www.pentaxforums.com/forums/pentax-k-5-forum/136324-k-5-1-03-versus-k...ml#post1421891

Um... that's exactly the test he has qualms about. :ugh:

Add me to the list of those noting variability in the tests.

Too much variability for my taste.

This is not an indictment of you. It IS the very reason I bought into the LensAlign system.

That said, I don't doubt you have issues. I am just in favor of more rigorous testing. I am waiting for Falk's results frankly.

woof!

sagiboy · 03-14-2011, 10:31 AM

Maybe a silly question...would dialing in +- EV affect the focussing??? I never thought of it like that???

ursamajor · 03-14-2011, 10:44 AM

Read HERE or HERE and you'll understand the concept. It's not about the exposure compensation, but about the quantity of light for a corect exposure, shortly explained.

Smeggypants · 03-14-2011, 11:39 AM

Originally posted by woof

Um... that's exactly the test he has qualms about. :ugh:

Add me to the list of those noting variability in the tests.

Too much variability for my taste.

This is not an indictment of you. It IS the very reason I bought into the LensAlign system.

That said, I don't doubt you have issues. I am just in favor of more rigorous testing. I am waiting for Falk's results frankly.

woof!

Oh I love rigorous testing when required.

My test here was enough to show by K-5's FF fault and that's all that matters fro this test.

And yup I'm looking forward to Falk's 1.03 tests. I'm already preparing to replace my my K-5 and buy a 2nd K-5 though.

Rin · 03-14-2011, 03:16 PM

unfortanely i still noticed some FF in certain conditions, it certainly has to do something with ellowish light, but i did not notice this before with firmware 1.03, so i am kinda thinking this problem occurred later.

Class A · 03-14-2011, 03:40 PM

Originally posted by Ray Pulley

I normally set my K20 AF adjust to a somewhat BF condition in daylight so that it will stay within the DOF in lower tungsten light, which causes a move to FF.

That's exactly what Smeggy did (intentionally or not). His K20D front focuses under Tungsten light, like every other K20 but due to the backfocus bias he dialed in, the front-focusing still leaves the target within the DOF.

Originally posted by Ray Pulley

On another note, if you have Photoshop, try this technique to easily see where the DOF begins and ends on your test shots ...

I suggested a similar technique (just based on oversharpening instead of embossing) in a previous post. I think the oversharpening halos work just as well.

Looking at the K-5 shots again, I noticed there are more than one shot suggesting FF for the K-5 under low light. Not knowing how the light levels were obtained, what care was taken in focusing, etc. it is difficult to assign a lot of weight to this. Assuming that the EV values were not calibrated by looking at the actual image brightness at the AF spot, they would actually have to be lower by roughly 2 EV (due to a missing 18% gray calibration). Sure, the tests might reveal a problem, they are just not as convincing as they could have been.

**Ray Pulley** · 03-14-2011, 07:41 PM

Originally posted by Class A

That's exactly what Smeggy did (intentionally or not). His K20D front focuses under Tungsten light, like every other K20 but due to the backfocus bias he dialed in, the front-focusing still leaves the target within the DOF.

I suggested a similar technique (just based on oversharpening instead of embossing) in a previous post. I think the oversharpening halos work just as well.

Looking at the K-5 shots again, I noticed there are more than one shot suggesting FF for the K-5 under low light. Not knowing how the light levels were obtained, what care was taken in focusing, etc. it is difficult to assign a lot of weight to this. Assuming that the EV values were not calibrated by looking at the actual image brightness at the AF spot, they would actually have to be lower by roughly 2 EV (due to a missing 18% gray calibration). Sure, the tests might reveal a problem, they are just not as convincing as they could have been.

I have a K5 due this week, which gives me a trifecta of: K20, K7, K5 to compare.

I also have a reasonable test setup that will allow me to check in Tungsten lighting from about EV2 to EV 10 or 11, and a pair of 5500k CFL lights for some tests in lighting that is less red.

I have a Sekonic L-358 to measure the EV level at the target and a color meter to measure the color at the target as well.

My setup is repeatible and accurate enough without going overboard and getting into real lab conditions and optical benches and so on. At the end of the day my real test is to use the camera indoors in normal lighting I encounter all the time and also outdoors in a similar fashion wherer the K20 has been ok for me for several years now.

FWIW, I also have no interest in trying and adjust for what "might" be seen by the various sensors in the camera, nor do I really care about tweaking EV readings to compensate for the lack of blue and green or whatever. I have the sort of tools that are commonly available and used in various forms of photography such as handheld meters and halogen modeling lights on my studio flashes, as well as a decent target that can be properly aligned and which also has a flat surface to focus upon.

I already know that my K20 will not hold DOF below about f4.0 in tungsten lighting at EV2. I also know that the specification for the K20 is lower than this. As I mentioned above, I have a way of calibrating the K20 to make the most of the available DOF, and in most cases I shoot bounce flash at f5.6 in this sort of light which has given me decent results with the K20. If the K5 can match this, I will be happy.

Hopefully the K5 will show up with 1.02 installed so I can run a direct 1.02 vs 1.03 comparison.

I will post some test images once the K5 arrives.

Ray

jaad75 · 03-14-2011, 09:00 PM - **1 Like**

Ok, so I have a little proof that at least some K-5s can focus properly at -1EV tungsten light:

f/1.4, 0.3s, ISO800 (pushed +1EV in Lightroom), AF assist light off

18 shots, defocus after each shot, 5 fails (4xFF, 1xBF), 7 perfects, 6 acceptable. Not bad in my book...
And no, my old K10D can't get anywhere close at this light level...

The only problem is pronounced FF with small light objects on dark/black background even at 1.5EV tungsten, which is some kind of bug I guess, but since I rarely shoot at such low light, I'm not too bothered... That's something to remember just in case.

Catalana · 03-14-2011, 09:09 PM

Brilliant...

. There can be no doubt as to EV.

Smeggypants · 03-14-2011, 11:02 PM

Apart from the the Low Light Front focus fault my K-5 can take pics. Art isn't bothered about Field Curvature

Gimbal · 03-15-2011, 12:32 AM

Originally posted by jaad75

Ok, so I have a little proof that at least some K-5s can focus properly at -1EV tungsten light:

f/1.4, 0.3s, ISO800 (pushed +1EV in Lightroom), AF assist light off

18 shots, defocus after each shot, 5 fails (4xFF, 1xBF), 7 perfects, 6 acceptable. Not bad in my book...
And no, my old K10D can't get anywhere close at this light level...

The only problem is pronounced FF with small light objects on dark/black background even at 1.5EV tungsten, which is some kind of bug I guess, but since I rarely shoot at such low light, I'm not too bothered... That's something to remember just in case.

Nice results, but how about wider focal lengths? The problem seems to escalate with wider focal lengths, at least for some of us. Mine is OK at 70mm but way out at 17mm. It got a lot better with 1.03 but it is still there, only at slightly lower EV.

Tommot1965 · 03-15-2011, 01:18 AM

Originally posted by Gimbal

Nice results, but how about wider focal lengths? The problem seems to escalate with wider focal lengths, at least for some of us. Mine is OK at 70mm but way out at 17mm. It got a lot better with 1.03 but it is still there, only at slightly lower EV.

I wonder why that is..Mines the same..perfect from about 35mm on, i tested my 70-200 Sigma last night. that was perfect too...it seems the wider you go the worse it is..

but at least I know that..so I stay away from the wide end of my sigma 17-50 if in low light.

if Im close 300mm away..even at 17 mm its ok...seem the further you get from the focal plane the worse it is too...very strange

so even though mines not good at very wide FL...Im considering it good enough for my needs..perhaps future FW fix will sort it for us

Nico13 · 03-15-2011, 03:16 AM

Originally posted by bogiesbad

These tests mean very little considering you don't have the charts anywhere near parallel to the sensor plane and some are not centered. Failed to convince me.

My sentiments also very poorly aligned charts.
Throw this test out and start again with everything properly aligned and you may get some meaningful results.

03-14-2011, 10:31 AM	#79
sagiboy Senior Member Join Date: Aug 2009 Location: LI Posts: 230	Why do we have to dial in the EV for?? Maybe a silly question...would dialing in +- EV affect the focussing??? I never thought of it like that???

03-14-2011, 07:41 PM	#84
Ray Pulley Site Supporter Join Date: Jan 2007 Location: Fallon Nevada Photos: Albums Posts: 504	K5 Incoming Originally posted by Class A That's exactly what Smeggy did (intentionally or not). His K20D front focuses under Tungsten light, like every other K20 but due to the backfocus bias he dialed in, the front-focusing still leaves the target within the DOF. I suggested a similar technique (just based on oversharpening instead of embossing) in a previous post. I think the oversharpening halos work just as well. Looking at the K-5 shots again, I noticed there are more than one shot suggesting FF for the K-5 under low light. Not knowing how the light levels were obtained, what care was taken in focusing, etc. it is difficult to assign a lot of weight to this. Assuming that the EV values were not calibrated by looking at the actual image brightness at the AF spot, they would actually have to be lower by roughly 2 EV (due to a missing 18% gray calibration). Sure, the tests might reveal a problem, they are just not as convincing as they could have been. I have a K5 due this week, which gives me a trifecta of: K20, K7, K5 to compare. I also have a reasonable test setup that will allow me to check in Tungsten lighting from about EV2 to EV 10 or 11, and a pair of 5500k CFL lights for some tests in lighting that is less red. I have a Sekonic L-358 to measure the EV level at the target and a color meter to measure the color at the target as well. My setup is repeatible and accurate enough without going overboard and getting into real lab conditions and optical benches and so on. At the end of the day my real test is to use the camera indoors in normal lighting I encounter all the time and also outdoors in a similar fashion wherer the K20 has been ok for me for several years now. FWIW, I also have no interest in trying and adjust for what "might" be seen by the various sensors in the camera, nor do I really care about tweaking EV readings to compensate for the lack of blue and green or whatever. I have the sort of tools that are commonly available and used in various forms of photography such as handheld meters and halogen modeling lights on my studio flashes, as well as a decent target that can be properly aligned and which also has a flat surface to focus upon. I already know that my K20 will not hold DOF below about f4.0 in tungsten lighting at EV2. I also know that the specification for the K20 is lower than this. As I mentioned above, I have a way of calibrating the K20 to make the most of the available DOF, and in most cases I shoot bounce flash at f5.6 in this sort of light which has given me decent results with the K20. If the K5 can match this, I will be happy. Hopefully the K5 will show up with 1.02 installed so I can run a direct 1.02 vs 1.03 comparison. I will post some test images once the K5 arrives. Ray

03-14-2011, 09:09 PM	#86
Catalana Senior Member Join Date: Feb 2010 Location: California Posts: 190	Brilliant.... There can be no doubt as to EV.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K20D versus K10D	Mapleleaf-Mick	Pentax DSLR Discussion	13	12-24-2010 10:58 AM
gx20 versus k20d!	kositoes	Pentax DSLR Discussion	22	09-25-2009 03:00 PM
Tests to determine if a new K20D is faulty	eman	Pentax DSLR Discussion	21	07-05-2008 10:30 PM
gx-20 versus k20d anyone?	davieFL	Photographic Technique	1	06-18-2008 12:23 PM
Some K20D Vs K10D ISO noise tests (100 to 6400)	RiceHigh	Pentax DSLR Discussion	15	02-25-2008 04:00 PM

03-13-2011, 10:50 PM	#76
melander Veteran Member Join Date: Jul 2008 Location: Sweden, Umea Posts: 876	I dont understand how your k20d can focus in anything below ev3, my k20d couldnt focus and hunted like crazy.

03-14-2011, 09:06 AM	#77
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536 Original Poster	Originally posted by Ray Pulley Hi Smeggy, I normally set my K20 AF adjust to a somewhat BF condition in daylight so that it will stay within the DOF in lower tungsten light, which causes a move to FF. As long as the image stays at an acceptable focus as the light changes, I see nothing wrong with that. In fact, setting the camera to be perfectly centered in one lighting condition or another removes a significant portion of the available range of the AF system from bright daylight to dim tungsten. It is the only possible work-around for the earlier cameras that slide to FF in low tungsten light other than changing the AF adjust every time you go from inside to outside or vice versa. Otherwise, you have to set it for the conditions you shoot in most often. I suppose a balance of color temp and light level could be calculated somehow and the AF point set in the center under a balanced condition, but most of us have no easy way to do that. The trick is to not set the daylight AF calibration point so deep that you get BF in more blue daylight conditions. Field shots to confirm are the best way to tell this, and if your K20 is holding acceptable focus from daylight to dim tungsten in the field, the BF cal is working for you. On another note, if you have Photoshop, try this technique to easily see where the DOF begins and ends on your test shots (I took the liberty of borrowing one of your test shots to illustrate, I hope that you do not mind): The steps are: Open the image Desaturate to B&W On the Filter menu select "Stylize" and then "emboss" Play with the angle to get the best level of embossing (this will depend upon where the main sources of light were when you shot it). Leave the height at 3 pixels and adjust the amount to make the lines look as clear as possible. Adjust contrast if the image was a bit soft (and the FA 50 @ f1.4 is always a bit soft!). EDITED TO ADD: This technique is obviously not at all needed on the Sigma test you posted above Ray Cheers Ray. I don't use Photoshop, I Use LIghtroom+Photopaint, but photopaint does embossing. I'll remember that technique if and when I do some moreaccurate analysis in the future. I've messaged my dealer now regarding replacing my K-5 and asked how long it would take. I'm planning on buying a 2nd K-5 in the next few weeks and it would be ideal to get that first so I'm not without one.

03-14-2011, 10:05 AM	#78
woof Inactive Account Join Date: Nov 2007 Location: In the present Photos: Gallery Posts: 1,870	Originally posted by Smeggypants Hi there. Please see this post for info. You may have missed it. https://www.pentaxforums.com/forums/pentax-k-5-forum/136324-k-5-1-03-versus-k...ml#post1421891 Um... that's exactly the test he has qualms about. :ugh: Add me to the list of those noting variability in the tests. Too much variability for my taste. This is not an indictment of you. It IS the very reason I bought into the LensAlign system. That said, I don't doubt you have issues. I am just in favor of more rigorous testing. I am waiting for Falk's results frankly. woof!

03-14-2011, 10:44 AM	#80
ursamajor Banned Join Date: Jan 2011 Posts: 242	Read HERE or HERE and you'll understand the concept. It's not about the exposure compensation, but about the quantity of light for a corect exposure, shortly explained.

03-14-2011, 11:39 AM	#81
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536 Original Poster	Originally posted by woof Um... that's exactly the test he has qualms about. :ugh: Add me to the list of those noting variability in the tests. Too much variability for my taste. This is not an indictment of you. It IS the very reason I bought into the LensAlign system. That said, I don't doubt you have issues. I am just in favor of more rigorous testing. I am waiting for Falk's results frankly. woof! Oh I love rigorous testing when required. My test here was enough to show by K-5's FF fault and that's all that matters fro this test. And yup I'm looking forward to Falk's 1.03 tests. I'm already preparing to replace my my K-5 and buy a 2nd K-5 though.

03-14-2011, 03:16 PM	#82
Rin Forum Member Join Date: Nov 2010 Posts: 87	unfortanely i still noticed some FF in certain conditions, it certainly has to do something with ellowish light, but i did not notice this before with firmware 1.03, so i am kinda thinking this problem occurred later.

03-14-2011, 03:40 PM	#83
Class A Pentaxian Join Date: Aug 2008 Location: Wellington, New Zealand Posts: 11,251	Originally posted by Ray Pulley I normally set my K20 AF adjust to a somewhat BF condition in daylight so that it will stay within the DOF in lower tungsten light, which causes a move to FF. That's exactly what Smeggy did (intentionally or not). His K20D front focuses under Tungsten light, like every other K20 but due to the backfocus bias he dialed in, the front-focusing still leaves the target within the DOF. Originally posted by Ray Pulley On another note, if you have Photoshop, try this technique to easily see where the DOF begins and ends on your test shots ... I suggested a similar technique (just based on oversharpening instead of embossing) in a previous post. I think the oversharpening halos work just as well. Looking at the K-5 shots again, I noticed there are more than one shot suggesting FF for the K-5 under low light. Not knowing how the light levels were obtained, what care was taken in focusing, etc. it is difficult to assign a lot of weight to this. Assuming that the EV values were not calibrated by looking at the actual image brightness at the AF spot, they would actually have to be lower by roughly 2 EV (due to a missing 18% gray calibration). Sure, the tests might reveal a problem, they are just not as convincing as they could have been.

03-14-2011, 09:00 PM - 1 Like	#85
jaad75 Veteran Member Join Date: Sep 2007 Location: Cracow Posts: 457	Ok, so I have a little proof that at least some K-5s can focus properly at -1EV tungsten light: f/1.4, 0.3s, ISO800 (pushed +1EV in Lightroom), AF assist light off 18 shots, defocus after each shot, 5 fails (4xFF, 1xBF), 7 perfects, 6 acceptable. Not bad in my book... And no, my old K10D can't get anywhere close at this light level... The only problem is pronounced FF with small light objects on dark/black background even at 1.5EV tungsten, which is some kind of bug I guess, but since I rarely shoot at such low light, I'm not too bothered... That's something to remember just in case.

03-14-2011, 11:02 PM	#87
Smeggypants Veteran Member Join Date: Sep 2010 Posts: 1,536 Original Poster	Apart from the the Low Light Front focus fault my K-5 can take pics. Art isn't bothered about Field Curvature

03-15-2011, 12:32 AM	#88
Gimbal Pentaxian Join Date: Apr 2007 Location: Sweden Posts: 2,106	Originally posted by jaad75 Ok, so I have a little proof that at least some K-5s can focus properly at -1EV tungsten light: f/1.4, 0.3s, ISO800 (pushed +1EV in Lightroom), AF assist light off 18 shots, defocus after each shot, 5 fails (4xFF, 1xBF), 7 perfects, 6 acceptable. Not bad in my book... And no, my old K10D can't get anywhere close at this light level... The only problem is pronounced FF with small light objects on dark/black background even at 1.5EV tungsten, which is some kind of bug I guess, but since I rarely shoot at such low light, I'm not too bothered... That's something to remember just in case. Nice results, but how about wider focal lengths? The problem seems to escalate with wider focal lengths, at least for some of us. Mine is OK at 70mm but way out at 17mm. It got a lot better with 1.03 but it is still there, only at slightly lower EV.

Thread Tools	Search this Thread
	Search this Thread: Advanced Search

03-15-2011, 01:18 AM	#89
Tommot1965 Veteran Member Join Date: Sep 2010 Location: Perth Western Australia Photos: Gallery \| Albums Posts: 1,026	Originally posted by Gimbal Nice results, but how about wider focal lengths? The problem seems to escalate with wider focal lengths, at least for some of us. Mine is OK at 70mm but way out at 17mm. It got a lot better with 1.03 but it is still there, only at slightly lower EV. I wonder why that is..Mines the same..perfect from about 35mm on, i tested my 70-200 Sigma last night. that was perfect too...it seems the wider you go the worse it is.. but at least I know that..so I stay away from the wide end of my sigma 17-50 if in low light. if Im close 300mm away..even at 17 mm its ok...seem the further you get from the focal plane the worse it is too...very strange so even though mines not good at very wide FL...Im considering it good enough for my needs..perhaps future FW fix will sort it for us

03-15-2011, 03:16 AM	#90
Nico13 Forum Member Join Date: Apr 2008 Location: Australia/Victoria Photos: Gallery Posts: 99	Originally posted by bogiesbad These tests mean very little considering you don't have the charts anywhere near parallel to the sensor plane and some are not centered. Failed to convince me. My sentiments also very poorly aligned charts. Throw this test out and start again with everything properly aligned and you may get some meaningful results.