Yet another K-r AF test (with both 1.00 and 1.01)

3drives · 02-18-2011, 10:14 PM

Here you go. These are taken handheld, no timer delay. Forget the colors, my printer went broke on me today. Concentrate only on focus for today.

At 18mm and 55mm, I followed the focus chart test with 45 degree angle and results are here:

Attached Images

View Picture EXIF

View Picture EXIF

ursamajor · 02-18-2011, 11:00 PM

Can you tell me the serial of it? See on the bottom of the camera.

3drives · 02-18-2011, 11:19 PM

Originally posted by ursamajor

Can you tell me the serial of it? See on the bottom of the camera.

Starts with 379xxxx. what does that imply?

Hypocorism · 02-18-2011, 11:59 PM

Originally posted by 3drives

Starts with 379xxxx. what does that imply?

To you, Nothing. (Unless one might have a very naive knowledge and understanding, meaning zero internal work experience, of the real logistics behind manufacturing industry volume production processes and its International market product distribution.) Think about it.

And why would you want to "hide" your S/R with silly xxx's? Do you think someone might use it to validate their copy of Windows?

Think about it. Don't just repeat doing or believing things because you see other people say or do it.

Fwiw: Mine is 3894260 (pls let me know if that bold red text isn't bright enough to see), and it can't properly focus in tungsten light (no surprises there) if its bloody life depended on it - if that matters to you. See the non-correlation now?

Ps. When you think about and accept these things you might like to consider adding yours to the P.F Serial Number Database. And while it would be equally naive to deduce that it's automatically related to the above, the data collection does and will serve a meaningful purpose in ways that might not be obvious to some, so the contribution will be appreciated.

Have phun experimenting with that K-r focusing. The amusement makes a real fun read.

.R.

If I had six hours to chop down a tree, I'd spend the first four sharpening the axe. -- Abraham Lincoln (1809-1865)

Last edited by Hypocorism; 02-19-2011 at 10:29 AM.

3drives · 02-19-2011, 05:42 AM

Hypocorism, The serial numbers are critical for software upgrades, they also work as identifiers in some cases. Yes, others could use my number, not you though. Sorry to say, but I am not going to reveal the full serial number.

3drives · 02-19-2011, 06:41 AM

Ok, guys, somebody in the other thread said the issue is not with tungsten WB but it is with AWB in tungsten light.

screw you all for creating so much confusion.

Anyway, I repeated the focus chart test with same settings but in AWB now. I find there is FF. But, that damn chart test is stupid as the letters are so small, I am not able to figure whether I got eyesight or whether letters are OOF or what. Plus I was always wondering if I am holding the camera in 45 degree angle. With the table and tripod I have, I was not able to use the tripod.

So, enough of this nuisance, I tried the test in a different way but completely scientific.

So, what I did was - I selected 3 subjects, put them all close together but in different planes. Like one's plane in front of another's plane. In the tests, my focus was on the word vaseline in both images, mind you.

Pic 1: Without focus adjustment . Unfortunately, the camera focussed on Stax and not on vaseline. Confirms FF issue.

Pic 2: Focus adjustment to -10. Removes the problem 90% and satisfied but it is not still perfect.

Your input is always welcome.

ursamajor · 02-19-2011, 07:26 AM

The issue is "available" in tungsten light, no matter the selection of the WB, AWB or Tungsten.

Look here at my pictures:

Imageshack - imgp0322a.jpg
Imageshack - imgp0323x.jpg

and to a set of pictures taken by a friend of mine, who own the Kr too:

http://proxtu.ro/focus/

Let's all accept that camera has problem (not only mine and other 3-4 cameras, but all) and see if Pentax manage to solve the issue, an extremely upsetting one.

3drives · 02-19-2011, 10:14 AM

Taken in Flourescent lighting in my library with AWB setting. I focussed on the blue mug on letters "high risk" and the camera did exactly the same. Pretty! This is a cropped image.

All 3 subjects are in 3 different planes, one in front of another plane to see if camera was back or front focussing.

Uploaded with ImageShack.us

ursamajor · 02-19-2011, 10:25 AM

Does anybody said there is a problem in fluorescent?

krebsy75 · 02-19-2011, 10:25 AM

Why bother posting results from a focus test that was taken handheld?

A test is only valid if done correctly.

Originally posted by 3drives

Here you go. These are taken handheld, no timer delay. Forget the colors, my printer went broke on me today. Concentrate only on focus for today.

At 18mm and 55mm, I followed the focus chart test with 45 degree angle and results are here:

3drives · 02-19-2011, 10:30 AM

Originally posted by krebsy97

Why bother posting results from a focus test that was taken handheld?

Because not everytime do I use a tripod. I will do handheld shots in this conditions.

You should take into consideration that the object in front was focussing fine and perfect without blur. If there was camera shake in my handheld, and movement, the front object won't focus fine - the whole pic will show some blur. If nothing was in focus, and everything blurry, then your point is valid.

I hope you understood.

3drives · 02-19-2011, 06:26 PM

Has anyone tried this K-x solution on the K-r? RiceHigh's Pentax Blog: K-x Debug Mode Tutorial - AF Adjustment

ursamajor · 02-20-2011, 12:47 AM

Don't work on Kr. And, yes, there is on other thread this link.
Kx owners have used it (the adjustement from the debug menu) for SMALL front/back focus IN ALL TYPES OF LIGHT, not only in tungsten. That debug menu (they don't have AF fine tuning) it was a posibility to make their cameras to focus perfectly, despite the imperfections of the body-lens system. Try to understand that there is a big difference between moderate ff (adjustable) in all type of lights and specific SEVERE tungsten ff of the Kr/K5.

torpesco · 02-21-2011, 12:12 AM

Originally posted by ursamajor

The issue is "available" in tungsten light, no matter the selection of the WB, AWB or Tungsten.

Look here at my pictures:

Imageshack - imgp0322a.jpg
Imageshack - imgp0323x.jpg

First - I also have a FF problem, so I'm not trying to knock your claims.

However: You're focusing on the wrong part of the chart. You're supposed to focus where it says "Focus Here". The point is that focus sensors are actually a bit bigger than a tiny little dot in the middle. There should be nothing in the immediate vicinity that could potentially "steal" the focus.

Jeffrey Friedl's Blog: Jeffrey’s Autofocus Test Chart - "...ensure that the autofocus locks on to the exact location you intend..."

See the samples in the "Improperly Performed Test" section.

ursamajor · 02-21-2011, 12:32 AM

OK, maybe this test wasn't made corectly. But it's not the only one. I made a lot of tests with or without charts. I wouldn't have care about chart tests if real life tests would have been OK.
Anyway, it's not admissible (even in the conditions I made the focus in that chart test) to get so big FF.

02-19-2011, 06:41 AM	#186
3drives Forum Member Join Date: Feb 2011 Posts: 60	Ok, guys, somebody in the other thread said the issue is not with tungsten WB but it is with AWB in tungsten light. screw you all for creating so much confusion. Anyway, I repeated the focus chart test with same settings but in AWB now. I find there is FF. But, that damn chart test is stupid as the letters are so small, I am not able to figure whether I got eyesight or whether letters are OOF or what. Plus I was always wondering if I am holding the camera in 45 degree angle. With the table and tripod I have, I was not able to use the tripod. So, enough of this nuisance, I tried the test in a different way but completely scientific. So, what I did was - I selected 3 subjects, put them all close together but in different planes. Like one's plane in front of another's plane. In the tests, my focus was on the word vaseline in both images, mind you. Pic 1: Without focus adjustment . Unfortunately, the camera focussed on Stax and not on vaseline. Confirms FF issue. Pic 2: Focus adjustment to -10. Removes the problem 90% and satisfied but it is not still perfect. Your input is always welcome. Last edited by 3drives; 02-19-2011 at 07:30 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
People Test...Seeing if I still know how to do this...	donallison13	Photo Critique	6	10-30-2010 09:08 AM
K-x SR test	dude2009	Pentax DSLR Discussion	11	01-04-2010 10:38 AM
test	Oldphoto678	Post Your Photos!	13	02-18-2009 04:51 PM
Interesting DR Test and BG Noise Test of the K20D	RiceHigh	Pentax DSLR Discussion	14	06-25-2008 09:49 AM

02-18-2011, 11:00 PM	#182
ursamajor Banned Join Date: Jan 2011 Posts: 242	Can you tell me the serial of it? See on the bottom of the camera.

02-18-2011, 11:19 PM	#183
3drives Forum Member Join Date: Feb 2011 Posts: 60	Originally posted by ursamajor Can you tell me the serial of it? See on the bottom of the camera. Starts with 379xxxx. what does that imply?

02-18-2011, 11:59 PM	#184
Hypocorism Veteran Member Join Date: Sep 2010 Location: Melbourne .au Posts: 623	Originally posted by 3drives Starts with 379xxxx. what does that imply? To you, Nothing. (Unless one might have a very naive knowledge and understanding, meaning zero internal work experience, of the real logistics behind manufacturing industry volume production processes and its International market product distribution.) Think about it. And why would you want to "hide" your S/R with silly xxx's? Do you think someone might use it to validate their copy of Windows? Think about it. Don't just repeat doing or believing things because you see other people say or do it. Fwiw: Mine is 3894260 (pls let me know if that bold red text isn't bright enough to see), and it can't properly focus in tungsten light (no surprises there) if its bloody life depended on it - if that matters to you. See the non-correlation now? Ps. When you think about and accept these things you might like to consider adding yours to the P.F Serial Number Database. And while it would be equally naive to deduce that it's automatically related to the above, the data collection does and will serve a meaningful purpose in ways that might not be obvious to some, so the contribution will be appreciated. Have phun experimenting with that K-r focusing. The amusement makes a real fun read. .R. If I had six hours to chop down a tree, I'd spend the first four sharpening the axe. -- Abraham Lincoln (1809-1865) Last edited by Hypocorism; 02-19-2011 at 10:29 AM.

02-19-2011, 05:42 AM	#185
3drives Forum Member Join Date: Feb 2011 Posts: 60	Hypocorism, The serial numbers are critical for software upgrades, they also work as identifiers in some cases. Yes, others could use my number, not you though. Sorry to say, but I am not going to reveal the full serial number.

02-19-2011, 07:26 AM	#187
ursamajor Banned Join Date: Jan 2011 Posts: 242	The issue is "available" in tungsten light, no matter the selection of the WB, AWB or Tungsten. Look here at my pictures: Imageshack - imgp0322a.jpg Imageshack - imgp0323x.jpg and to a set of pictures taken by a friend of mine, who own the Kr too: http://proxtu.ro/focus/ Let's all accept that camera has problem (not only mine and other 3-4 cameras, but all) and see if Pentax manage to solve the issue, an extremely upsetting one.

02-19-2011, 10:14 AM	#188
3drives Forum Member Join Date: Feb 2011 Posts: 60	Taken in Flourescent lighting in my library with AWB setting. I focussed on the blue mug on letters "high risk" and the camera did exactly the same. Pretty! This is a cropped image. All 3 subjects are in 3 different planes, one in front of another plane to see if camera was back or front focussing. Uploaded with ImageShack.us

02-19-2011, 10:25 AM	#189
ursamajor Banned Join Date: Jan 2011 Posts: 242	Does anybody said there is a problem in fluorescent?

02-19-2011, 10:25 AM	#190
krebsy75 Veteran Member Join Date: Jun 2010 Location: Chester County, Pa. Posts: 804	Why bother posting results from a focus test that was taken handheld? A test is only valid if done correctly. Originally posted by 3drives Here you go. These are taken handheld, no timer delay. Forget the colors, my printer went broke on me today. Concentrate only on focus for today. At 18mm and 55mm, I followed the focus chart test with 45 degree angle and results are here:

02-19-2011, 10:30 AM	#191
3drives Forum Member Join Date: Feb 2011 Posts: 60	Originally posted by krebsy97 Why bother posting results from a focus test that was taken handheld? Because not everytime do I use a tripod. I will do handheld shots in this conditions. You should take into consideration that the object in front was focussing fine and perfect without blur. If there was camera shake in my handheld, and movement, the front object won't focus fine - the whole pic will show some blur. If nothing was in focus, and everything blurry, then your point is valid. I hope you understood.

02-19-2011, 06:26 PM	#192
3drives Forum Member Join Date: Feb 2011 Posts: 60	Has anyone tried this K-x solution on the K-r? RiceHigh's Pentax Blog: K-x Debug Mode Tutorial - AF Adjustment

02-20-2011, 12:47 AM	#193
ursamajor Banned Join Date: Jan 2011 Posts: 242	Don't work on Kr. And, yes, there is on other thread this link. Kx owners have used it (the adjustement from the debug menu) for SMALL front/back focus IN ALL TYPES OF LIGHT, not only in tungsten. That debug menu (they don't have AF fine tuning) it was a posibility to make their cameras to focus perfectly, despite the imperfections of the body-lens system. Try to understand that there is a big difference between moderate ff (adjustable) in all type of lights and specific SEVERE tungsten ff of the Kr/K5.

02-21-2011, 12:12 AM	#194
torpesco Forum Member Join Date: Oct 2010 Posts: 87	Originally posted by ursamajor The issue is "available" in tungsten light, no matter the selection of the WB, AWB or Tungsten. Look here at my pictures: Imageshack - imgp0322a.jpg Imageshack - imgp0323x.jpg First - I also have a FF problem, so I'm not trying to knock your claims. However: You're focusing on the wrong part of the chart. You're supposed to focus where it says "Focus Here". The point is that focus sensors are actually a bit bigger than a tiny little dot in the middle. There should be nothing in the immediate vicinity that could potentially "steal" the focus. Jeffrey Friedl's Blog: Jeffrey’s Autofocus Test Chart - "...ensure that the autofocus locks on to the exact location you intend..." See the samples in the "Improperly Performed Test" section.

02-21-2011, 12:32 AM	#195
ursamajor Banned Join Date: Jan 2011 Posts: 242	OK, maybe this test wasn't made corectly. But it's not the only one. I made a lot of tests with or without charts. I wouldn't have care about chart tests if real life tests would have been OK. Anyway, it's not admissible (even in the conditions I made the focus in that chart test) to get so big FF.