k5 afc Case Study

telfish · 10-22-2010, 06:29 AM

Originally posted by imtheguy

Its hard to draw any conclusions from the first half of the shots as I look at the trees it does not seem to change focus very often. The closer shots look like its working! I am trying to do a similar test with a sea gull flying towards me but no complete results yet.

Were you holding the shutter button fully down the entire time in 7fps mode or working from the half-pressed mode and clicking off individual shots quickly?

I kept my finger pressed so it was taking shots as fast as it could camera set on continuous hi mode. I am uploading the full size files this morning so people can zoom into the farther shots. 800x600 shots are difficult at distance.

Terry

rudemood11 · 10-22-2010, 06:30 AM

Which version of Sigma lens is that? Screw drive one or the newer HSM II?

telfish · 10-22-2010, 06:31 AM

Originally posted by rudemood11

Which version of Sigma lens is that? Screw drive one or the newer HSM II?

It's the first HSM version.

Terry

Conan · 10-22-2010, 06:57 AM

AFC test on moving cars is not a challenge. I used to use my K200D's AFC in a drifting event and it was quite easy since the target is so big.

JamieP · 10-22-2010, 07:04 AM

Sorry if I missed this but, were you shooting in focus or speed priority?

kenyee · 10-22-2010, 07:05 AM

If you have a little league baseball match nearby, you might want to plant yourself near first base and fire up when the kid runs toward the base.
I still remember a Canon 1dmkIII user complaining that only 60% of the shots were in focus when they all were in focus w/ his 1DmkII (this was Canon's mirror box misalignment problem that caused them to do a recall)...

telfish · 10-22-2010, 07:15 AM

Originally posted by JamieP

Sorry if I missed this but, were you shooting in focus or speed priority?

Focus priority.

I am not sure I used the optimum camera settings for this test, perhaps others should chime in and I will repeat it. Should I have stopped down to increase DOF. Should I have used only center AF point to prevent the camera focusing on trees as I panned?

I wanted to make it as difficult as I could. Full size files are uploading and I will give the address to see them soon.

Terry

telfish · 10-22-2010, 07:16 AM

Originally posted by kenyee

If you have a little league baseball match nearby, you might want to plant yourself near first base and fire up when the kid runs toward the base.
I still remember a Canon 1dmkIII user complaining that only 60% of the shots were in focus when they all were in focus w/ his 1DmkII (this was Canon's mirror box misalignment problem that caused them to do a recall)...

Will do, and I will post the results. Let me know what settings you prefer for the test.

telfish · 10-22-2010, 08:02 AM

Full size files can be seen here. Index of /telfish Many thanks to Priyantha Bleeker for the web hosting

From the shots the camera did not fully lock until about the third frame. I am thinking that was because the car was in deep shadow and the 5 point af locked on the transition between the sunlit and shadowed road surface. Fast moving stuff is not my forte I usually do landscapes and macro so please excuse any lack of technique in these shots.

From then on it found focus OK. Not an ideal test and I will do it again at a child’s baseball game.

**jpzk** · 10-22-2010, 08:59 AM

Originally posted by telfish

Full size files can be seen here. Index of /telfish Many thanks to Priyantha Bleeker for the web hosting

From the shots the camera did not fully lock until about the third frame. I am thinking that was because the car was in deep shadow and the 5 point af locked on the transition between the sunlit and shadowed road surface. Fast moving stuff is not my forte I usually do landscapes and macro so please excuse any lack of technique in these shots.

From then on it found focus OK. Not an ideal test and I will do it again at a child’s baseball game.

Telfish,

Thanks a bunch for the sample AF shots.

I think that what some people were saying in that a car really is not the best subject for AF-C tests is that it is large, usually quite contrasty and easier to follow than a child running to base in baseball!

I don't own the K5 (yet) [I have the K7 and K20D] so I can't vouch for your testing but I believe that trying on running subject TOWARD you would be better.
Also, the smaller the subject the more difficult it is for the AF to "track", so maybe if you are close enough to some birds (gulls?) you could give it a try?

Also, so you know, I started a thread on K5 AF testing here:
https://www.pentaxforums.com/forums/pentax-dslr-discussion/118553-k5-auto-focus-tests.html
and I thought perhaps it would be a good idea to post some of your pics there as well, as the risk of being redundant perhaps, but with different sample shots?

Anyway, great work you are doing for all of us who will buy the K5 soon.

Cheers!

JP

telfish · 10-22-2010, 09:07 AM

Originally posted by jpzk

Telfish,

Thanks a bunch for the sample AF shots.

I think that what some people were saying in that a car really is not the best subject for AF-C tests is that it is large, usually quite contrasty and easier to follow than a child running to base in baseball!

I don't own the K5 (yet) [I have the K7 and K20D] so I can't vouch for your testing but I believe that trying on running subject TOWARD you would be better.
Also, the smaller the subject the more difficult it is for the AF to "track", so maybe if you are close enough to some birds (gulls?) you could give it a try?

Also, so you know, I started a thread on K5 AF testing here:
https://www.pentaxforums.com/forums/pentax-dslr-discussion/118553-k5-auto-focus-tests.html
and I thought perhaps it would be a good idea to post some of your pics there as well, as the risk of being redundant perhaps, but with different sample shots?

Anyway, great work you are doing for all of us who will buy the K5 soon.

Cheers!

JP

thanks jp.

Will try and find some kids or seagulls to shoot. Can you give me optimal settings to get the best out of the camera in this situation.

I would guess stopping down to f5.6 or f8 would be better and should I still go five point af so I get some chance of one af point being on the bird?

Terry

jaieger · 10-22-2010, 09:22 AM

Yknow that black cloud thing from Lost? Shoot that in low light at f1.4, center point only, with it coming directly at you at full speed. That should give the K-5 a workout

jake14mw · 10-22-2010, 09:32 AM

Telfish, I also thank you very much for posting your results. It's very much appreciated. My first reply to you is to not stop down to f 5.6 or 8. It will easier with shallow DOF to really determine whether the focus tracking is really working. So, I think you were correct in the way you did it. My wish would be to test this with somebody running towards the camera on grass. The reason I say on grass is that grass has texture to it, and it's easy to look at the grass and determine where the focus point really was.

To me, the other important variable is testing with centerpoint vs. auto 5 point like you chose. In field sports, centerpoint is often more desirable to use. Again, thank you very much. I'm sure you will get all different kinds of advice from many different users. I apreciate your efforts and those of others. This issue is important to me.

telfish · 10-22-2010, 09:46 AM

I have just taken some shots of my dog running towards me on grass at 2.8 I will post these for you. Some are in focus some seem behind the head I was aiming for, that may be tecnique.

Posted in the AF k5 tests thread here https://www.pentaxforums.com/forums/pentax-dslr-discussion/118553-k5-auto-foc...ml#post1233650

**jpzk** · 10-22-2010, 09:54 AM

Originally posted by telfish

thanks jp.

Will try and find some kids or seagulls to shoot. Can you give me optimal settings to get the best out of the camera in this situation.

I would guess stopping down to f5.6 or f8 would be better and should I still go five point af so I get some chance of one af point being on the bird?

Terry

Here's a good way to really test it as laid out by Terry: use the lens wide open under poor lighting conditions, and use centre point!
Of course, by using F5.6 to F8 would be better to help the AF but that's not the point here.
Best of luck in your "homework"!

JP

10-22-2010, 08:02 AM	#24
telfish Veteran Member Join Date: Aug 2009 Location: NY Posts: 407 Original Poster	Full size files can be seen here. Index of /telfish Many thanks to Priyantha Bleeker for the web hosting From the shots the camera did not fully lock until about the third frame. I am thinking that was because the car was in deep shadow and the 5 point af locked on the transition between the sunlit and shadowed road surface. Fast moving stuff is not my forte I usually do landscapes and macro so please excuse any lack of technique in these shots. From then on it found focus OK. Not an ideal test and I will do it again at a child’s baseball game. Last edited by telfish; 10-22-2010 at 08:08 AM.

10-22-2010, 09:46 AM	#29
telfish Veteran Member Join Date: Aug 2009 Location: NY Posts: 407 Original Poster	I have just taken some shots of my dog running towards me on grass at 2.8 I will post these for you. Some are in focus some seem behind the head I was aiming for, that may be tecnique. Posted in the AF k5 tests thread here https://www.pentaxforums.com/forums/pentax-dslr-discussion/118553-k5-auto-foc...ml#post1233650 Last edited by telfish; 10-22-2010 at 10:04 AM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
K-5, DNG/RAW ISO Case Study (6400-51200)...	JohnBee	Pentax K-5 & K-5 II	47	10-29-2010 09:14 PM
K-5, ISO80 vs ISO6400 Early Case Study(JPG)...	JohnBee	Pentax K-5 & K-5 II	67	10-18-2010 06:53 PM
FPS slows down during AFC	opiedog	Pentax DSLR Discussion	28	06-22-2010 07:52 PM
K-7 ISO3200 Advanced NR Case Study	JohnBee	Digital Processing, Software, and Printing	3	06-18-2010 01:02 AM
K7's afc is for real.	Torphoto	Pentax DSLR Discussion	20	07-27-2009 12:59 PM

10-22-2010, 06:29 AM	#16
telfish Veteran Member Join Date: Aug 2009 Location: NY Posts: 407 Original Poster	Originally posted by imtheguy Its hard to draw any conclusions from the first half of the shots as I look at the trees it does not seem to change focus very often. The closer shots look like its working! I am trying to do a similar test with a sea gull flying towards me but no complete results yet. Were you holding the shutter button fully down the entire time in 7fps mode or working from the half-pressed mode and clicking off individual shots quickly? I kept my finger pressed so it was taking shots as fast as it could camera set on continuous hi mode. I am uploading the full size files this morning so people can zoom into the farther shots. 800x600 shots are difficult at distance. Terry

10-22-2010, 06:30 AM	#17
rudemood11 New Member Join Date: Oct 2010 Posts: 4	Which version of Sigma lens is that? Screw drive one or the newer HSM II?

10-22-2010, 06:31 AM	#18
telfish Veteran Member Join Date: Aug 2009 Location: NY Posts: 407 Original Poster	Originally posted by rudemood11 Which version of Sigma lens is that? Screw drive one or the newer HSM II? It's the first HSM version. Terry

10-22-2010, 06:57 AM	#19
Conan Forum Member Join Date: Jun 2008 Location: Philippines Posts: 99	AFC test on moving cars is not a challenge. I used to use my K200D's AFC in a drifting event and it was quite easy since the target is so big.

10-22-2010, 07:04 AM	#20
JamieP Inactive Account Join Date: Sep 2008 Location: Cleveland Posts: 132	Sorry if I missed this but, were you shooting in focus or speed priority?

10-22-2010, 07:05 AM	#21
kenyee Veteran Member Join Date: Mar 2007 Location: Boston, PRofMA Photos: Albums Posts: 3,026	If you have a little league baseball match nearby, you might want to plant yourself near first base and fire up when the kid runs toward the base. I still remember a Canon 1dmkIII user complaining that only 60% of the shots were in focus when they all were in focus w/ his 1DmkII (this was Canon's mirror box misalignment problem that caused them to do a recall)...

10-22-2010, 07:15 AM	#22
telfish Veteran Member Join Date: Aug 2009 Location: NY Posts: 407 Original Poster	Originally posted by JamieP Sorry if I missed this but, were you shooting in focus or speed priority? Focus priority. I am not sure I used the optimum camera settings for this test, perhaps others should chime in and I will repeat it. Should I have stopped down to increase DOF. Should I have used only center AF point to prevent the camera focusing on trees as I panned? I wanted to make it as difficult as I could. Full size files are uploading and I will give the address to see them soon. Terry

10-22-2010, 07:16 AM	#23
telfish Veteran Member Join Date: Aug 2009 Location: NY Posts: 407 Original Poster	Originally posted by kenyee If you have a little league baseball match nearby, you might want to plant yourself near first base and fire up when the kid runs toward the base. I still remember a Canon 1dmkIII user complaining that only 60% of the shots were in focus when they all were in focus w/ his 1DmkII (this was Canon's mirror box misalignment problem that caused them to do a recall)... Will do, and I will post the results. Let me know what settings you prefer for the test.

10-22-2010, 08:59 AM	#25
jpzk Site Supporter Join Date: Jan 2009 Location: Québec Photos: Gallery \| Albums Posts: 8,251	Originally posted by telfish Full size files can be seen here. Index of /telfish Many thanks to Priyantha Bleeker for the web hosting From the shots the camera did not fully lock until about the third frame. I am thinking that was because the car was in deep shadow and the 5 point af locked on the transition between the sunlit and shadowed road surface. Fast moving stuff is not my forte I usually do landscapes and macro so please excuse any lack of technique in these shots. From then on it found focus OK. Not an ideal test and I will do it again at a child’s baseball game. Telfish, Thanks a bunch for the sample AF shots. I think that what some people were saying in that a car really is not the best subject for AF-C tests is that it is large, usually quite contrasty and easier to follow than a child running to base in baseball! I don't own the K5 (yet) [I have the K7 and K20D] so I can't vouch for your testing but I believe that trying on running subject TOWARD you would be better. Also, the smaller the subject the more difficult it is for the AF to "track", so maybe if you are close enough to some birds (gulls?) you could give it a try? Also, so you know, I started a thread on K5 AF testing here: https://www.pentaxforums.com/forums/pentax-dslr-discussion/118553-k5-auto-focus-tests.html and I thought perhaps it would be a good idea to post some of your pics there as well, as the risk of being redundant perhaps, but with different sample shots? Anyway, great work you are doing for all of us who will buy the K5 soon. Cheers! JP

10-22-2010, 09:07 AM	#26
telfish Veteran Member Join Date: Aug 2009 Location: NY Posts: 407 Original Poster	Originally posted by jpzk Telfish, Thanks a bunch for the sample AF shots. I think that what some people were saying in that a car really is not the best subject for AF-C tests is that it is large, usually quite contrasty and easier to follow than a child running to base in baseball! I don't own the K5 (yet) [I have the K7 and K20D] so I can't vouch for your testing but I believe that trying on running subject TOWARD you would be better. Also, the smaller the subject the more difficult it is for the AF to "track", so maybe if you are close enough to some birds (gulls?) you could give it a try? Also, so you know, I started a thread on K5 AF testing here: https://www.pentaxforums.com/forums/pentax-dslr-discussion/118553-k5-auto-focus-tests.html and I thought perhaps it would be a good idea to post some of your pics there as well, as the risk of being redundant perhaps, but with different sample shots? Anyway, great work you are doing for all of us who will buy the K5 soon. Cheers! JP thanks jp. Will try and find some kids or seagulls to shoot. Can you give me optimal settings to get the best out of the camera in this situation. I would guess stopping down to f5.6 or f8 would be better and should I still go five point af so I get some chance of one af point being on the bird? Terry

10-22-2010, 09:22 AM	#27
jaieger Veteran Member Join Date: Jul 2010 Photos: Gallery \| Albums Posts: 586	Yknow that black cloud thing from Lost? Shoot that in low light at f1.4, center point only, with it coming directly at you at full speed. That should give the K-5 a workout

10-22-2010, 09:32 AM	#28
jake14mw Veteran Member Join Date: Jan 2008 Location: Connecticut, USA Posts: 977	Telfish, I also thank you very much for posting your results. It's very much appreciated. My first reply to you is to not stop down to f 5.6 or 8. It will easier with shallow DOF to really determine whether the focus tracking is really working. So, I think you were correct in the way you did it. My wish would be to test this with somebody running towards the camera on grass. The reason I say on grass is that grass has texture to it, and it's easy to look at the grass and determine where the focus point really was. To me, the other important variable is testing with centerpoint vs. auto 5 point like you chose. In field sports, centerpoint is often more desirable to use. Again, thank you very much. I'm sure you will get all different kinds of advice from many different users. I apreciate your efforts and those of others. This issue is important to me.

10-22-2010, 09:54 AM	#30
jpzk Site Supporter Join Date: Jan 2009 Location: Québec Photos: Gallery \| Albums Posts: 8,251	Originally posted by telfish thanks jp. Will try and find some kids or seagulls to shoot. Can you give me optimal settings to get the best out of the camera in this situation. I would guess stopping down to f5.6 or f8 would be better and should I still go five point af so I get some chance of one af point being on the bird? Terry Here's a good way to really test it as laid out by Terry: use the lens wide open under poor lighting conditions, and use centre point! Of course, by using F5.6 to F8 would be better to help the AF but that's not the point here. Best of luck in your "homework"! JP