Lens correction for manual focus lenses?

aleonx3 · 03-05-2015, 05:17 PM

Originally posted by stevebrot

This should be the case, but there are a large number of users who depend on focus confirmation via the AF system when using manual focus lenses. As might imagine, the results are often a bit mixed with faster primes.

Steve

I have been using my auto-topcor lenses (m42) for a fairly long time now... and my personal experience is that I need to use the "apply all" adjustment to make the camera's focus confirmation to align accurately. For example, if I cancel the focus adjustment, I get 'front focus' using my m42 lens.... but after -8 or -9 adjustment, I got the focus fairly accurately. This change (of -8 apply all) does not affect any of my DA or FA lenses. So, I am happy about that.

So, I am not sure what other people's experience with AF adjustment, but I do believe that it helps also with manual focus lens as well especially with focus confirmation and catch-in-focus feature. So, it works for me but YMMV.

stevebrot · 03-05-2015, 05:30 PM

Originally posted by rburgoss

Can someone shed some light here?

Read comments above...

Seriously.

Focus confirmation for manual focus uses the AF system and works for both AF and manual focus lenses
Per lens AF adjustment of focus confirmation is not possible with manual focus lenses due to lack of lens identifier. That is a known.
Global changes to AF settings bias the AF system base functionality. That is why they call it global and why it affects focus confirmation for manual focus lenses. What the AF global settings do not change is the viewfinder focus performance (duh)
AF system calibration for manual focus lenses is a recurring question on this site. Why, I do not know.

In regards to the last point, it was explained to me on one thread that the user found manual focus difficult due poor eyesight, but really enjoyed using the manual focus lenses. My recommendations?

Forget attempting to tune the AF system to your manual focus glass. Doing so carries with it the risk of screwing up the performance of your AF lenses. Use the optical viewfinder or magnified live view for manual focus.
If you have poor eyesight, make sure the diopter is properly adjusted. A viewfinder magnifier has also proven valuable for many users on this site.
For fine focus, a properly dialed in aftermarket screen is superior to the stock screen
If you don't have money or inclination to fiddle with a proper focus screen, then magnified live view is your best option for fine focus.

Steve

stevebrot · 03-05-2015, 05:32 PM

Originally posted by aleonx3

This change (of -8 apply all) does not affect any of my DA or FA lenses. So, I am happy about that.

This was not my experience when I first tried it with my K-3, though things might have changed with firmware versions and I might not have been doing it right

Steve

stevebrot · 03-05-2015, 05:40 PM

Originally posted by aleonx3

I do believe that it helps also with manual focus lens as well especially with focus confirmation and catch-in-focus feature. So, it works for me but YMMV.

May I ask why you use the AF system for manual focus lenses. The AF system (focus confirm) is painfully imprecise when compared with a decent focus screen.

Steve

aleonx3 · 03-05-2015, 05:42 PM

Originally posted by stevebrot

This was not my experience when I first tried it with my K-3, though things might have changed with firmware versions and I might not have been doing it right

Steve

Steve, I am using DA* lenses frequently in event shooting and both are quite accurate without any 'individual' adjustment at all, I also have DA18-135 plus my 3 FA limited lenses. None of those requires any adjustment at all, and this is on my k-3 and k-5IIs.

stevebrot · 03-05-2015, 05:58 PM - **2 Likes**

Originally posted by rburgoss

Of course, there is still focus confirmation via PDAF or CDAF, but its bases on AF sensor positioning (PD) or actual CMOS image on sensor.

Originally posted by rburgoss

To my understanding, the AF adjustment affects on how long to move the AF drive (either screw or SDM) in order to achieve focus, but it does not actually change the AF adjustment (AF sensors or CMOS position) in relation to lens.

OK...I just reread your comment carefully and think I found the disconnect.

PDAF Focus confirm uses exactly the same mechanism as the PDAF auto focus
The system attains focus by detection, not by prediction. That being said, the point of detection for PDAF may not correspond to the true focus point due to the optical characteristics of the lens (flat, indistinct, or asymmetrical phase peak).
My understanding is that AF adjustment is a simple bias onto the system (shifts the detector off-set*) from the factory setting. Physical sensor position is not altered.

One last thing...AF adjustment applies to PDAF only. CDAF and focus peaking in live view are not adjustable.

Steve

* I have no information as to the exact process mechanism, but I have a few ideas.

aleonx3 · 03-05-2015, 07:17 PM - **1 Like**

Originally posted by stevebrot

May I ask why you use the AF system for manual focus lenses. The AF system (focus confirm) is painfully imprecise when compared with a decent focus screen.

Steve

Actually my experience using catch-in-focus, focus priority with focus confirmation is pretty accurate and with very high success rate, equal or better than using AF lenses. I used to rely on my eye to see if it locked focus, this is painfully inaccurate (I suppose better focus screen could help too), but that too can be due to user error. Besides, I don't have a good eye neither. Therefore, tuning the camera to get accurate focus is important, hence I play a lot with the adjustment. Another point about manual focus is that if the camera has persistently exhibiting front-focus behaviour, manual focus helps a great deal if I turn the focus from far-to-near until it snaps. Turing the focus ring the other way from near-to-far does not work that well as it would not be as accurate.

---------- Post added 03-05-2015 at 09:22 PM ----------

Originally posted by stevebrot

One last thing...AF adjustment applies to PDAF only. CDAF and focus peaking in live view are not adjustable.

Steve

* I have no information as to the exact process mechanism, but I have a few ideas.

I agree with you, the AF adjustment does not work for CDAF (as focus peaking in live view). I generally don't find focus peaking help (at least in my case) unless you have a good eye; it is hard for me to determine the optimal point of the high intensity in the outline when the intended subject is in focus.

stevebrot · 03-05-2015, 07:29 PM

Originally posted by aleonx3

I generally don't find focus peaking help (at least in my case) unless you have a good eye; it is hard for me to determine the optimal point of the high intensity in the outline when the intended subject is in focus.

I have found that it is most useful in the magnified view.

Steve

aleonx3 · 03-05-2015, 07:52 PM

Originally posted by stevebrot

I have found that it is most useful in the magnified view.

Steve

Agree... however, in event/stage photography it is hard to do when the subjects move around on stage. I have my k-01, but I don't use it for event photography, just for still objects.

emalvick · 03-06-2015, 10:08 AM

Originally posted by stevebrot

OK...I just reread your comment carefully and think I found the disconnect.

PDAF Focus confirm uses exactly the same mechanism as the PDAF auto focus
The system attains focus by detection, not by prediction. That being said, the point of detection for PDAF may not correspond to the true focus point due to the optical characteristics of the lens (flat, indistinct, or asymmetrical phase peak).
My understanding is that AF adjustment is a simple bias onto the system (shifts the detector off-set*) from the factory setting. Physical sensor position is not altered.

One last thing...AF adjustment applies to PDAF only. CDAF and focus peaking in live view are not adjustable.

Steve

* I have no information as to the exact process mechanism, but I have a few ideas.

This is all correct...

The big problem I see with trying to adjust for manual lenses is that the focus indicator does not necessarily indicate at the perfect focus point. If you play with manual focus, you see the focus confirmation begin and end over a range not at a point. This has led me to issues in relying on the focus indication when I'm shooting in situations with a narrow depth of field. Focus can end up front or back focused, and it depends on which direction I'm focusing from (front to back or back to front). It can be very difficult to know you are in the middle of the range, which would be the ideal.

Adjusting the global focus adjustment applies a bias to win the indicator lights up. I suppose the ideal would be to potentially dial the adjustment so the first indicator occurs at just the correct point, but I believe that would be a much larger bias than is truly needed and it would compromise the focus adjustments for autofocus lenses which are "in theory" hitting the perfect focus point. This would also require discipline in making sure you always adjust to focus from one side of perfect focus.

That being said, if a person can rely on the indicator and make it work for them without the complications listed above, more power to them. I'd guess that if you are often shooting at slower apertures the issues above are less of a problem.

For myself, I'd rather make my focus adjustments via installing a split prism and using the shims to get the focus correct.

stevebrot · 03-06-2015, 11:05 AM

Originally posted by emalvick

If you play with manual focus, you see the focus confirmation begin and end over a range not at a point.

Yes, this is definitely the case. What you are seeing is the limit of focus sensitivity for the active sensor. That "slop"/imprecision is the dirty secret of AF photography. Most of the time, you get reasonably good focus and all is good, but at wider taking apertures, the rate of missed focus goes up.

Originally posted by emalvick

For myself, I'd rather make my focus adjustments via installing a split prism and using the shims to get the focus correct.

...and there you have touched on the dirty secret of using the stock focus screen. Its focus sensitivity is a tad less than the center points on the K-3 (~ f/4 vs. f/2.8).

Steve

emalvick · 03-06-2015, 11:51 AM

Originally posted by stevebrot

...and there you have touched on the dirty secret of using the stock focus screen. Its focus sensitivity is a tad less than the center points on the K-3 (~ f/4 vs. f/2.8).

Steve

But, the K3 is the one camera I haven't yet changed the focus screen on, and I am finding that it is working well enough for me. I'm not advocating for that or even that the stock screen is good, but I am finding I am doing better with it than I have with the stock screens on my K5 or K10d (both of which I had switched to a split prism).

I haven't switched the K3 mostly because I've cut back significantly on manual focus since I've gotten all the limited FA lenses now, but that was somewhat driven by issues with MF, the focus indicator, and issues I've had with auto-exposure with a 3rd party focus screen.

I feel like photography pushes us to OCD a bit too much... or for me OCD led me to photography after failed attempts at other art.

03-05-2015, 05:30 PM	#17
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by rburgoss Can someone shed some light here? Read comments above... Seriously. Focus confirmation for manual focus uses the AF system and works for both AF and manual focus lenses Per lens AF adjustment of focus confirmation is not possible with manual focus lenses due to lack of lens identifier. That is a known. Global changes to AF settings bias the AF system base functionality. That is why they call it global and why it affects focus confirmation for manual focus lenses. What the AF global settings do not change is the viewfinder focus performance (duh) AF system calibration for manual focus lenses is a recurring question on this site. Why, I do not know. In regards to the last point, it was explained to me on one thread that the user found manual focus difficult due poor eyesight, but really enjoyed using the manual focus lenses. My recommendations? Forget attempting to tune the AF system to your manual focus glass. Doing so carries with it the risk of screwing up the performance of your AF lenses. Use the optical viewfinder or magnified live view for manual focus. If you have poor eyesight, make sure the diopter is properly adjusted. A viewfinder magnifier has also proven valuable for many users on this site. For fine focus, a properly dialed in aftermarket screen is superior to the stock screen If you don't have money or inclination to fiddle with a proper focus screen, then magnified live view is your best option for fine focus. Steve Last edited by stevebrot; 03-05-2015 at 05:37 PM.

03-05-2015, 05:58 PM - 2 Likes	#21
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by rburgoss Of course, there is still focus confirmation via PDAF or CDAF, but its bases on AF sensor positioning (PD) or actual CMOS image on sensor. Originally posted by rburgoss To my understanding, the AF adjustment affects on how long to move the AF drive (either screw or SDM) in order to achieve focus, but it does not actually change the AF adjustment (AF sensors or CMOS position) in relation to lens. OK...I just reread your comment carefully and think I found the disconnect. PDAF Focus confirm uses exactly the same mechanism as the PDAF auto focus The system attains focus by detection, not by prediction. That being said, the point of detection for PDAF may not correspond to the true focus point due to the optical characteristics of the lens (flat, indistinct, or asymmetrical phase peak). My understanding is that AF adjustment is a simple bias onto the system (shifts the detector off-set) from the factory setting. Physical sensor position is not altered. One last thing...AF adjustment applies to PDAF only. CDAF and focus peaking in live view are not adjustable. Steve I have no information as to the exact process mechanism, but I have a few ideas. Last edited by stevebrot; 03-05-2015 at 06:04 PM.

03-06-2015, 10:08 AM	#25
emalvick Veteran Member Join Date: Feb 2008 Location: Davis, CA Photos: Gallery Posts: 1,642	Originally posted by stevebrot OK...I just reread your comment carefully and think I found the disconnect. PDAF Focus confirm uses exactly the same mechanism as the PDAF auto focus The system attains focus by detection, not by prediction. That being said, the point of detection for PDAF may not correspond to the true focus point due to the optical characteristics of the lens (flat, indistinct, or asymmetrical phase peak). My understanding is that AF adjustment is a simple bias onto the system (shifts the detector off-set) from the factory setting. Physical sensor position is not altered. One last thing...AF adjustment applies to PDAF only. CDAF and focus peaking in live view are not adjustable. Steve I have no information as to the exact process mechanism, but I have a few ideas. This is all correct... The big problem I see with trying to adjust for manual lenses is that the focus indicator does not necessarily indicate at the perfect focus point. If you play with manual focus, you see the focus confirmation begin and end over a range not at a point. This has led me to issues in relying on the focus indication when I'm shooting in situations with a narrow depth of field. Focus can end up front or back focused, and it depends on which direction I'm focusing from (front to back or back to front). It can be very difficult to know you are in the middle of the range, which would be the ideal. Adjusting the global focus adjustment applies a bias to win the indicator lights up. I suppose the ideal would be to potentially dial the adjustment so the first indicator occurs at just the correct point, but I believe that would be a much larger bias than is truly needed and it would compromise the focus adjustments for autofocus lenses which are "in theory" hitting the perfect focus point. This would also require discipline in making sure you always adjust to focus from one side of perfect focus. That being said, if a person can rely on the indicator and make it work for them without the complications listed above, more power to them. I'd guess that if you are often shooting at slower apertures the issues above are less of a problem. For myself, I'd rather make my focus adjustments via installing a split prism and using the shims to get the focus correct.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Viewfinder ( focus screen) options for manual lenses	Dlanor Sekao	Pentax K-30 & K-50	6	12-01-2014 11:58 AM
Reversing Ring for Pentax K Manual Focus Lenses only CDN$ 1,192.85	SointulArt	Pentax Price Watch	12	11-22-2014 09:49 AM
New body for manual focus lenses	richard64	Pentax DSLR Discussion	13	03-23-2014 12:35 PM
Photoshop Lens correction filters for Pentax lenses	Photodana	Pentax SLR Lens Discussion	2	01-05-2014 10:43 AM
645D focus assist with Manual Focus Lenses?	uintaangler	Pentax Medium Format	11	04-12-2013 08:32 PM

03-05-2015, 05:17 PM	#16
aleonx3 Veteran Member Join Date: Jul 2007 Location: Brampton, Ontario Photos: Gallery \| Albums Posts: 3,996	Originally posted by stevebrot This should be the case, but there are a large number of users who depend on focus confirmation via the AF system when using manual focus lenses. As might imagine, the results are often a bit mixed with faster primes. Steve I have been using my auto-topcor lenses (m42) for a fairly long time now... and my personal experience is that I need to use the "apply all" adjustment to make the camera's focus confirmation to align accurately. For example, if I cancel the focus adjustment, I get 'front focus' using my m42 lens.... but after -8 or -9 adjustment, I got the focus fairly accurately. This change (of -8 apply all) does not affect any of my DA or FA lenses. So, I am happy about that. So, I am not sure what other people's experience with AF adjustment, but I do believe that it helps also with manual focus lens as well especially with focus confirmation and catch-in-focus feature. So, it works for me but YMMV.

03-05-2015, 05:32 PM	#18
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by aleonx3 This change (of -8 apply all) does not affect any of my DA or FA lenses. So, I am happy about that. This was not my experience when I first tried it with my K-3, though things might have changed with firmware versions and I might not have been doing it right Steve

03-05-2015, 05:40 PM	#19
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by aleonx3 I do believe that it helps also with manual focus lens as well especially with focus confirmation and catch-in-focus feature. So, it works for me but YMMV. May I ask why you use the AF system for manual focus lenses. The AF system (focus confirm) is painfully imprecise when compared with a decent focus screen. Steve

03-05-2015, 05:42 PM	#20
aleonx3 Veteran Member Join Date: Jul 2007 Location: Brampton, Ontario Photos: Gallery \| Albums Posts: 3,996	Originally posted by stevebrot This was not my experience when I first tried it with my K-3, though things might have changed with firmware versions and I might not have been doing it right Steve Steve, I am using DA* lenses frequently in event shooting and both are quite accurate without any 'individual' adjustment at all, I also have DA18-135 plus my 3 FA limited lenses. None of those requires any adjustment at all, and this is on my k-3 and k-5IIs.

03-05-2015, 07:17 PM - 1 Like	#22
aleonx3 Veteran Member Join Date: Jul 2007 Location: Brampton, Ontario Photos: Gallery \| Albums Posts: 3,996	Originally posted by stevebrot May I ask why you use the AF system for manual focus lenses. The AF system (focus confirm) is painfully imprecise when compared with a decent focus screen. Steve Actually my experience using catch-in-focus, focus priority with focus confirmation is pretty accurate and with very high success rate, equal or better than using AF lenses. I used to rely on my eye to see if it locked focus, this is painfully inaccurate (I suppose better focus screen could help too), but that too can be due to user error. Besides, I don't have a good eye neither. Therefore, tuning the camera to get accurate focus is important, hence I play a lot with the adjustment. Another point about manual focus is that if the camera has persistently exhibiting front-focus behaviour, manual focus helps a great deal if I turn the focus from far-to-near until it snaps. Turing the focus ring the other way from near-to-far does not work that well as it would not be as accurate. ---------- Post added 03-05-2015 at 09:22 PM ---------- Originally posted by stevebrot One last thing...AF adjustment applies to PDAF only. CDAF and focus peaking in live view are not adjustable. Steve * I have no information as to the exact process mechanism, but I have a few ideas. I agree with you, the AF adjustment does not work for CDAF (as focus peaking in live view). I generally don't find focus peaking help (at least in my case) unless you have a good eye; it is hard for me to determine the optimal point of the high intensity in the outline when the intended subject is in focus.

03-05-2015, 07:29 PM	#23
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by aleonx3 I generally don't find focus peaking help (at least in my case) unless you have a good eye; it is hard for me to determine the optimal point of the high intensity in the outline when the intended subject is in focus. I have found that it is most useful in the magnified view. Steve

03-05-2015, 07:52 PM	#24
aleonx3 Veteran Member Join Date: Jul 2007 Location: Brampton, Ontario Photos: Gallery \| Albums Posts: 3,996	Originally posted by stevebrot I have found that it is most useful in the magnified view. Steve Agree... however, in event/stage photography it is hard to do when the subjects move around on stage. I have my k-01, but I don't use it for event photography, just for still objects.

03-06-2015, 11:05 AM	#26
stevebrot Otis Memorial Pentaxian Join Date: Mar 2007 Location: Vancouver (USA) Photos: Gallery \| Albums Posts: 42,007	Originally posted by emalvick If you play with manual focus, you see the focus confirmation begin and end over a range not at a point. Yes, this is definitely the case. What you are seeing is the limit of focus sensitivity for the active sensor. That "slop"/imprecision is the dirty secret of AF photography. Most of the time, you get reasonably good focus and all is good, but at wider taking apertures, the rate of missed focus goes up. Originally posted by emalvick For myself, I'd rather make my focus adjustments via installing a split prism and using the shims to get the focus correct. ...and there you have touched on the dirty secret of using the stock focus screen. Its focus sensitivity is a tad less than the center points on the K-3 (~ f/4 vs. f/2.8). Steve

03-06-2015, 11:51 AM	#27
emalvick Veteran Member Join Date: Feb 2008 Location: Davis, CA Photos: Gallery Posts: 1,642	Originally posted by stevebrot ...and there you have touched on the dirty secret of using the stock focus screen. Its focus sensitivity is a tad less than the center points on the K-3 (~ f/4 vs. f/2.8). Steve But, the K3 is the one camera I haven't yet changed the focus screen on, and I am finding that it is working well enough for me. I'm not advocating for that or even that the stock screen is good, but I am finding I am doing better with it than I have with the stock screens on my K5 or K10d (both of which I had switched to a split prism). I haven't switched the K3 mostly because I've cut back significantly on manual focus since I've gotten all the limited FA lenses now, but that was somewhat driven by issues with MF, the focus indicator, and issues I've had with auto-exposure with a 3rd party focus screen. I feel like photography pushes us to OCD a bit too much... or for me OCD led me to photography after failed attempts at other art.