Catch in focus help/questions

FernandoBatista · 02-11-2013, 07:09 AM

Hello,

I have some questions about Pentax’s CIF and would appreciate some help please.

A bit of background first, I’m not a Pentax user and never used a Pentax body, I was a Nikon user for about 13 years until recently.
I mostly use long MF lenses (>300mm), until now I’ve used several Nikon bodies digital and analog, the last being two D90s.
With my D90 bodies I used all my lenses chipped with dandelion chips. For those who don’t know they’re a kind of AF emulator chip, completely tunable for the lens specs (focal lenght, aperture, even back/front focus adjustment).
It all worked perfectly, but both my bodies were getting old with several shots made and had to be replaced, so I went the obvious way and tried a D7000 and D300.
But the more advanced Nikon bodies have a different AF system, where the AF confirmation (and Trap focus) isn’t so precise as the basic bodies (nothing to do with it’s af capabilities witch are much better than basic bodies).
What happens is that the camera confirms focus and takes a photo even being slightly out of focus, it happens in both ends of DOF and has nothing to do with front or backfocus. There seems to be a higher tolerance to the exact focus point in the D7000/300 than there was in the cheaper/older bodies.

So I’ve been looking for other choices. I really don’t care what brand I use, what I want is the best body for manual focus, with a decent viewfinder.

I’ve been using a Sony A65 now, mostly for it’s focus peaking together with an amazing EVF witch is really nice. But trap focus while slightly better than D7000 is still not as precise as D90’s.

So, now I’m wondering about Pentax (specially the K30), how accurate does CIF work? Does it release the trigger really only when the image is focused? or is there some tolerance and it will fire even slightly OOF? Is it the same in all Pentax cameras, or there’s differences between the various Pentax bodies?

I’d really love to try this for myself, but Pentax users are rare around here, so I’d be really grateful if someone can help me out.

Thank you.

Regards,
Fernando

Salagubang · 02-11-2013, 08:07 AM

I have no experience with Nikon CIF and K30 but my experiences with K10, K200, KM, KX have all been similar to what you've experienced with what seem to be higher triggering tolerance. The problem seems to be caused by the size of the middle focus sensor (lets just call it square), it will fire the shutter if focus is detected at any of its edges; and these sensors are cross type.

Pentax older models has what is called debug mode where you can adjust front or back focus to compensate where you plan on which direction you will turn the focus ring.

As alternative, my technique is to set the debug mode to vertical sensors only as active being horizontal subjects tends to stand straight up hence not much affected by the size of the focus sensor.

FernandoBatista · 02-11-2013, 08:46 AM

Hi,

Thank you for your input.

I assume what you mean is in real life subjects, correct? I think what you're saying is caused by the tridimensionality of objects, AF can detect any point of the subject, front or back.

In my case, the tests I made aren't affected by the sensor size, or the sensors being crossed or not. Actually my results with nikon where very consistent with all focus points.
I adjust trapfocus settings using a tool similar to the LensAlign. The AF point is completely covered by the target, so it should only detect what is on that exact plane, nothing in the front or back. Hope I'm explaining myself correctly :P

**TER-OR** · 02-11-2013, 08:47 AM

The CIF experience I have with the K-5 is that it's great when the subject stands out from the background. So, for flowers etc, it's good. For something sitting on the ground it's much trickier - and you will have better luck walking the focus forward etc - triggering while you rotate the focus.

carpenter bee | Flickr - Photo Sharing!
Using the old M50 f/2 lens.

**jatrax** · 02-11-2013, 10:43 AM

I've no Nikon experience (or K-30) but on both my k-x and k-5 the CIF has some 'slop' in it. On some lenses it is better to start at a distance and focus toward close, on some lenses the reverse works better. Without experience on both systems I think it would be very hard for anyone to say one is better than the other. I can say that it works on Pentax and well enough to get used regularly. But I use it mostly on shorter focal lengths, never tried it on 300mm+.

rockmaster1964 · 02-11-2013, 11:51 AM

I thought it was very hard to use on the K10, but it works great on my K5.

IchabodCrane · 02-11-2013, 03:03 PM

Just a hunch... CIF likely works best with a fast lens that's stopped down a good amount. The fast lens should decrease the "slop" in the focus while the smaller aperture upon actuation will allow higher tolerance of what slop in the focus accuracy remains. Any merit to this?

**jatrax** · 02-11-2013, 03:49 PM

Maybe, but I think my issues have been more because of the relatively large central focus 'point' so you are never quite sure what it will lock on to. Stopped down it usually works but most of what I've tried have been closeups and sometimes it just catches the wrong place.

Laurentiu Cristofor · 02-11-2013, 05:20 PM

Originally posted by FernandoBatista

So I’ve been looking for other choices. I really don’t care what brand I use, what I want is the best body for manual focus, with a decent viewfinder.

At the moment, for me, that is the Olympus E-M5.

Originally posted by FernandoBatista

So, now I’m wondering about Pentax (specially the K30), how accurate does CIF work? Does it release the trigger really only when the image is focused? or is there some tolerance and it will fire even slightly OOF? Is it the same in all Pentax cameras, or there’s differences between the various Pentax bodies?

CIF is only as good as the camera AF is. With the K10D and K-7, I always noticed that the focusing confirmation stayed on for a bit while I was adjusting focus - this is a bit like focus peaking which can continue to cover some area while you keep adjusting focus. So my answers to all your quoted questions are: CIF is as accurate as the camera's AF, it can release the trigger when the focus is not perfectly acquired but is close, and there are differences between the bodies, because the AF algorithms are continuously being improved.

causey · 02-11-2013, 08:49 PM

Quote:

CIF is only as good as the camera AF is.

That's why one can use the focus adjust function to improve CIF performance. My K-x doesn't officially allow focus adjustment, but I get in the debug mode to deliberately make lenses front-focus a bit. With CIF usually my problem is back-focus.

This is a set of pics for which I used Rokinon 85/1.4 on K-x, with CIF:
https://www.pentaxforums.com/forums/post-your-photos/215046-streets-chinese-y...ngton-d-c.html

bullitt_60 · 02-11-2013, 09:40 PM

It works great for me, but I haven't been using it as it startles the bejesus out of me.

FernandoBatista · 02-12-2013, 01:40 PM

Hello everyone,
Thank you all very much for your thoughts and ideas, they're much appreciated.

Some comments to some of your answers following:

Originally posted by jatrax

.... On some lenses it is better to start at a distance and focus toward close, on some lenses the reverse works better. ...

For what I understand from your statement it seems to me the behavior is similar to what I get with the D7000 or A65.
With the D90 it didn't matter where I was coming from, the camera allowed the shutter to be released only when the image was in focus, this made a world of difference when shooting fast birds. In the D90 there is a very slight difference in the focus position coming from back or front, but the subject was always within DOF... always.

Originally posted by IchabodCrane

Just a hunch... CIF likely works best with a fast lens that's stopped down a good amount. The fast lens should decrease the "slop" in the focus while the smaller aperture upon actuation will allow higher tolerance of what slop in the focus accuracy remains. Any merit to this?

Well, yes, it's a good suggestion, it's a workaround to this limitation. But it only works if the used lens have auto aperture, and if I'm willing to use it stopped down, witch happens rarely

Originally posted by Laurentiu Cristofor

At the moment, for me, that is the Olympus E-M5.

Ok, but why? I also have a G5, witch for MF usage is not that different than the EM5. But when it comes to MF I far prefer the Sony. Peaking puts it on another level, so does the EVF.

Originally posted by Laurentiu Cristofor

CIF is only as good as the camera AF is.

That was my expectation, but unfortunately not my experience, The D7000 and D300 AF are both very good, much better than D90, but still I have problems with trap focus on these bodies, the D90 on the other hand is much more reliable.
IMO a good AF doesn't warrant a good CIF/trap focus.

Originally posted by Laurentiu Cristofor

With the K10D and K-7, I always noticed that the focusing confirmation stayed on for a bit while I was adjusting focus - this is a bit like focus peaking which can continue to cover some area while you keep adjusting focus.
So my answers to all your quoted questions are: CIF is as accurate as the camera's AF, it can release the trigger when the focus is not perfectly acquired but is close, and there are diferences between the bodies, because the AF algorithms are continuously being improved.

Having peaking showing doesn't mean it's the point of focus...true. But with peaking you can adjust back and forth and look for the best point, and used like that is 100% acurate and also very fast. This at least is my experience with it.

CIF on the other hand cannot be used like that, to be reliable it should only allow the shutter to be released if the chosen focus point is at least somewhere within DOF, that was what happened with my older bodies. Is the newer bodies with newer and improved AF algorithms that are giving me problems.

Getting back to peaking, as good as I find it, in it's current state it's only really good for static or slow moving shots, for anything moving fast I still preferred my D90s trap focus.

Laurentiu Cristofor · 02-12-2013, 03:23 PM

Originally posted by causey

That's why one can use the focus adjust function to improve CIF performance.

The problem I was describing goes beyond what you can adjust. Even with perfect calibration, CIF and AF have problems because of the size of the AF sensors. They focus accurately, but not where you wanted them to.

Originally posted by FernandoBatista

Ok, but why? I also have a G5, witch for MF usage is not that different than the EM5. But when it comes to MF I far prefer the Sony. Peaking puts it on another level, so does the EVF.

Because I like the EVF. I thought the Sony EVF was great, but that was before I got the E-M5. And I found focus peaking more distracting than useful for my manual focusing, so I have little interest in it. YMMV, which is why I qualified my statement with "for me".

Originally posted by FernandoBatista

That was my expectation, but unfortunately not my experience, The D7000 and D300 AF are both very good, much better than D90, but still I have problems with trap focus on these bodies, the D90 on the other hand is much more reliable.
IMO a good AF doesn't warrant a good CIF/trap focus.

OK, let me reformulate. What I was trying to say was that the CIF cannot be better than the AF. I guess it can be worse.

Originally posted by FernandoBatista

Having peaking showing doesn't mean it's the point of focus...true. But with peaking you can adjust back and forth and look for the best point, and used like that is 100% acurate and also very fast. This at least is my experience with it.

That is a different subject now. My point was that just because the AF confirmation point lights up, it does not mean that focus was achieved where you wanted (I compared with focus peaking for this point only). This happens because the size of the AF point doesn't reflect necessarily the size of the AF sensor, so the camera may have achieved focus somewhere in that area, but not exactly in the point you wanted it to be focused. This is why focus adjustments don't help and why I could see focus points stay lit up despite me still moving the focusing ring - some area of the AF sensor was still detecting focus.

Because of these kind of issues, I prefer to not use AF at all. After initially trying to use the AF confirmation points as helpers for my manual focusing, I ended up ignoring them and just relying on my eyes. Now that I don't even use SLRs that often, I got rid of their distraction for good.

FernandoBatista · 02-12-2013, 03:56 PM

Quote:

That is a different subject now. My point was that just because the AF confirmation point lights up, it does not mean that focus was achieved where you wanted (I compared with focus peaking for this point only). This happens because the size of the AF point doesn't reflect necessarily the size of the AF sensor, so the camera may have achieved focus somewhere in that area, but not exactly in the point you wanted it to be focused. This is why focus adjustments don't help and why I could see focus points stay lit up despite me still moving the focusing ring - some area of the AF sensor was still detecting focus.

Because of these kind of issues, I prefer to not use AF at all. After initially trying to use the AF confirmation points as helpers for my manual focusing, I ended up ignoring them and just relying on my eyes. Now that I don't even use SLRs that often, I got rid of their distraction for good.

Well, I never had any issues with the size of the AF points, if there's something in the back it can surely confirm focus on it, but it's rarely a problem with my type of shooting. Usually I have too much separation between my subjects and any background.

That said, I understand your points, and for static shots I really could live with just a good EVF, but I like to shoot action and using an MF lens with accurate results without some kind of aid for fast action is not exactly easy.

BTW, just to give an idea, this is what I usually shoot:
http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-KtzjLqK/0/O/2012_03_03_1.jpg

http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-WvvxLrP/0/O/2011_11_27_5.jpg

http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-n6GpQ7G/0/O/2011_04_23_1.jpg

http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-4SgqVbv/0/O/2011_04_16_1.jpg

Doing images like this with the D90 was almost as easy as any AF lens, or even easier in some points... but doing it without the help of a reliable trapfocus, no matter how good the viewfinder or peaking, is just something I'm not seeing happening.

Laurentiu Cristofor · 02-13-2013, 12:51 PM

Originally posted by FernandoBatista

Well, I never had any issues with the size of the AF points, if there's something in the back it can surely confirm focus on it, but it's rarely a problem with my type of shooting. Usually I have too much separation between my subjects and any background.

That said, I understand your points, and for static shots I really could live with just a good EVF, but I like to shoot action and using an MF lens with accurate results without some kind of aid for fast action is not exactly easy.

Yes, I guess I had my scenarios in mind more than yours. My "focus challenge scenario" is available light portraiture and in that scenario, with thin DOF it makes a difference if you focus on the eye or on the brow - the focusing point can often overlap both, as it is not easy to keep it just over the eye if the subject is moving

. This is also where focus peaking was failing - it was highlighting both the pupil and the brow and the highlight over the pupil was not easy to notice either.

Originally posted by FernandoBatista

BTW, just to give an idea, this is what I usually shoot:
http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-KtzjLqK/0/O/2012_03_03_1.jpg

http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-WvvxLrP/0/O/2011_11_27_5.jpg

http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-n6GpQ7G/0/O/2011_04_23_1.jpg

http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-4SgqVbv/0/O/2011_04_16_1.jpg

Doing images like this with the D90 was almost as easy as any AF lens, or even easier in some points... but doing it without the help of a reliable trapfocus, no matter how good the viewfinder or peaking, is just something I'm not seeing happening.

Very nice work! I never had a lens that is suitable for this scenario. Speaking of lenses, which Pentax lens would you plan to use for such work?

02-11-2013, 07:09 AM	#1
FernandoBatista New Member Join Date: Feb 2013 Posts: 8	Catch in focus help/questions Hello, I have some questions about Pentax’s CIF and would appreciate some help please. A bit of background first, I’m not a Pentax user and never used a Pentax body, I was a Nikon user for about 13 years until recently. I mostly use long MF lenses (>300mm), until now I’ve used several Nikon bodies digital and analog, the last being two D90s. With my D90 bodies I used all my lenses chipped with dandelion chips. For those who don’t know they’re a kind of AF emulator chip, completely tunable for the lens specs (focal lenght, aperture, even back/front focus adjustment). It all worked perfectly, but both my bodies were getting old with several shots made and had to be replaced, so I went the obvious way and tried a D7000 and D300. But the more advanced Nikon bodies have a different AF system, where the AF confirmation (and Trap focus) isn’t so precise as the basic bodies (nothing to do with it’s af capabilities witch are much better than basic bodies). What happens is that the camera confirms focus and takes a photo even being slightly out of focus, it happens in both ends of DOF and has nothing to do with front or backfocus. There seems to be a higher tolerance to the exact focus point in the D7000/300 than there was in the cheaper/older bodies. So I’ve been looking for other choices. I really don’t care what brand I use, what I want is the best body for manual focus, with a decent viewfinder. I’ve been using a Sony A65 now, mostly for it’s focus peaking together with an amazing EVF witch is really nice. But trap focus while slightly better than D7000 is still not as precise as D90’s. So, now I’m wondering about Pentax (specially the K30), how accurate does CIF work? Does it release the trigger really only when the image is focused? or is there some tolerance and it will fire even slightly OOF? Is it the same in all Pentax cameras, or there’s differences between the various Pentax bodies? I’d really love to try this for myself, but Pentax users are rare around here, so I’d be really grateful if someone can help me out. Thank you. Regards, Fernando

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Focus Trapping (aka Catch-In Focus)	regor	Pentax K-01	7	01-05-2013 11:32 PM
catch in focus	jennverr	Pentax DSLR Discussion	20	04-07-2012 09:08 AM
Help! The Bats are Back! Catch In Focus	seachunk2	Troubleshooting and Beginner Help	9	07-08-2011 03:21 PM
Catch in Focus	mand	Pentax DSLR Discussion	18	02-25-2011 04:38 PM
K7 & catch-in-focus... help!	Nass	Troubleshooting and Beginner Help	5	09-27-2009 12:36 PM

02-11-2013, 08:07 AM	#2
Salagubang Senior Member Join Date: Oct 2011 Location: San Jose Del Monte Photos: Gallery \| Albums Posts: 198	I have no experience with Nikon CIF and K30 but my experiences with K10, K200, KM, KX have all been similar to what you've experienced with what seem to be higher triggering tolerance. The problem seems to be caused by the size of the middle focus sensor (lets just call it square), it will fire the shutter if focus is detected at any of its edges; and these sensors are cross type. Pentax older models has what is called debug mode where you can adjust front or back focus to compensate where you plan on which direction you will turn the focus ring. As alternative, my technique is to set the debug mode to vertical sensors only as active being horizontal subjects tends to stand straight up hence not much affected by the size of the focus sensor.

02-11-2013, 08:46 AM	#3
FernandoBatista New Member Join Date: Feb 2013 Posts: 8 Original Poster	Hi, Thank you for your input. I assume what you mean is in real life subjects, correct? I think what you're saying is caused by the tridimensionality of objects, AF can detect any point of the subject, front or back. In my case, the tests I made aren't affected by the sensor size, or the sensors being crossed or not. Actually my results with nikon where very consistent with all focus points. I adjust trapfocus settings using a tool similar to the LensAlign. The AF point is completely covered by the target, so it should only detect what is on that exact plane, nothing in the front or back. Hope I'm explaining myself correctly :P

02-11-2013, 08:47 AM	#4
TER-OR Site Supporter Join Date: Apr 2009 Location: Dundee, IL Photos: Gallery \| Albums Posts: 6,699	The CIF experience I have with the K-5 is that it's great when the subject stands out from the background. So, for flowers etc, it's good. For something sitting on the ground it's much trickier - and you will have better luck walking the focus forward etc - triggering while you rotate the focus. carpenter bee \| Flickr - Photo Sharing! Using the old M50 f/2 lens.

02-11-2013, 10:43 AM	#5
jatrax Site Supporter Join Date: May 2010 Location: Washington Cascades Photos: Gallery \| Albums Posts: 12,991	I've no Nikon experience (or K-30) but on both my k-x and k-5 the CIF has some 'slop' in it. On some lenses it is better to start at a distance and focus toward close, on some lenses the reverse works better. Without experience on both systems I think it would be very hard for anyone to say one is better than the other. I can say that it works on Pentax and well enough to get used regularly. But I use it mostly on shorter focal lengths, never tried it on 300mm+.

02-11-2013, 11:51 AM	#6
rockmaster1964 Veteran Member Join Date: Oct 2008 Location: Frisco Texas Photos: Gallery Posts: 382	I thought it was very hard to use on the K10, but it works great on my K5.

02-11-2013, 03:03 PM	#7
IchabodCrane Veteran Member Join Date: Sep 2012 Photos: Gallery \| Albums Posts: 1,728	Just a hunch... CIF likely works best with a fast lens that's stopped down a good amount. The fast lens should decrease the "slop" in the focus while the smaller aperture upon actuation will allow higher tolerance of what slop in the focus accuracy remains. Any merit to this?

02-11-2013, 03:49 PM	#8
jatrax Site Supporter Join Date: May 2010 Location: Washington Cascades Photos: Gallery \| Albums Posts: 12,991	Maybe, but I think my issues have been more because of the relatively large central focus 'point' so you are never quite sure what it will lock on to. Stopped down it usually works but most of what I've tried have been closeups and sometimes it just catches the wrong place.

02-11-2013, 05:20 PM	#9
Laurentiu Cristofor Banned Join Date: Nov 2009 Location: WA Photos: Albums Posts: 3,055	Originally posted by FernandoBatista So I’ve been looking for other choices. I really don’t care what brand I use, what I want is the best body for manual focus, with a decent viewfinder. At the moment, for me, that is the Olympus E-M5. Originally posted by FernandoBatista So, now I’m wondering about Pentax (specially the K30), how accurate does CIF work? Does it release the trigger really only when the image is focused? or is there some tolerance and it will fire even slightly OOF? Is it the same in all Pentax cameras, or there’s differences between the various Pentax bodies? CIF is only as good as the camera AF is. With the K10D and K-7, I always noticed that the focusing confirmation stayed on for a bit while I was adjusting focus - this is a bit like focus peaking which can continue to cover some area while you keep adjusting focus. So my answers to all your quoted questions are: CIF is as accurate as the camera's AF, it can release the trigger when the focus is not perfectly acquired but is close, and there are differences between the bodies, because the AF algorithms are continuously being improved.

02-11-2013, 08:49 PM	#10
causey Veteran Member Join Date: Feb 2009 Location: Arlington, VA Photos: Albums Posts: 3,757	Quote: CIF is only as good as the camera AF is. That's why one can use the focus adjust function to improve CIF performance. My K-x doesn't officially allow focus adjustment, but I get in the debug mode to deliberately make lenses front-focus a bit. With CIF usually my problem is back-focus. This is a set of pics for which I used Rokinon 85/1.4 on K-x, with CIF: https://www.pentaxforums.com/forums/post-your-photos/215046-streets-chinese-y...ngton-d-c.html

02-11-2013, 09:40 PM	#11
bullitt_60 Veteran Member Join Date: Nov 2012 Location: Atlanta, GA Posts: 320	It works great for me, but I haven't been using it as it startles the bejesus out of me.

02-12-2013, 01:40 PM	#12
FernandoBatista New Member Join Date: Feb 2013 Posts: 8 Original Poster	Hello everyone, Thank you all very much for your thoughts and ideas, they're much appreciated. Some comments to some of your answers following: Originally posted by jatrax .... On some lenses it is better to start at a distance and focus toward close, on some lenses the reverse works better. ... For what I understand from your statement it seems to me the behavior is similar to what I get with the D7000 or A65. With the D90 it didn't matter where I was coming from, the camera allowed the shutter to be released only when the image was in focus, this made a world of difference when shooting fast birds. In the D90 there is a very slight difference in the focus position coming from back or front, but the subject was always within DOF... always. Originally posted by IchabodCrane Just a hunch... CIF likely works best with a fast lens that's stopped down a good amount. The fast lens should decrease the "slop" in the focus while the smaller aperture upon actuation will allow higher tolerance of what slop in the focus accuracy remains. Any merit to this? Well, yes, it's a good suggestion, it's a workaround to this limitation. But it only works if the used lens have auto aperture, and if I'm willing to use it stopped down, witch happens rarely Originally posted by Laurentiu Cristofor At the moment, for me, that is the Olympus E-M5. Ok, but why? I also have a G5, witch for MF usage is not that different than the EM5. But when it comes to MF I far prefer the Sony. Peaking puts it on another level, so does the EVF. Originally posted by Laurentiu Cristofor CIF is only as good as the camera AF is. That was my expectation, but unfortunately not my experience, The D7000 and D300 AF are both very good, much better than D90, but still I have problems with trap focus on these bodies, the D90 on the other hand is much more reliable. IMO a good AF doesn't warrant a good CIF/trap focus. Originally posted by Laurentiu Cristofor With the K10D and K-7, I always noticed that the focusing confirmation stayed on for a bit while I was adjusting focus - this is a bit like focus peaking which can continue to cover some area while you keep adjusting focus. So my answers to all your quoted questions are: CIF is as accurate as the camera's AF, it can release the trigger when the focus is not perfectly acquired but is close, and there are diferences between the bodies, because the AF algorithms are continuously being improved. Having peaking showing doesn't mean it's the point of focus...true. But with peaking you can adjust back and forth and look for the best point, and used like that is 100% acurate and also very fast. This at least is my experience with it. CIF on the other hand cannot be used like that, to be reliable it should only allow the shutter to be released if the chosen focus point is at least somewhere within DOF, that was what happened with my older bodies. Is the newer bodies with newer and improved AF algorithms that are giving me problems. Getting back to peaking, as good as I find it, in it's current state it's only really good for static or slow moving shots, for anything moving fast I still preferred my D90s trap focus.

02-12-2013, 03:23 PM	#13
Laurentiu Cristofor Banned Join Date: Nov 2009 Location: WA Photos: Albums Posts: 3,055	Originally posted by causey That's why one can use the focus adjust function to improve CIF performance. The problem I was describing goes beyond what you can adjust. Even with perfect calibration, CIF and AF have problems because of the size of the AF sensors. They focus accurately, but not where you wanted them to. Originally posted by FernandoBatista Ok, but why? I also have a G5, witch for MF usage is not that different than the EM5. But when it comes to MF I far prefer the Sony. Peaking puts it on another level, so does the EVF. Because I like the EVF. I thought the Sony EVF was great, but that was before I got the E-M5. And I found focus peaking more distracting than useful for my manual focusing, so I have little interest in it. YMMV, which is why I qualified my statement with "for me". Originally posted by FernandoBatista That was my expectation, but unfortunately not my experience, The D7000 and D300 AF are both very good, much better than D90, but still I have problems with trap focus on these bodies, the D90 on the other hand is much more reliable. IMO a good AF doesn't warrant a good CIF/trap focus. OK, let me reformulate. What I was trying to say was that the CIF cannot be better than the AF. I guess it can be worse. Originally posted by FernandoBatista Having peaking showing doesn't mean it's the point of focus...true. But with peaking you can adjust back and forth and look for the best point, and used like that is 100% acurate and also very fast. This at least is my experience with it. That is a different subject now. My point was that just because the AF confirmation point lights up, it does not mean that focus was achieved where you wanted (I compared with focus peaking for this point only). This happens because the size of the AF point doesn't reflect necessarily the size of the AF sensor, so the camera may have achieved focus somewhere in that area, but not exactly in the point you wanted it to be focused. This is why focus adjustments don't help and why I could see focus points stay lit up despite me still moving the focusing ring - some area of the AF sensor was still detecting focus. Because of these kind of issues, I prefer to not use AF at all. After initially trying to use the AF confirmation points as helpers for my manual focusing, I ended up ignoring them and just relying on my eyes. Now that I don't even use SLRs that often, I got rid of their distraction for good.

02-12-2013, 03:56 PM	#14
FernandoBatista New Member Join Date: Feb 2013 Posts: 8 Original Poster	Quote: That is a different subject now. My point was that just because the AF confirmation point lights up, it does not mean that focus was achieved where you wanted (I compared with focus peaking for this point only). This happens because the size of the AF point doesn't reflect necessarily the size of the AF sensor, so the camera may have achieved focus somewhere in that area, but not exactly in the point you wanted it to be focused. This is why focus adjustments don't help and why I could see focus points stay lit up despite me still moving the focusing ring - some area of the AF sensor was still detecting focus. Because of these kind of issues, I prefer to not use AF at all. After initially trying to use the AF confirmation points as helpers for my manual focusing, I ended up ignoring them and just relying on my eyes. Now that I don't even use SLRs that often, I got rid of their distraction for good. Well, I never had any issues with the size of the AF points, if there's something in the back it can surely confirm focus on it, but it's rarely a problem with my type of shooting. Usually I have too much separation between my subjects and any background. That said, I understand your points, and for static shots I really could live with just a good EVF, but I like to shoot action and using an MF lens with accurate results without some kind of aid for fast action is not exactly easy. BTW, just to give an idea, this is what I usually shoot: http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-KtzjLqK/0/O/2012_03_03_1.jpg http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-WvvxLrP/0/O/2011_11_27_5.jpg http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-n6GpQ7G/0/O/2011_04_23_1.jpg http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-4SgqVbv/0/O/2011_04_16_1.jpg Doing images like this with the D90 was almost as easy as any AF lens, or even easier in some points... but doing it without the help of a reliable trapfocus, no matter how good the viewfinder or peaking, is just something I'm not seeing happening.

02-13-2013, 12:51 PM	#15
Laurentiu Cristofor Banned Join Date: Nov 2009 Location: WA Photos: Albums Posts: 3,055	Originally posted by FernandoBatista Well, I never had any issues with the size of the AF points, if there's something in the back it can surely confirm focus on it, but it's rarely a problem with my type of shooting. Usually I have too much separation between my subjects and any background. That said, I understand your points, and for static shots I really could live with just a good EVF, but I like to shoot action and using an MF lens with accurate results without some kind of aid for fast action is not exactly easy. Yes, I guess I had my scenarios in mind more than yours. My "focus challenge scenario" is available light portraiture and in that scenario, with thin DOF it makes a difference if you focus on the eye or on the brow - the focusing point can often overlap both, as it is not easy to keep it just over the eye if the subject is moving . This is also where focus peaking was failing - it was highlighting both the pupil and the brow and the highlight over the pupil was not easy to notice either. Originally posted by FernandoBatista BTW, just to give an idea, this is what I usually shoot: http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-KtzjLqK/0/O/2012_03_03_1.jpg http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-WvvxLrP/0/O/2011_11_27_5.jpg http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-n6GpQ7G/0/O/2011_04_23_1.jpg http://fernandobatista.smugmug.com/Galeria/Aves-Birds/i-4SgqVbv/0/O/2011_04_16_1.jpg Doing images like this with the D90 was almost as easy as any AF lens, or even easier in some points... but doing it without the help of a reliable trapfocus, no matter how good the viewfinder or peaking, is just something I'm not seeing happening. Very nice work! I never had a lens that is suitable for this scenario. Speaking of lenses, which Pentax lens would you plan to use for such work?