PDA

View Full Version : C-AF disappointing


Tordan58
27th May 2017, 07:25 PM
Have been trying the E-M1 Mark 2 during this weekend.

First impression was great. I tested the camera with the 300/4 (also for test) and my own lenses (300/2.8 and Canon 400/5.6) and the AF-S accuracy was much, much more accurate/reliable than the E-M1. Based on these promising start I would have been tempted by the camera. However I must say I am very disappointed with the C-AF. Some 500+ shots resulted in zero keepers. Zero, all shots OOF. This in spite of good light conditions, clean background, camera acquiring and confirming focus.

I tried following:
AF points, central 9, central 5, single AF.
C-AF, C-AF + tracking.
IS-1, IS off.

So just to be sure nothing was wrong with the camera I did some final tests using S-AF and the results were as good as the first test.

Anyone else?

pdk42
27th May 2017, 07:43 PM
Have been trying the E-M1 Mark 2 during this weekend.

First impression was great. I tested the camera with the 300/4 (also for test) and my own lenses (300/2.8 and Canon 400/5.6) and the AF-S accuracy was much, much more accurate/reliable than the E-M1. Based on these promising start I would have been tempted by the camera. However I must say I am very disappointed with the C-AF. Some 500+ shots resulted in zero keepers. Zero, all shots OOF. This in spite of good light conditions, clean background, camera acquiring and confirming focus.

I tried following:
AF points, central 9, central 5, single AF.
C-AF, C-AF + tracking.
IS-1, IS off.

So just to be sure nothing was wrong with the camera I did some final tests using S-AF and the results were as good as the first test.

Anyone else?

I'm no BIF expert Torden, but I did manage to get some very sharp action shots of birds at a workshop I did recently at Bempton Cliffs in Yorkshire. You'll find my write up, with plenty of images here:

http://e-group.uk.net/forum/showthread.php?t=45094

Greytop
27th May 2017, 09:40 PM
Not my experience either I'm glad to say.
I'm finding the E-M1 MkII to be a real revelation with continuous auto focus, mostly with the 40-150 Pro (with and without the MC-14).

Examples http://e-group.uk.net/forum/showthread.php?t=44611

Phill D
27th May 2017, 09:58 PM
I was also really impressed with the C-AF shooting a fast jet at Duxford using the 40-150. I've not looked super closely yet at all my shots but there are only two or three focus misses. Didn't try the tracking, it sounded a bit of a step too far when it was explained to be honest. I'll go back and look at the suggested settings for you Tord to see if there is anything there that may help.

c12402
27th May 2017, 10:12 PM
Many of us had the same first impresion so, be confident in the machine... it's just a matter of learning curve and the right settings. There are quite a few entries in internet, and some in this site, to help you with the right settings.

I'm using now regularly the c-af for BIF with good results using s-af (also s-af/manual) but not tracking mode.

Have a look https://flic.kr/p/UC9xT2

Tordan58
27th May 2017, 10:18 PM
Here is a sample, one of the sharpest dug out from hundreds. An easy and cooperative subject, hovering still, perfect light etc and should be easy to lock AF on. As you can see there is not much in focus. And it keeps on like this for every single CAF frame. I am not too impressed.


http://www.e-group.uk.net/gallery/data/2042/P5260320.jpg

http://www.e-group.uk.net/gallery/data/2042/P5260320-2.jpg

DavyG
27th May 2017, 10:28 PM
Tord, C-AF and C-AF + Tr are different on the E-M1 II than on the E-M1.

I used C-AF + Tr most of the time on the E-M1, I only use it for subjects which are actually moving a bit on the MKII.

The E-M1 II is a great piece of kit but does require a different approach than the E-M1.

There are a few examples of static subjects and bif on my Flickr page, mostly taken with C-AF and C-AF + Tr, here's a link if you're interested:

https://www.flickr.com/photos/144249381@N05/

Dave

pault
28th May 2017, 06:43 AM
Tord
This maybe a silly point, but, when using the 300 f4 did the aperture used take you into defraction territory?

wornish
28th May 2017, 07:08 AM
I can see from your 500px gallery that you have taken some great shots using the Mk1, so I think it is a Mk2 setting issue as you say.

What shutter setting did you use. Mechanical or Silent. I found Silent Sequential Low worked best (❤︎ ❏ L). and Image stabilisation set to S-IS Auto.

Phill D
28th May 2017, 07:17 AM
Yes there was a diffraction discussion at Duxford and Drmarkf was also having trouble with that I think from his earlier post.
The settings for fast aircraft that were suggested are below but on reflection I think you are way beyond this simple starting point.
C-AF using 1,5, or 9 a/f areas,
centre weighted metering,
shutter 1/1000 or more
"sequential low" setting set to 6fps
RLS for C-AF set to off
IS also set to off. Although drmarf and beagletorque found it better to have it on IS-auto.
Probably not much help but I hope you get it sorted out.

Tordan58
28th May 2017, 07:47 AM
Tord
This maybe a silly point, but, when using the 300 f4 did the aperture used take you into defraction territory?
This is not a diffraction issue, it's an OOF issue. Pictures shot at full aperture.

pault
28th May 2017, 07:48 AM
It Drmarkf's comments and same image quality of some of my pics, that reminded me of diffraction.

Tordan58
28th May 2017, 08:00 AM
I can see from your 500px gallery that you have taken some great shots using the Mk1, so I think it is a Mk2 setting issue as you say.

What shutter setting did you use. Mechanical or Silent. I found Silent Sequential Low worked best (❤︎ ❏ L). and Image stabilisation set to S-IS Auto.
I have a hard time believing it's a setting issue, as all shots are OOF. I get the feeling there is a consisent failure in acquiring focus in C-AF.

I used mechanical shutter, H fps (from 8 to max) and IS Auto, except the end of the session when I disabled IS to see if that was the culprit - no improvement seen, though.

wornish
28th May 2017, 08:07 AM
I have a hard time believing it's a setting issue, as all shots are OOF. I get the feeling there is a consisent failure in acquiring focus in C-AF.

I used mechanical shutter, H fps (from 8 to max) and IS Auto, except the end of the session when I disabled IS to see if that was the culprit - no improvement seen, though.


I agree with you the shots are definitely out of focus and it doesn't look like motion blur.

Maybe it would be worth doing some tests with the camera mounted on a tripod and use a lens test chart as a target. Then comparing two shots one taken with S-AF and another taken using C-AF. I think eliminating as many variables as possible is the only way to track down what the cause is.

Tordan58
28th May 2017, 08:15 AM
Here is a test I did demonstrating the OOF issue.

Single AF point, positioned so that it aims at the central tip of the central leaf
Shutter release prio OFF.
Camera locked focus immediately.
1/250s, F/4

This one is with S-AF. Pretty much spot on.
http://www.e-group.uk.net/gallery/data/2042/P5270414.jpg

This one is with C-AF. Note the severe out of focus issue (front focus), a tad in front of the left leaf.
http://www.e-group.uk.net/gallery/data/2042/P5270413.jpg

wornish
28th May 2017, 08:20 AM
Here is a test I did demonstrating the OOF issue.

Single AF point, positioned so that it aims at the central tip of the central leaf
Shutter release prio OFF.
Camera locked focus immediately.
1/250s, F/4




The shots really show the issue.
One last check would be to try it with a different lens, then if thats the same I would take the body back under warranty and have it repaired.

Phill D
28th May 2017, 09:31 AM
That test looks pretty conclusive to me, as Dave said probably worth trying it on another lens to see if it's different. but it's an interesting test I'm probably going to try it myself now as you've made me curious.

wornish
28th May 2017, 10:07 AM
Just tried the comparison and I can't see any difference both focus methods were spot on.


C-AF left. S-AF Right

http://e-group.uk.net/gallery/data/2016/C-AF_v_S-AF.jpg (http://e-group.uk.net/gallery/showphoto.php/photo/93319)

IainMacD
28th May 2017, 10:21 AM
I don't know the camera, but have read somewhere that it has the facility for focus adjustment like the older 4/3 cameras (e.g. E-5) is it possible that S-AF and C-AF use different methods of focusing and the lens/camera combo need fine tuning?

Tordan58
28th May 2017, 10:57 AM
OK so I tested using a focus chart. Distance to subject something like 2 meters. (On a side note this test also shows that the camera horizontal gauge is not calibrated however it does not really impact the test result as I aimed at the middle of the horizontal line, just where it intersects the vertical line).

S-AF: Pretty much spot on.
http://www.e-group.uk.net/gallery/data/2042/P5280001.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280002.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280003.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280004.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280005.JPG

C-AF: Some spread is seen, within +/-1
http://www.e-group.uk.net/gallery/data/2042/P5280006.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280007.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280008.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280009.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280010.JPG

C-AF+TR: Spread increases, offset -3
http://www.e-group.uk.net/gallery/data/2042/P5280011.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280012.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280013.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280014.JPG
http://www.e-group.uk.net/gallery/data/2042/P5280015.JPG

c12402
28th May 2017, 10:59 AM
I was in the same doubt, and did a test using a tripod and a calibration Chart, testing all modes. Conclusión: no difference. Can you share with us the exif of the flowers' pictures? That may help in finding something...

Tordan58
28th May 2017, 11:00 AM
I don't know the camera, but have read somewhere that it has the facility for focus adjustment like the older 4/3 cameras (e.g. E-5) is it possible that S-AF and C-AF use different methods of focusing and the lens/camera combo need fine tuning?
Hi Ian,
This sound like a long shot to me... If that turned to be true it would be a bad thing.
/Tord

pdk42
28th May 2017, 11:04 AM
Hi Ian,
This sound like a long shot to me... If that turned to be true it would be a bad thing.
/Tord

Agreed - thought I'd left that behind when I ditched Canon!

In any case, if focus calibration were the issue then it would show in the S-AF shots.

c12402
28th May 2017, 11:15 AM
Hi Tordan,

Can you please do a total reset of the camera? I mean save your settings, go to factory reset and reload again.

In a number of occasions that fixed the problem.

Tordan58
28th May 2017, 11:17 AM
I was in the same doubt, and did a test using a tripod and a calibration Chart, testing all modes. Conclusión: no difference. Can you share with us the exif of the flowers' pictures? That may help in finding something...
Attached. I had to ZIP because of forum file size constraints. The ZIP contains two XML files with the EXIF.

Tordan58
28th May 2017, 11:19 AM
Hi Tordan,

Can you please do a total reset of the camera? I mean save your settings, go to factory reset and reload again.

In a number of occasions that fixed the problem.

Hi,

How do I do to save the settings, prior reset?
/Tord

c12402
28th May 2017, 11:29 AM
Using the free app Olympus viewer from Olympus. Go to menu-help-update camera. there you have three options for saving, reload, etc,..

Save first.

BTW, even if it has no effect in this test as it is static, I do recommend to set the reset lens parameter to off in all modes. I assume you also have read the local / global parameters logic in the camera, it's convenient to have in mind.

Ross the fiddler
28th May 2017, 12:20 PM
I'm assuming you've done the latest updates on lens & body then?
http://dl01.olympus-imaging.com/ww/ud2/ENU/0001/1250/index02a.html
http://dl01.olympus-imaging.com/ww/ud2/ENU/0004/1024/index02a.html

As found here. http://cs.olympus-imaging.jp/en/support/imsg/digicamera/download/software/firm/e1/

Miketoll
28th May 2017, 01:23 PM
"H fps (from 8 to max)"

My understanding is that for C-AF on the mkii you have to have L fps set. In high only the first frame is focussed. Still does not explain the oof flower though.

GyRob
28th May 2017, 04:24 PM
I'm finding that S-AF with 1 focus point is just about spot on every time but on C-AF with 1 focus point and shooting say a 20 frame burst on say a duck in flight you get some sharp then slightly soft then back to sharp again tracking +2 fast .
I'm SURE It may well be grabbing the wing tip at times or I move off the bird just a little but I am finding it is more often Me that is at fault .
I just did a static test tripod, gimbal 3200sec f4 iso 400 IS on
70 shot burst on a lamp post S-af 1 focus point all sharp but 3 a little soft IS on

same but C-AF about 5 were just a little soft IS on
same again but with IS off about 7 just a little soft
All silent electronic shutter Low setting .
When using more focus point's thing's can vary quite a lot but for birding (BIF ) you often have to and just risk the right point get the head .
My test's today have made me even more happy with the camera / lens Yes it could be a little better But we will always want that :)
Rob.

Tordan58
28th May 2017, 04:49 PM
Camera and lens were updated with the latest FW since the first day.

Next I did a full camera reset, and gave it a last try this afternoon, a few hundred more frames. Still, zero(!) subjects in focus, the camera failed even in the easiest situations. Strange since the acquisition feels snappy and the camera seems to re-acquire focus when lost, even is background is somewhat busy (easiest with 5 AF points). However, there is an OOF issue.

It that seems many of you are happy with the C-AF performance, so my conclusion is that this copy of the camera is a piece of rubbish (sorry for being harsh but that is exactly how I feel). Product was new when I started testing it in case you wonder. How could it have passed the quality check?

GyRob
28th May 2017, 04:55 PM
Just did the same test with the 1.4tc results about the same and I see no loss of sharpness with the 1.4 either .
Rob.

Walti
28th May 2017, 06:13 PM
Camera and lens were updated with the latest FW since the first day.

Next I did a full camera reset, and gave it a last try this afternoon, a few hundred more frames. Still, zero(!) subjects in focus, the camera failed even in the easiest situations. Strange since the acquisition feels snappy and the camera seems to re-acquire focus when lost, even is background is somewhat busy (easiest with 5 AF points). However, there is an OOF issue.

It that seems many of you are happy with the C-AF performance, so my conclusion is that this copy of the camera is a piece of rubbish (sorry for being harsh but that is exactly how I feel). Product was new when I started testing it in case you wonder. How could it have passed the quality check?

Tordan,

I found that my "out of the box" first use was a similar experience to yours, and ended up re-setting the camera (full reset) twice, then it jumped into working properly, I put the down to the camera having been played with by the dealer (don't know if that was true) but, I had to do multiple resets before it started working properly! NB when doing a reset you have to power off and back on with slight delays without touching ANYTHING, it really seems that the reset needs the power off without any input being made to take properly...

Give it a go - it may make difference it seemed to for me!

Tordan58
28th May 2017, 07:36 PM
Thanks for advise, but is not the game I want to play. Image buying a new car, collect it from the dealer... and you find out the engine refuses to revolve above 1200 rpm except in first gear. To get the car to deliver what it is capable of you need to find how to disconnect the fuel injection control box from the DC supply. Then disconnect it, reconnect it and disconnect a few times and then hopefully...

Miketoll
28th May 2017, 07:49 PM
This an interesting link about the abilities of the EM1 mkII versus other cameras for wildlife and BIF. While not up to the likes of a Nikon D500 it does pretty well.
http://mirrorlesscomparison.com/best/mirrorless-cameras-for-wildlife-and-bird-photography/

wornish
28th May 2017, 09:14 PM
Thanks for advise, but is not the game I want to play. Image buying a new car, collect it from the dealer... and you find out the engine refuses to revolve above 1200 rpm except in first gear. To get the car to deliver what it is capable of you need to find how to disconnect the fuel injection control box from the DC supply. Then disconnect it, reconnect it and disconnect a few times and then hopefully...

I said this in an earlier post but will repeat.

Take it back and ask them to fix it or replace it, thats what warranty is all about.

I think you have more than enough proof it is not working to spec.

Olybirder
28th May 2017, 10:01 PM
Hi Tord. In your original post you say that you have been trying the E-M1 Mark 2 this weekend. Does that mean that you haven't actually bought it at the moment? If that is the case, I think you should return it and try some of the alternatives to see how you get on with them. From the gear you already own and use, it sounds as if weight and size are not an issue for you. A dedicated action camera such as the Nikon D500 or Canon 7D II might suit your requirements better.

I bought an E-M1 MkII on Thursday as it seems to be the best option for me. I looked at the Nikon and Canon offerings while I was in the store but they are much to big and heavy for me to use. They are not an option for me at all.

I haven't had a chance to use the MkII much yet and I am sure it is not set up properly at the moment. However, my first attempts at action shots were reasonably encouraging. I am sure they can be improved upon when I know what I am doing. However, perhaps I am not the most discerning of users and am easily pleased. As a new owner your post was not what I wanted to hear but, as I explained, there are not any viable alternatives for me. Your case is different though. If you are not happy with the camera, I would return it and try something else. As you say, you shouldn't have to work so hard to get it working for you.

Ron

pdk42
28th May 2017, 10:35 PM
Unless I'm missing something, the AF setup on the mkii isn't really all that complicated. If we ignore Tracking mode (and most people seem to think that is a good idea), there really are only two C-AF things that can be fiddled with (ignoring basic stuff like focus point number and position):

- C-AF Lock - sets how quickly the C-AF will re-focus when it detects the subject moves from in-focus to out-of-focus. The setting is variable from -2 to +2. +2 is "loose" meaning that it'll refocus rapidly (i.e. loose lock on the subject). -2 is "tight" meaning that it'll wait a little while before refocusing (i.e. tight lock on the subject). I would guess that for BIF where there is little chance of foreground intervention and where you are confident of keeping the bird in the focus zone that +2 is probably the best option. I can see -2 being better for sports where the rate of change is lower and where people passing in front of the subject will happen more often.

- AF-Scanner - sets how aggressively the AF runs and re-runs if it can't find focus. Seems from my testing that it only really makes any difference in extremely low lighting levels. I doubt it makes much difference to BIF.

So - I don't think that Tord's issues are to do with camera setup. In fact, my best guess is that there's some sort of lens issue. From my experience with both the 12-100 and 300 f4 the AF is pretty accurate most of the time, even for CAF on moving subjects.

Interestingly, my 50-200 (non SWD) is much worse than either of these lenses and pretty often misses focus.

Bengeo
28th May 2017, 11:52 PM
Tord,

Did you see Mike's comment here: http://e-group.uk.net/forum/showpost.php?p=415611&postcount=29

He makes an important point about the fps setting which is different from the Mark I.

Ross the fiddler
29th May 2017, 12:10 AM
Camera and lens were updated with the latest FW since the first day.

Next I did a full camera reset, and gave it a last try this afternoon, a few hundred more frames. Still, zero(!) subjects in focus, the camera failed even in the easiest situations. Strange since the acquisition feels snappy and the camera seems to re-acquire focus when lost, even is background is somewhat busy (easiest with 5 AF points). However, there is an OOF issue.

It that seems many of you are happy with the C-AF performance, so my conclusion is that this copy of the camera is a piece of rubbish (sorry for being harsh but that is exactly how I feel). Product was new when I started testing it in case you wonder. How could it have passed the quality check?

When you review the images in camera, thumbnail size, it will show where the AF point is located. Are those AF points showing where you intended?

GyRob
29th May 2017, 08:23 AM
The focus square is a lot bigger than indicated and a little movement such as pressing the shutter can make the camera grab onto something even though the center of the green square is right on the subject .
Canon is similar that is why they let you be able to make the focus square smaller for critical work .
Rob.

GyRob
29th May 2017, 08:38 AM
Edit. by the way my test subject was a lamp post with writing on it about 50yds away also when I get new gear like the 300f4 I set a cereal box 50 to 60 ft away in the garden as its fine print and graphics clearly show sharpness in a real world test.

Tordan58
29th May 2017, 08:53 AM
Hi Tord. In your original post you say that you have been trying the E-M1 Mark 2 this weekend. Does that mean that you haven't actually bought it at the moment? If that is the case, I think you should return it and try some of the alternatives to see how you get on with them. From the gear you already own and use, it sounds as if weight and size are not an issue for you. A dedicated action camera such as the Nikon D500 or Canon 7D II might suit your requirements better.

I bought an E-M1 MkII on Thursday as it seems to be the best option for me. I looked at the Nikon and Canon offerings while I was in the store but they are much to big and heavy for me to use. They are not an option for me at all.

I haven't had a chance to use the MkII much yet and I am sure it is not set up properly at the moment. However, my first attempts at action shots were reasonably encouraging. I am sure they can be improved upon when I know what I am doing. However, perhaps I am not the most discerning of users and am easily pleased. As a new owner your post was not what I wanted to hear but, as I explained, there are not any viable alternatives for me. Your case is different though. If you are not happy with the camera, I would return it and try something else. As you say, you shouldn't have to work so hard to get it working for you.

Ron
Hi Ron
I was intending to buy the Mark II and before doing so got the opportunity to test a copy from my local dealer as, I am a loyal returning customer. That copy was brand new and intended to be used for display and demo purpose afterwards. Returned the copy this morning explained what happened, no obligation. Have contacted Olymupus support who have offered to inspect the samples to see what has gone wrong.

Will keep you updated.

Tordan58
29th May 2017, 09:05 AM
"H fps (from 8 to max)"

My understanding is that for C-AF on the mkii you have to have L fps set. In high only the first frame is focussed. Still does not explain the oof flower though.

If that was correct then at least some of the first frames would have been in focus. I got zero (sic) frames in focus, out of 500+.

TonyR
29th May 2017, 10:55 AM
If that was correct then at least some of the first frames would have been in focus. I got zero (sic) frames in focus, out of 500+.

No, I don't think so. If Rls Priority-C was set to the default On, then the camera starts shooting before focus is achieved. So, if you were using Sequnetial-H mode, your results are as I would expect. You must use Sequential-L to get C-AF with burst shooting.

It can do this:

http://e-group.uk.net/gallery/data/500/P2028335_s.jpg (http://e-group.uk.net/gallery/showphoto.php/photo/93339)

Tordan58
29th May 2017, 11:30 AM
No, I don't think so. If Rls Priority-C was set to the default On, then the camera starts shooting before focus is achieved. So, if you were using Sequnetial-H mode, your results are as I would expect. You must use Sequential-L to get C-AF with burst shooting.

It can do this:

http://e-group.uk.net/gallery/data/500/P2028335_s.jpg (http://e-group.uk.net/gallery/showphoto.php/photo/93339)

I had shutter release prio set to Off from the beginning.

drmarkf
29th May 2017, 12:14 PM
I'm almost certain CAF is actually related to the actual frame rate you choose, not whether it is set to L or H.

So, if you limit H to, say, 15fps you'll get CAF between each frame. That's certainly how mine behaves, anyway.

TonyR
29th May 2017, 12:51 PM
I'm almost certain CAF is actually related to the actual frame rate you choose, not whether it is set to L or H.

So, if you limit H to, say, 15fps you'll get CAF between each frame. That's certainly how mine behaves, anyway.

Look at Page 46 of the manual.

Sequential High -

"Pictures are taken at up to about 15 frames per second (fps)
while the shutter button is pressed all the way down. Focus,
exposure, and white balance are fi xed at the values for the
fi rst shot in each series."

We are talking about the E-M1 MkII right?

GyRob
29th May 2017, 12:58 PM
I agree with TonyR I always use Low for CAF .
Rob.

Tordan58
29th May 2017, 01:53 PM
Then H FPS and CAF would be meaningless when used together. And the statement regarding CAF on page 43 is contradictory with page 123. I use mode 2 most of the time, sometimes mode 1 depending on light angle and background.

And still does not explain why I get zero frames in focus.

drmarkf
29th May 2017, 02:55 PM
Look at Page 46 of the manual.

Sequential High -

"Pictures are taken at up to about 15 frames per second (fps)
while the shutter button is pressed all the way down. Focus,
exposure, and white balance are fi xed at the values for the
fi rst shot in each series."

We are talking about the E-M1 MkII right?

With mechanical shutter, yes.

Where does the manual talk about these issues with the electronic shutter?

I have been using L for ages (too many images with H to deal with!) so I'm relying on memory, but I still believe it will CAF at 18fps with the electronic shutter and I thought you could only reach that on H?

TonyR
29th May 2017, 03:09 PM
With mechanical shutter, yes.

Where does the manual talk about these issues with the electronic shutter?

I have been using L for ages (too many images with H to deal with!) so I'm relying on memory, but I still believe it will CAF at 18fps with the electronic shutter and I thought you could only reach that on H?

No, 18 fps is the highest burst rate in Sequnetial-L with silent shutter. The manual omits to specifically talk about how AF works with Sequential-H and silent shutter but both H with mechanical and Pro Capture-H say the AF reverts to S-AF if C-AF is set.

Also, I just tested Sequential-H, silent shutter at 15 fps and it only focuses once at the beginning.

pdk42
29th May 2017, 03:10 PM
With mechanical shutter, yes.

Where does the manual talk about these issues with the electronic shutter?

I have been using L for ages (too many images with H to deal with!) so I'm relying on memory, but I still believe it will CAF at 18fps with the electronic shutter and I thought you could only reach that on H?

I used Seq low at 18fps on all the shots on Bempton and for sure CAF was active all the time.

birdboy
29th May 2017, 03:20 PM
Hi Tord I would not have thought that your experience with CAF would have been a setting issue given the images that you post. You are a season Olympus pro. I am happy with my copy of the MKII and 300 F4 but I only use drive modes with antishock (diamond) and LF with CAF. At first I thought this may have been the issue but your static CAF tests shows a front focusing issue.
Do you think the image you show is an oof issue or a shutter shock issue?
Which drive modes were you using? Will be interested what comes back from Olympus support.

drmarkf
29th May 2017, 03:24 PM
No, 18 fps is the highest burst rate in Sequnetial-L with silent shutter. The manual omits to specifically talk about how AF works with Sequential-H and silent shutter but both H with mechanical and Pro Capture-H say the AF reverts to S-AF if C-AF is set.

Also, I just tested Sequential-H, silent shutter at 15 fps and it only focuses once at the beginning.

Thanks - I'll correct my brain should I ever need to use those speeds!

TonyR
29th May 2017, 03:26 PM
Then H FPS and CAF would be meaningless when used together. And the statement regarding CAF on page 43 is contradictory with page 123. I use mode 2 most of the time, sometimes mode 1 depending on light angle and background.

And still does not explain why I get zero frames in focus.

Yes, Sequential-H and C-AF is meaningless when used together. But that is why the manual says the camera reverts to S-AF when you set both of these. The question is, were you set to any (silent, normal or Pro capture) of the Sequential-H settings? If you were you would probably not get any sharp shots of a moving subject like a bird.

The manual is cr@p ;)! We have to deduce what it means and what the camera really does. The stuff on page 43 assumes you have used the default settings. C-AF starts when you half-press the shutter. The table on page 123 is very confusing but sort of works if you assume that the photographer is using single shot and not sequnetial. So, the AF starts continuously focusing when you half-press the shutter and stops (or locks) when you fully press it, in mode 1 or 2. In sequential shooting it only "locks" momentarily to take the shot before continuing to C-AF between shots (and then only if you are in an L mode).

Anyway, as a long-time E-M1 MkI and MkII user, I can assure you that the MkII must be in sequential-L to use C-AF (unlike the MkI after firmware update 3.0!). The results are not perfect but you can get a number of well-focused shots from a burst. On static subjects, you will see a variety of sharpness as the AF hunts around. I don't use C-AF for static subjects, ever.

TonyR
29th May 2017, 03:41 PM
Hi Tord I would not have thought that your experience with CAF would have been a setting issue given the images that you post. You are a season Olympus pro. I am happy with my copy of the MKII and 300 F4 but I only use drive modes with antishock (diamond) and LF with CAF. At first I thought this may have been the issue but your static CAF tests shows a front focusing issue.
Do you think the image you show is an oof issue or a shutter shock issue?
Which drive modes were you using? Will be interested what comes back from Olympus support.

I suspect that the front-focusing issue with C-AF comes from the way that S-AF and C-AF differ in operation. The following is mostly a guess based on a few snippets from Olympus slides and web sites.

S-AF uses hybrid PDAF and CDAF so it probably uses CDAF to fine-tune the focus. CDAF favours the highest contrast part of the subject under the focus point rather than the closest part.

C-AF is looking for movement and is much more dependent on PDAF (see https://www.olympus.com.au/Products/Interchangeable-Lens-Cameras/Olympus-OM-D/E-M1-Mark-II/Features/High-speed-Performance - High-speed AF system) so it may be less accurate and it may have an algorithm that hunts back and forth a little (I think it does this more with AF Lock set to higher values). The PDAF system is designed to favour the nearest subject which is something that CDAF cannot do. So, when you use C-AF on the focus target used above it may pick out the part of the subject that is nearer than S-AF would. So, on average, with this target, you may get front-focus.

Tordan58
29th May 2017, 10:08 PM
When you review the images in camera, thumbnail size, it will show where the AF point is located. Are those AF points showing where you intended?
I wish I could check this however don't have the camera any longer. I still have the files (RAW). Is there a way to find out using Olympus Viewer?

TonyR
29th May 2017, 10:17 PM
I have a hard time believing it's a setting issue, as all shots are OOF. I get the feeling there is a consisent failure in acquiring focus in C-AF.

I used mechanical shutter, H fps (from 8 to max) and IS Auto, except the end of the session when I disabled IS to see if that was the culprit - no improvement seen, though.

I just noticed that you said this in an earlier post. Mystery solved. It will not be doing C-AF.

It is pretty stupid of Olympus to have these arcane, complex rules. Perhaps if the AF displayed in the viewfinder was actually S-AF when you use Sequential-H it would be better. Or some message gets displayed when you choose C-AF and Sequential-H. Anyway, it is poor user interface design. It would be easy to miss in the manual and lots of people expect to be able to use a camera without reading it.

Tordan58
29th May 2017, 10:32 PM
OK so reviewed the pictures once more and focused on the first frame in sequence. There are something like 1000 of pictures, time and patience are limited so cannot inspect every single one. Most are crap anyway.

I found one exception. It's not perfect but pretty accurate from focus point of view. I also know for sure that the E-M1 would have nailed it, so it's not a tremendous achievement considering the Mark2 performance on paper.
http://www.e-group.uk.net/gallery/data/2042/P5280136.jpg

Then more samples showing the mediocre results that the C-AF (non TR) produces. The Tern is progressing slowly against a steady head wind, I fired a burst hoping to capture a nice dive.
http://www.e-group.uk.net/gallery/data/2042/P5280191.jpg

And one showing how bad it gets with C-AF + TR. I tracked the duck and fired a burst as it had slowed down almost to a halt, expecting to freeze some nice splashes.
http://www.e-group.uk.net/gallery/data/2042/P5270097.jpg

Both have clean background, good contrast, subject occupies a large enough part of the frame, all conditions met for the AF to acquire focus instantly. Which it did, and still not capable of focusing correctly.

Can anyone explain what is going on?

Tordan58
29th May 2017, 10:49 PM
I just noticed that you said this in an earlier post. Mystery solved. It will not be doing C-AF.

It is pretty stupid of Olympus to have these arcane, complex rules. Perhaps if the AF displayed in the viewfinder was actually S-AF when you use Sequential-H it would be better. Or some message gets displayed when you choose C-AF and Sequential-H. Anyway, it is poor user interface design. It would be easy to miss in the manual and lots of people expect to be able to use a camera without reading it.
Hi Tony,
Thanks for input but I am not sure I follow what you write.
I am 100% certain that the camera did focus continuously when in H-FPS, focus was maintained while tracking subjects with shutter release button half-pressed. Clearly visible when shooting against sky. The issue is that I do not get what I see - at shutter release focus is altered, with one exception (see previous post). See also the focus chart samples earlier in this thread.

/Tord

Tordan58
29th May 2017, 10:52 PM
Hi Tord I would not have thought that your experience with CAF would have been a setting issue given the images that you post. You are a season Olympus pro. I am happy with my copy of the MKII and 300 F4 but I only use drive modes with antishock (diamond) and LF with CAF. At first I thought this may have been the issue but your static CAF tests shows a front focusing issue.
Do you think the image you show is an oof issue or a shutter shock issue?
Which drive modes were you using? Will be interested what comes back from Olympus support.
It can't be shutter shock issue at 1/800s. (I could have added that piece of information).
What do you mean with "drive mode"? Memory card? I use one card only.

Just shared the files with Olympus support. Can't wait to get their reply.
/Tord

TonyR
29th May 2017, 11:02 PM
Hi Tony,
Thanks for input but I am not sure I follow what you write.
I am 100% certain that the camera did focus continuously when in H-FPS, focus was maintained while tracking subjects with shutter release button half-pressed. Clearly visible when shooting against sky. The issue is that I do not get what I see - at shutter release focus is altered, with one exception (see previous post). See also the focus chart samples earlier in this thread.

/Tord


Ah-ha! I see what you're saying, and I have just given it a try with my camera. I have been sloopy with my terminology! When I said Sequential-H uses S-AF I am wrong :o. What actually happens is that C-AF behaves differently with Sequential-H (in a way a bit similar to S-AF :)).

The manual says for Sequential-H "Focus, exposure, and white balance are fixed at the values for the first shot in each series." I translated that as S-AF but of course, as you point out, it does do C-AF while the shutter button is half-pressed. It just stops doing C-AF after the first shot in the series!

Sorry for the confusion. The result is pretty much the same and the reason that your images are not focused is that the camera stops focusing after the first shot. The first shot in a burst is often not the best.

Hope that helps. The answer is to use Sequential-L of any variety - normal, silent, Pro Capture.

birdboy
31st May 2017, 03:09 PM
It can't be shutter shock issue at 1/800s.

That's what I thought at first but cannot find any oly info qualifying what shutter speeds this applies to p47 manual states;

"To prevent camera shake caused by the small vibrations that occur during shutter operations, shooting is performed using an electronic front-curtain shutter."
The info display on the back of the camera states "When shutter speed is 1/320 or less, the first shutter curtain is switched to electronic. Select the post-shutter release delay time." I do not under stand what this really means.

If it was ineffective at above 1/320sec I would have expected the antishock mode to be greyed out when set above this speed if it wasn't needed. All I can say is that I have noticed an improvement in the appearance of sharpness of the image using the 0sec anti shock mode compared to the normal mode. I have had so many bad images using the normal drive modes that I would question what are they good for?


What do you mean with "drive mode"?

Its not very clear in the manual but I have often seen the setting that sets single, Sequential High etc as the drive mode, p 46 manual show some. There are 15 drive modes on the super control panel p50 marked as 19 Sequential shooting/Self-timer. I do not think this helps as that mode also includes single, Anti-Shock(diamond),Silent (heart).ProCapH, ProCapL & High Res Shot. Drive Mode is a term that has been used on E5 for sure in David Busch's book. I do not know of a better way of identifying that group of selections.

Tordan58
1st June 2017, 10:57 PM
I was able to test the E-M1 Mark 2 body once more (last) time after explaining a possible reason behind my earlier disappointment.

I used L-FPS this time, and indeed it seems that you are right in that C-AF does not work with H-FPS. A quick look at the results indicates that some are quite good. Will take some time to develop and evaluate, and when done I will share some photos later for you to feedback.

I have a quick question: is there an easy way to find out which AF point that was actually used for each frame (and ideally visualize it in Lightroom)?

Thanks
Tord

mik
2nd June 2017, 02:49 AM
Will this one work

https://petapixel.com/2014/09/19/simple-lightroom-plugin-shows-you-what-focus-points-your-camera-used-when-you-pressed-the-shutter/

wornish
2nd June 2017, 07:13 AM
The original one does not support Olympus and is no longer available but there is a link further down the page linked below that takes you to a new one that does support the EM1mk2.

https://petapixel.com/2017/01/03/free-lightroom-plugin-reveals-active-focus-point-photos/


The only frustrating thing is that you have to close it and reopen again for each image, you can't simply step through your images.

Final point it only works on Mac

TonyR
2nd June 2017, 07:53 AM
I was able to test the E-M1 Mark 2 body once more (last) time after explaining a possible reason behind my earlier disappointment.

I used L-FPS this time, and indeed it seems that you are right in that C-AF does not work with H-FPS. A quick look at the results indicates that some are quite good. Will take some time to develop and evaluate, and when done I will share some photos later for you to feedback.

I have a quick question: is there an easy way to find out which AF point that was actually used for each frame (and ideally visualize it in Lightroom)?

Thanks

Tord

Good. Glad you are getting somewhere. It is crazy of Olympus to allow such confusion to be possible with their cameras. They must lose loads of business when people try out their cameras in the shop.

birdboy
2nd June 2017, 09:56 AM
I have a quick question: is there an easy way to find out which AF point that was actually used for each frame (and ideally visualize it in Lightroom)?

Thanks
Tord

The quick way is in camera. Go to view and press info button until it shows the display with the histogram. The focus point should show as a small green square.

Tordan58
4th June 2017, 11:43 PM
Below are some samples showing what the camera is capable of:
(already posted in other thread, posting here as well)



C-AF (non-TR)
L-FPS
5 central AF points.
C-AF Release prio OFF
C-AF lock normal (0 on the scale -2 ... +2)
These are with the m.Zuiko 300 at F/4.


These birds are really swift and the time from they start diving until they emerge and take off is really short . With FPS set at 10 you will get maybe 5 frames from point of impact until subject is out of frame even when doing your best to track them. Typically the camera would produce 2 frames with correct focus, at some occasions it managed to keep focus throughout the sequence. All sequences, half a dozen of them, had at least one good frame. Quite impressive I must say.

Still I need to figure out how to make C-AF + TR work the way I imagine it should work - lock the target and keep it in focus.

http://www.e-group.uk.net/gallery/data/1167/P5260418.jpg

http://www.e-group.uk.net/gallery/data/1167/P6010542.jpg

http://www.e-group.uk.net/gallery/data/1167/P6010248.jpg

TonyR
5th June 2017, 07:45 AM
Below are some samples showing what the camera is capable of:
(already posted in other thread, posting here as well)



C-AF (non-TR)
L-FPS
5 central AF points.
C-AF Release prio OFF
C-AF lock normal (0 on the scale -2 ... +2)
These are with the m.Zuiko 300 at F/4.


These birds are really swift and the time from they start diving until they emerge and take off is really short . With FPS set at 10 you will get maybe 5 frames from point of impact until subject is out of frame even when doing your best to track them. Typically the camera would produce 2 frames with correct focus, at some occasions it managed to keep focus throughout the sequence. All sequences, half a dozen of them, had at least one good frame. Quite impressive I must say.

Still I need to figure out how to make C-AF + TR work the way I imagine it should work - lock the target and keep it in focus.

...





Bravo! But save yourself the effort with C-AF + Tr. It isn't reliable. I never use it. Fo r bif againsta clear sky, All-points with cluster display works very well and tracks the subject far better. If the background is cluttered, you can sometimes use the focus limiter to isolate the subject from the foreground and background whilst using all points.

Tordan58
5th June 2017, 01:57 PM
Hi Tony,

This AF limiter feature sounds quite useful for this application area. Will try.
Also just found the AF Target display, well hidden/disguised as "ON2".

Thanks
Tord

TonyR
5th June 2017, 03:31 PM
Hi Tony,

This AF limiter feature sounds quite useful for this application area. Will try.
Also just found the AF Target display, well hidden/disguised as "ON2".

Thanks
Tord

I have set my AF limiter up as a function button. By default, when I press it, it limits focus to to between 10m and 70m. This is for use with the 300 f/4 with/without teleconverter. If the bird is closer than 10m, I will probably not be able to follow it. If the bird is more than 70m away, I probably won't want the photo anyway! For a "record" shot, I just press the function button to revert to the full focus range. Note that the focus limiter switch on the lens has to be in the central position (ie full focus range) for any of this to work.

I use the focus limiter with 5, 9 and all points as it helps to stop the AF wandering off when there is something else in the frame and/or I drift of the subject. The limiter is a bit fuzzy in that the minimum and maximum focus distances are not exact but, with a bit of experimentation, it can be a useful addition.

When I press the function button the first time, the EVF displays "C-AF Limit" and when I press it again, it displays just "C-AF" so I always know where I am.

Hope that helps!

drmarkf
5th June 2017, 03:45 PM
Hi Tony,

This AF limiter feature sounds quite useful for this application area. Will try.


Yes, good idea.

However...
...a trouble I foresee with this in some real-life situations is that you'd be happily waiting with your AF limiter set for those pretty Pochards to fly in front of the reeds, when you suddenly spotted 4 bitterns, 2 water rails and a decoration of reed buntings peering out from among the reedy background for 10 seconds (before disappearing forever).

Cue frantic scrabble with the camera controls to allow it to focus on them (although you might think to have something suitable set up on a 'C' or button, I suppose).

[I see that Tony has already provided one version of a solution, but it is another complication to remember, especially for a camera used for multiple types of photography!)