Of late, there had been some discussion on the best settings for autofocus with the Olympus system. There are a number of settings that can be varied, and they may have an effect of the AF accuracy and consistency.
When I had gone on a tour to North-East India, I was mainly shooting static birds. Of course, they weren’t still; they were moving a bit as birds will, but for all intents and purposes, it was a static subject. I noticed in a burst of 8-10 images, a few would be out of focus,
then a few would be just off
and a few would be spot on.
This was worrying because where there was no chance of a burst, there was very possibility of having mis-focused images. I thought I should investigate this so when I came back, I set up a test regime.
My initial endeavor was to come up with a group of settings that would be of most use for a static subject. When I say static, it means the subject isn’t moving much, like a bird. However, it should be recognized that even if the bird or insect is static, there could be some movement of the branches due to wind or your own small movements so the AF lock may change slightly.
The subject of my scrutiny was this garden ornament. I chose it because it had a high level of detail in the grassy hair so point of focus could be accurately gauged. Also the eye had a good contrast so it was the right object for focus point. I was shooting with the em1x and 300/4. To eliminate chances of motion blur, shutter speed was chosen at 1/800s (going on the basic principle of 1/focal length)
After looking through the menu, I selected a few parameters to vary, which I thought would have the most impact on the focus accuracy and consistency. I ran through 11 groups of settings and then a further 3 with lower shutter speed (1/250s) so IS would be a factor. The object was about 3-4 inches tall and it was 21 feet from me. I was sitting on a swing with both feet on the ground to give myself a slight unstable platform. I felt this was the right things to do, to simulate handheld photography while walking around.
Just to be clear, the below image is in-focus
and the below image is just off-focus. the difference is very marginal but u can see the eye area in the first image is just slightly sharper
Set 1: I started with this basic set and got a keeper rate of around 78%
| SET 1 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | OFF | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 2: on the second set, I changed lens IS priority to ON and got around 81% perfect focus. However, the images lacked crispness somehow. More images were in the borderline sharp than not, but weren’t quite unsharp…. Its hard to explain but I didn’t like the output
| SET 2 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | OFF | |
| LENS IS PRIORITY | ON | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 3: On this I set the lens IS manually to OFF. There had been discussion on one of the Indian forums that IS is actually degrading the AF performance so I manually set the IS switch on the lens to OFF. I got a keeper rate of 53% with 15% images being actually out of focus. The image also jumped around too much making the AF performance even worse. I would definitely recommend keeping IS ON for static subjects
| SET 3 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | OFF | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | OFF | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 4: I thought engaging IS when pressing the shutter might be beneficial so I put tat to O and got better results Than earlier. I think the IS engages just as you press the shutter and it further compensates for shake. The keeper rate was 85%. But the 90% focus images could just as easily have been in the sharp category. The differences were very marginal
| SET 4 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | ON | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 5: In this setting, lens IS priority was ON. While the images were stable, the images jumped around in the viewfinder too much as the camera the camera to compensate for small movements. Keeper rate was around 66%. I would recommend keeping lens IS priority OFF.
| SET 5 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | ON | |
| LENS IS PRIORITY | ON | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 6: In this, the IS was set to IS priority rather than FPS priority. I did not find anything detrimental about this. Just the keeper rate dropped slightly to 69&. Again, the images aren’t un-sharp, they are just less crisp.
| SET 6 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | IS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | OFF | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 7: In this setting the C-AF release priority was turned to on. Again, theres not much difference in the results and its kind of ok to use this one if you like. Keeper rate was 78%
| SET 7 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | ON | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | ON | |
| LENS IS PRIORITY | ON | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 8: In this setting I basically used setting 4 which I felt gave the best results so far but I used an AF sensitivity of -2. I had always felt that part of the mis-focus could be because the camera was trying to do too much. It was attempting to refocus when in fact it wasn’t really crisper and more consistent
| SET 8 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | -2 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | ON | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 9: So I also tried the other way, i.e. IS sensitivity of +2. Keeper rate was 81%. It’s a bit close to call but I think for static subjects which may move slightly like birds or insects of something, a sensitivity of 0 or -1 may be the best choice.
| SET 9 | FOCUS MODE | C-AF |
| FOCUS SENSITIVITY | 2 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | ON | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 10: Its usually recommended to use S-AF for static subjects so I tried that out too. Keeper rate was 79%. I think if u analyze the images, you realize that the first image is always spot o. Since I was sitting on an unstable platform, I was definitely moving a little. And its these small movements that never got compensated for, once focus was locked. I took a total of 29 images in 3 sets. So I refocused thrice and shot off 10 frames in each refocus. Keeper rate was lower.
| SET 10 | FOCUS MODE | S-AF |
| FOCUS SENSITIVITY | 2 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | ON | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Set 11: This was with AF+Tr. Whilst the camera was able to largely keep the eye in the square (I deliberately moved the camera so that tracking all over the frame was forced). Keeper rate dropped to 64%. One thing I noticed was that the square was too large to accurately tie-in into the eye. It was like a zone around the eye. Maybe that’s the reason for the slight mis-focus.
| SET 11 | FOCUS MODE | C-AF+TR |
| FOCUS SENSITIVITY | 0 | |
| LENS IS | ON | |
| C-AF RELEASE PRIORITY | OFF | |
| IS | FPS PRIORITY | |
| HALFWAY RELEASE SHUTTER IS | ON | |
| LENS IS PRIORITY | OFF | |
| SHUTTER SPEED | 1/800S | |
| AF SCANNER | MODE 3 |
Results Summary
| shots | in focus | 90% focus | out of focus | |
| set1 | 9 | 77.8% | 22.2% | 0.0% |
| set2 | 11 | 81.8% | 18.2% | 0.0% |
| set3 | 13 | 53.8% | 30.8% | 15.4% |
| set4 | 14 | 85.7% | 14.3% | 0.0% |
| set5 | 12 | 66.7% | 33.3% | 0.0% |
| set6 | 16 | 68.8% | 31.3% | 0.0% |
| set7 | 14 | 78.6% | 21.4% | 0.0% |
| set8 | 15 | 86.7% | 13.3% | 0.0% |
| set9 | 22 | 81.8% | 18.2% | 0.0% |
| set10 | 29 | 79.3% | 24.1% | 0.0% |
| set11 | 31 | 64.5% | 25.8% | 9.7% |
From the above analysis, I think set 4 with a AF sensitivity of -1 would be the best option for static images. A setting of -1 would give good stickiness but just enough flexibility to re-focus if the need arose. Of course, these aren’t real world images but I think I have covered most of the bases. I think maybe AF scanner mode 2 also needs to be looked into but I don’t think it will have a significant effect.
The other 3 sets I tried were at 1/250s to force IS into the equation. Results weren’t really different. All I can say is that the IS WORKS!!!!
I wish someone would use these settings and come up with their own set of numbers so we can corroborate.
3 Responses
Hi Girish,
I can say our settings match almost identically with one exception…Lens IS Priority. I do not disagree with you advice…I just don’t know exactly what the setting is. I’ll look it up in the manual.
I did shoot (EM1X 4/300 1.4X) some perched birds (low light /cloudy in a forest) and had -2…I found AF/lock up too slow (with and without Bird AI) and/or too many ‘just off’ AF images so I went the other way to +2 and left it at that for the remainder of the session with a slight improvement.
from the manual…
“Image Stabilization for IS Lenses (Lens I.S. Priority)
When using a third-party lens with an on-board image stabilizer, choose whether
image stabilization gives priority to the camera or the lens. This option has no effect
if the lens is equipped with an IS switch that can be used to enable or disable image
stabilization.”
that is what i thought “Lens IS Priority” was but after reading your advising disabling it…I second guessed my own feelings on it’s importance and/or relevance.
I suppose since the Oly 100-400 does not have Sync IS one should be inclined to test Len I.S. Priority with that lens/camera combo. I believe Robin Wong stated that he got better results (re: Oly 100-400) when using the lens’ IS. I am guessing the lens’ IS system has more ‘travel’ to better compensate for lens movement at 800mm equivalent.