• Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.

    Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
ZEISS DTI thermal imaging cameras. For more discoveries at night, and during the day.

Micro Four-Thirds (1 Viewer)

Tord,
I don't see any focus change in the shots you posted. The bird moved and that is deceptive. In the second one, the tip of the beak is not in focus, but the birds left eye is. Look at the green wire in the upper left corner. Same on both shots.
You can't test lenses like that. It has to be done systematically to get any sort of reliable information. Easy enough to do.
 
I tend to go with Dan on this. Tord and I should not rely on bird photos but on a 45 degree ruler to test front/back focus.

If the f/b focus test confirms the problem, then you must find the culprit. IMO, it is not that difficult if you have the required parts:

1- Test the body with another M43 lens
2- Test the lens on a Canon camera
3- Test the body and lens with another Metabones adapter

Finding an EM-1 and a Canon camera shouldn't be too difficult. Finding a MB adapter will be a problem... At least, do the first 2 tests to eliminate the camera and the lens.

If the first 2 tests are OK, then the problem has to be with the MB adapter: it is either defective or incompatible with the 400mm lens.

You seem to think that the play between the components may be causing the problem. Make sure that you test the play between the lens and camera when you do the first 2 tests. If they are normal, then the play is caused by the MB adapter.

Since Dan and I also have the same setup, we should also make a f/b focus test to confirm that it focuses correctly. I'll try to do mine today if I can find a way to do it inside - snow today and tomorrow.
 
Keep in mind that in a burst, the first is almost always sharper due to vibrations within the camera, unless you are using silent shutter or VERY high shutter speeds. Also, any movement in either camera or subject is not noticed because of the blackouts. A little bird hears the first shot and reacts, as seems to be the case in the example above.
In S-AF, the body normally locks the focus on the first shot. This has been also confirmed by my Oly contacts in Germany.
Here is my very simple, slightly crude but no less effective, home made setup:
P1301629.JPG
The shot is un-cropped. The small single focus point is right on the focus target. The line on the long ruler to the side is lined up (more or less ;)) with the plane of the target. Distance is about 10 meters.
The first thing to do is see if the AF is accurate with single shot, either silent or AS=0, on a tripod and released remotely, lens wide open. If it has been ascertained that the focus is spot on, (I would not expect otherwise with the on-chip AF), with micro adjustment if need be, then do a check to see if there is any change in the focus in burst.
There are so many factors that can blur shots, they have to be ruled out one by one to find the true cause of the problem.
Mine scores 10 out of 10 in single, re-focusing every time, and in burst. I would be very much surprised if yours were different.
Also, all my other lenses, FT or mFT are equally accurate. The E-M1 AF might be a bit slow, but it IS accurate!
 
Last edited:
I tested the 400mm/MB focus this afternoon and here are the results:

Settings: ISO 800, f/5.7, 1/125 s., electronic shutter, remote using smartphone, distance about 8 m.. No processing except for CA removal, cropping and resizing. Manual de-focus between shots and half-press for focus.

1- Single shots: the focus is not stable from shot to shot. It varies around the perfect focus mark by a good margin, about +/- 3 cm. On average, I estimate it back focuses by about 1 cm.

2- 5 shot bursts: low speed, 5 fps. The first shot can be off by a good margin front or back like above, but the remaining 4 photos are exactly similar to the first one.

Attached are 2 examples of single shots. 1 unit is about 12mm.

:C It is worse than I thought :C
 

Attachments

  • 160131ab001-2.jpg
    160131ab001-2.jpg
    135.8 KB · Views: 225
  • 160131ab007.jpg
    160131ab007.jpg
    137.9 KB · Views: 240
  • 160131ab003.jpg
    160131ab003.jpg
    121.8 KB · Views: 190
What is your focus target, the line on the paper?
If so, then no wonder! The PDAF points on the sensor of the E-M1 are not cross type and are notoriously insensitive to horizontal lines. Without a good focus target set up square with the sensor, roughly 45° to the scale, the E-M1 has no chance. Try the same test you did here with the camera in portrait or at a 45° angle and see the difference.
Even a DSLR with cross type AF points would have trouble with this setup. The AF points are larger than one would expect, and they cannot second guess your intentions.
 
Last edited:
What is your focus target, the line on the paper?
If so, then no wonder! The PDAF points on the sensor of the E-M1 are not cross type and are notoriously insensitive to horizontal lines. Without a good focus target set up square with the sensor, roughly 45° to the scale, the E-M1 has no chance. Try the same test you did here with the camera in portrait or at a 45° angle and see the difference.
Even a DSLR with cross type AF points would have trouble with this setup. The AF points are larger than one would expect, and they cannot second guess your intentions.


Come on... Dan ! Don't you think I know more. I read all posts in this thread and I even used a target you suggested yourself:
http://petapixel.com/2013/03/12/ghet...enabled-dslrs/

The target is exactly at a 45 degree angle and its center is exactly at the same height as the camera. Spot focus, single large AF point.
 

Attachments

  • 160131ab003.jpg
    160131ab003.jpg
    138.4 KB · Views: 232
Sorry, but if you don't explain your test procedure, what are we to believe?


IMO, the procedure was clearly detailed in the post: "Settings: ISO 800, f/5.7, 1/125 s., electronic shutter, remote using smartphone, distance about 8 m.. No processing except for CA removal, cropping and resizing. Manual de-focus between shots and half-press for focus." The only omission was the target.
 
But that is in this case a big omission as there are "focus charts" out there that basically look like the samples you posted and they are guaranteed to give inaccurate focus. How is anyone to know you were in fact using a proper target?
Maybe try again in better light, though 1/100-125 at ISO 800 should be enough. I checked mine briefly this afternoon inside at 1/15, ISO 800, and though it was naturally slower than normal it still nailed the focus. Outside I was also getting 1/100-125 at 800 on random objects and it was border line of what is comfortable for a 400mm lens. Mostly it worked fine but it was balky on very low contrast objects that it has no problem with in better light. Still, I have gotten plenty of very sharp shots with the 400+1.4 Ext III at 1/50, ISO 400. It depends so much on the subject mater.
Also try with single point small and focus priority on and off and see if there is a difference. There has got to be a reason for what your are observing and testing is the only way to narrow it down.
 
E-m5 + Mb + 400/5.6

I did some quick tests today of the MB + 400/5.6 on the E-M5. Metabones don't mention this camera as supported (it uses contrast detect AF only). Still, I found it performed quite OK.

The AF speed was surprisingly fast, as long as hunting was not the case. Much faster than the E-M5 with 4/3 lenses, and almost as fast as when using the MB+400/6.6 on the E-M1.

The setup is however prone to deliver back-focused photos if you rely on one picture only. If you take a few picture and re-focus for each it will find focus though, as shown on these samples taken in crappy light.
 

Attachments

  • P2140022.jpg
    P2140022.jpg
    472.7 KB · Views: 235
  • P2140025.jpg
    P2140025.jpg
    475.5 KB · Views: 243
For bad light they are good. Good light is a prerequisite for fast and accurate AF, and when the light is not so good, the E-M1 should out perform the E-M5. It is surprising to me that the E-M5 works as well on as it does.
I don't see a back focusing problem here, or are these not examples of what you were seeing? If you look at the bark on the branch, which has not moved in relation to the bird, which has moved, you can see where the plane of focus lies. Also look at the feet.
It is hard for the focus points to know what you want in focus, and they will always look for sharp contours, and I can see the speckled texture on the branch being easier to grab than the fuzzy feathers.
I wonder sometimes about how exact the size of the AF box in the EVF is. I would assume very accurate, but I don't know. I know that on a lot of DSLRs, the actual point is quite a bit bigger than the box and can lead to problems. I like to have it small and accurate for those little guys.
 
These samples are with correct focus. I can upload samples demonstrating back-focus if you want.
 
Last edited:
No, no need. I believe you!;)
The question is, how does it believe in better light in a controlled test.

I am actually surprised. I was under the impression that the Metabones only worked with phase detect. Pleasant surprise.
 
No, no need. I believe you!;)
The question is, how does it believe in better light in a controlled test.

I am actually surprised. I was under the impression that the Metabones only worked with phase detect. Pleasant surprise.
I had a look at the Metabones web site and here is what they state:

------------
Also added to Metabones firmware update version 1.8 is fast contrast detect AF for other Olympus cameras including E-M5II, E-M5, E-M10 and E-P5. Contrast-detect AF accuracy on Olympus is lens-dependent with most Canon-brand lenses introduced over the last 10 years working satisfactorily. However, some older Canon lenses and some third party EF-mount lenses may not be able to achieve pinpoint-accurate focus on Olympus cameras with contrast-detect AF. These lenses should be used on OM-D E-M1 instead.
------------

Which seems in line with my observations of the E-M5/MB/EF 400 5.6 setup being back-focus prone, but also capable of focusing spot on. I guess it depends on variations in the texture, contrast, maybe also colors of the subject.

Weather permitting I will do some more serious tests with the E-M5.
 
Have updated my MB adapter to the latest FW. Hopefully I will be able to try with the E-M5 the upcoming weekend.
 
Have updated my MB adapter to the latest FW. Hopefully I will be able to try with the E-M5 the upcoming weekend.

I suggest you use the electronic shutter. It allowed me to discover how good my Panasonic 100-300mm lens is. I'm eager to try the 400mm with it.

Here are 2 photos taken on the same branch moments apart. Both at 300mm, ISO 800. f/5.6 for the Chickadee, f/7.1 for the Gray Jay. Distance 7 to 8 meters.

All photos in the series were in perfect focus, equally sharp. I'm discovering a new lens...
 

Attachments

  • 001-160223ab018kf.jpg
    001-160223ab018kf.jpg
    194.3 KB · Views: 245
  • 002-160223ab043kf.jpg
    002-160223ab043kf.jpg
    285.8 KB · Views: 256
I love the electronic shutter for birds.... as long as nothing moves. Rolling shutter distortion can be pretty bad, but it is not so noticable on single shots.
 
Warning! This thread is more than 7 years ago old.
It's likely that no further discussion is required, in which case we recommend starting a new thread. If however you feel your response is required you can still do so.

Users who are viewing this thread

Back
Top