C200: Strange Problem: Frame Jumping

D

dharty

Active member
Hello,

Shot my first interviews with my new C200 and had an odd problem: periodically the frame would "jump" 3 - 4 pixels to the left. It appears to happen when the interviewee moves their head (head shoulder framing), so my first guess is something odd with the autofocus using a Sigma 24-105 F/4 with the latest lens firmware with the stabilization turned off. Interview shot from tripod. Saw the problem in two different interviews shot two separate days.

Here were my camera settings:

MP4 4K

AF - continuous
AF Frame - selectable
AF Frame - small
Apply AF Speed Limit - always
AF Speed - 0
Face Detection - On
Face AF - Face Priority

I noticed the frame jump while shooting the interview so changed AF to One Shot, but still got the odd frame jump.

Anyone experienced this problem and have any suggestions?

THanks, Drew Harty
 
I have this lens but haven't tried it on the C200 yet. If I get time I'll test it but yes, posting footage would really help. Also, they are still working on a new firmware for this lens like they released for the other Sigma Art lenses that improves DPAF compatibility.

It's curious cuz this sounds like the OIS of the lens being enabled and somehow the face tracking is pulling the lens a little...
 
It sounds like the helical or lens element guide is not as tight as it needs to be. Whenever the focus motor changes directions during AF, it causes the elements to shift slightly and change the image that hits the sensor.
 
I occasionally see something like this with my 70-200 2.8 II IS when the IS is turned on. I just flip the IS off and on, and it fixes it. If you already have the IS off, this doesn't sound like the same thing.
 
If it only happens while focusing, then I am pretty sure that it is an issue with the lens element guides in the lens. I see it with older lenses that have become loose. Focusing the lens one way and then the other causes the lens elements to shift inside the guides as the direction of the force from the helical changes. You can test this by just manually focusing the lens one way and the other on a locked off shot and see if the frame jumps.
 
Hello,

Here is a clip showing how the frame jumps. It occurs about 2/3 in and last about 2-3 frames.

<iframe src="https://player.vimeo.com/video/268826645" width="640" height="360" frameborder="0" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>
 
Here is a second clip that shows the frame shift. Watch the lower left of frame and you will see a highlight on metal jump out of frame.

<iframe src="https://player.vimeo.com/video/269043726" width="640" height="360" frameborder="0" allowfullscreen></iframe>
 
Sorry to dig this up but wanted to add - I saw something similar with this lens and the C200 on a shoot last night but I had the lens stabilization turned ON. It seemed to coincide when I would use AF Lock once the DPAF had locked onto a target I was shooting. I should note that even though I had a shoulder rig, I was 100% handheld so it's possible what I was seeing was just the OIS trying to compensate for so much movement.

I've noticed that when shooting stills with this lens on an 80D, the OIS will "lock on" when you are using auto focus and press the shutter button half-way down to initiate auto focus.

I wonder if somehow the OIS switch is being ignored on the C200 and it's always on? That seems impossible but I can't explain it otherwise. They do have an update for this lens in the pipeline that will bring it support of DPAF 2 like they released in March for a large portion of the Sigma Art lenses. I'm guessing (lots of guessing, I know!) the OIS is the reason for the staggered releases.

PS- the DPAF of this lens worked a treat with the C200, even in lowlight conditions where I was shooting with ISO 3200
 
You must log in or register to reply here.
Back
Top