Rolling shutter effect from lens IS

greeches

greeches

Veteran
So, I was shooting with my 28-135mm Canon lens the other day, all the way to 135mm, and I started to get jello effect rolling through the picture as my IS was working. Anyone else noticed this with IS lenses? I will post some sample footage in a bit, but this is the second time I've seen it.
 
Technically this should not occur as the image is supposedly being stabilized in the lens before projecting the image to the sensor, unless of course the subject was moving very quickly in the frame from left to right or up to down.

Can you post some examples?
 
not sure if IS has anything to do with it, it's the sensor...that's what CMOS is good at
 
I have some footage at home. I feel as if the vibrations created by the IS is effecting the captured image. I've only noticed it at longer zooms. It rolls up the sensor. Will post some footage soon. (in a day or so)
 
greeches said:
It rolls up the sensor.
Click to expand...
If it's rolling up the screen, it's almost definitely the CMOS shutter problem. Were you shooting at 1/60th? If not, and you were shooting under fluos or HMIs or streetlights or stadium lights, then ... yeah, that would explain it.

Stick to 1/60th in the USA, 1/50th in Europe.
 
My guess is that your shooting hand-held. The more you zoom-in, the more small camera movements become amplified. IS really helps not have the jello effect. Just try turning off the IS and see how bad the effect becomes. I can only shoot hand-held with this camera up to about 50mm with IS. Anything more and I want a tri-pod.
 
IMPORTANT NOTICE: No media files are hosted on these forums. By clicking the link below you agree to view content from an external website. We can not be held responsible for the suitability or legality of this material. If the video does not play, wait a minute or try again later.
I AGREE


Tell me what you think, its at the very end (7 seconds) of the clip. I only see this zoomed out. I believe it has to do with the amplification, as you RyanFilms mentioned, but I think it is some frequency/vibration from the IS motor. This shot was shot outside in cloudy daylight, at Chichen Itza in Mexico.
 
Very strange problem. Has it happened at a wider field of view with the 28-135, or with other lenses?
 
The only time I have seen it is when IS is on and I'm nearly all the way out on the lens. I have no other long lenses that have IS.

Anyone else care to test?
 
My guess is that the IS is correcting the extremely minute shakes but can't quite get all of it out. To the best of my understanding, Canon IS isn't supposed to keep your image still, it just lessens the shake as much as it can.

In its intended use, it's supposed to stabilize your shaky hands enough to lower the shutter speed by around 3-4 stops, depending on the lens. It's not even meant for video work.

My conclusion? Nothing wrong with the lens or the camera. Get a rig, a tripod or stop drinking coffee :cheesy:
 
Well, greeches, that is indeed bizarre. Does it happen only handheld? Does it happen on a tripod?

It seems like it could potentially be a combination of microjitters just happening to perfectly coincide with the scan rate, but that seems farfetched. If you can make it happen every time, then I say try to borrow someone else's lens and rule out whether it's a hardware issue with your particular lens.
 
It just looks like CMOS "Jello" to me. Not uncommon with all these cameras. Maybe I'm missing something. I've seen some on my shots do some weird things on long zooms even on a tripod. Anytime the image bounces around, "jello" can show up.
Most of the time I'm just re-positioning the frame when it happens, stuff I would edit out of most projects anyway. "IS" is your friend most of the time. This image just may have confused the IS processor. If that is the end of the take, it may just be extra camera shake as you hit the stop button.
 
Last edited:
greeches said:
The only time I have seen it is when IS is on and I'm nearly all the way out on the lens. I have no other long lenses that have IS.

Anyone else care to test?
Click to expand...

What other long non IS lenses are you using, and are they used on a rig or tripod?
 
I'll try to recreate the problem. I notice with long lenses, it amplifies jello certainly, as a 'microbump' is multiplied when far out. I feel like its the gyro mechanism(whatever steadies the glass) creating the scanning effect.

I wouldn't mind it if I'm going for an old tv effect. :beer:
 
I have also had this problem occur however and specifically with the 28-135 lens.

When the IS was operating, it seemed like there was heavy vibration in the lens. Very heavy noticeable vibration which was gone when IS was turned off. IS in lenses seem to each have a different way of operating per lens as some lenses are constructed differently with different element groups and so on so I'm wondering if this is a common MO for the 28-135.

Perhaps to try another lens of the same type and see if the effect is lessened/removed? Because the vibrations from the IS unit were definitely putting extra jello effect on the image as when I tried the 17-55 IS USM -- that was very very smooth.

I'm just thinking the IS part of this particular lens either is a bad component somewhere or just in general not the best for this lens.
 
I agree. You can hear the noise of the IS in the video aswell.

I think I've noticed it more pointing the lens up or down, maybe suggesting gravity affects something internally to cause the pattern to show up...
 
You must log in or register to reply here.
Back
Top