Does the HC3 show th3 HC1 "rolling shutter" artifact

[Steve Mullen]

The rolling shutter artifact has been well document.

But looking over the last 8 pages -- I don't see any comments on whether the HC3 also has the artifact? Does it?

[Stu Holmes]

I think it may do, BUT to a MUCH lesser extent than the HC1/A1. I heard from someone who tested it that it does show rolling shutter but you REALLY have to provoke it for that effect to be noticeable at all.

Perhaps an HC3 owner (even better, an HC3 owner who has or has had an HC1 or A1) can 100% confirm this is the case.

I'm guessing that you'd like to know this as the V1's chip is similar design..

[Mikko Lopponen]

I watched an hc3 clip and it seemed to show the same effect.

[Scott Myers]

I bought an HC3 recently. What is the rolling shutter artifact? I don't think I've noticed anything unusual in my footage so far.

[Mikko Lopponen]

The top part of the image is recorded at a different time than the lower part of the image. If something goes fast across the screen it will look distorted. Shaking the camera will result in wobbling.

The a1/hc1 have this effect (I have the hc1, but not the hc3).

[Scott Myers]

When I first bought the camera, I deliberately did a lot of panning, both quickly and slowly, and did not see this effect. I'll test it some more with faster pans and shakes.

I will say that I saw some strange artifacts after editing/rendering a time or two but the original footage did not show those artifacts. I tried to replicate those results in the same editor and could not, so something was temporarliy out of whack with the editor.

[Scott Myers]

I just took some more footage with and without the stabilizer on and there's nothing wrong with the footage. This was DV because I've been doing some multicam tests with my editor (my other camera is analog and I didn't want to transcode or downconvert). In the earlier HDV footage I shot which had the stablizer on, I also did not see anything wrong with the footage.

The only time I've seen anything resembling what you described is when playing the tape in either slow or fast modes or when the camera is doing some special function such as playing the last seconds of the recording.

I've seen no problems when playing the footage normally on the camera's display or in an editor after capture.

Are you talking about a display artifact on the camera whle recording (or after) or an artifact in the recorded footage? I've never seen this happen even on the camera while taping.

[Mikko Lopponen]

If you watch it in an interlaced format with about 50/60 fields per second then it will be more difficult to spot. Once you deinterlace the video and go to 25/30 frames per second it is easier to spot.

Higher shutter speeds will also make it more apparent. If your shutter is at 1/50 or 1/60 then it will be harder to spot because moving objects will be blurred anyway. With a higher shutter they are not blurred so the slanting is obvious to my eyes anyway.

I hate the effect. If you still can't seem to see it I'll post a clip which shows it pretty badly.

[Scott Myers]

Mikko,

I understand what you're saying. To be as sure as possible to be able to see whatever this is, I have panned on objects with vertical lines so if there was any problem those objects would have a discontinuity - the straight vertical lines would be broken whether they were blurred or not. I have not seen any evidence of this so far.

I'll test it a little more sometime.

If you can provide a clip, I'd appreciate it. I'd like to know what this is so that I can watch for it.

[Mikko Lopponen]

Here are two clips.

http://hmcindie.pp.fi/rolling/

You can see the effect very clearly in the city clip. Shot with a high shutter. Makes doing action scenes a real pain. Cheap cellphone videocams have a rolling shutter too but with a worse effect.

[Graham Mullis]

Another clip

Sony HC3 Rolling Shutter

[Scott Myers]

Thanks, Mikko. Are you talking about the slanting of the image? When you described it earlier, I thought it was something completely different.

I guess I'm so used to seeing artifacts like this that it didn't stick out as a problem on the HC3. I think I have seen that on mine but it seemed normal to me.

I've seen all kinds of artifacts with still and motion picture cameras depending on their shutter design and the relative motion between camera and subject. This one reminds me of a very old photograph of a race car with the wheels appearing oval and slanting in the direction of travel while the background slanted in the opposite direction.

[Scott Myers]

Thanks, Graham. I'm assuming that it's the slanting we're talking about. I see it but that's one of those things that I haven't seen as a problem. Just about every camera I've ever used (still, motion picture, video, and whether film or digital) has had some idiosyncracy in its design.

My first thought is that this is due to the way the image sensor is scanned, maybe in the scan rate. I'm interested now in finding out more about the design.

[Mikko Lopponen]

Originally Posted by Scott Myers

My first thought is that this is due to the way the image sensor is scanned, maybe in the scan rate. I'm interested now in finding out more about the design.

Yup. For some reason the image is scanned line by line starting from the top. I'd much rather trade in some image quality to get a real still frame of the moment than a skewed image. When doing slowmotion it tends to get in the way and I happen to shoot a lot of handheld highspeed stuff anyway. Camera shaking looks a lot worse with a rolling shutter camera.