Shutter Workshop

[Jim Arthurs]

Hi all... with all the talk of rolling shutters and global shutters and various shutter signatures, I thought it would be informative to open a thread and tap into the brain trust here to help define what makes the "look" of a paticular shutter.

Since one of the goals of digital cinema is the styling of an electronic look equal to the classic look of a mechnical film shutter I decided to do an animation showing how the standard 180 degree shutter builds up an image on each frame. This is not a specific camera, but covers all the classic rotating shutters.

I'd like to include any other film shutters commonly used in motion picture work that would create a significantly different look, so please help me out with resources to locate and model animations for them.

Here's my first pass at it, I'm looking for comments and feedback to improve it and increase the accuracy of what I'm trying to convey.

http://ftp.datausa.com/imageshoppe/outgoing/SHUTTER_STUDIES/FILM_SHUTTER_SKEW_CLASSIC.mov

I'll refine as time permits (others are welcome to do better versions (please!)), but this kicks it off. My focus here is specifically defining how skew is created in a film shutter, since so much of the CMOS rolling shutter dialog is in comparing it to film skew.

In this example, I've chosen a "worst case" study of an object moving through the FOV in a single frame. Note that potential skew happens only at the beginning and end of the shutter open cycle.

Regards,

[Anders Holck]

Actually I think you have to account for the angle of the shutter as well.
A mirror shutter is not positioned at the film plane but it at a 45° angle, and the topmost edge is a few mm in front of the filmplane. This makes the shutter blade edge much more diffuse than illustated on your animation.

[Anders Holck]

I did this Animation a while ago in the rolling shutter thread:

Rolling shutter
http://holckowen.com/red/Shutter_rolling.gif

Mirror shutter
http://holckowen.com/red/Shutter_Mirror.gif

Actually in reality I think the shutter blade shadow is actually much more diffuse at most practical apertures.
And of course, the more diffuse it is the more it takes the form of a general image attenuation, instead of a wipe.

[Anders Holck]

Here is the Arri 416 Mirror/gate block:
http://holckowen.com/red/Mirrorhousing.png
It clearly shows how the shutter plane is positioned in relation to the film gate.
Also note the the gate is not centered above the mirror motor shaft, but instead the right edge of the gate is flush to the shaft position, making the shutter motion more diagonal.

[Jim Arthurs]

Great tips, Anders... In this run through, I was just referencing my old B&H 2709 and Wall cameras, where the shutter is perfectly perpendicular to the film plane, same with the rack-over Mitchells. My old Cameflex CM3 had a diagonal mirror shutter, but it was in-line with the shutter shaft, if I remember correctly. I should update to more reflect modern designs.

I absolutely agree with you about the degree of diffusion as the shutter wipes across the frame, which would dilute the effect of any skew as the shutter advances and retreats over the film plane.

I hope people can take a few things from this... mainly that there are two periods of potential skew as well as a significant period with NO skew, and that the whole will somewhat balance itself out over the full exposure. Of course, greater than 180 degree shutters will have even less skew, but overall more blur...

Great find with that 416 photo... I'll have to incorporate the offset as well...