[opendtv] Re: CMOS sensors and rolling shutters
----- Original Message ----- From: "Dave Bittner" <dave@xxxxxxxxxxxxxxxxx>
What I'm wondering, for the seasoned engineers in the crowd, is if these kinds of artifacts were also present with tube cameras.
Yes, absolutely, and for the same reasons. The target of the pickup tube is charged by light from the lens, and is discharged one line at a time by the electron beam striking the rear of the target. The electron beam discharged the target one line at a time, so just like the "rolling shutter" CMOS effect, the bottom of the image represents a different point in time than the top of the image.
Yes, I recall the "leaning flagpole" effect when quickly panning a tube camera. However, my guess is that these effects are more pronounced in PAL than they are in NTSC, due to the framerate differences. They would probably be even more pronounced in a 24 fps video environment (and your article seemed to be aimed at a Digital Cinema audience.) This effect would only be seen during slo-mo replays or stills, however. When viewing the video in real time, the imaging tube and the display CRT were in lockstep, so the image looked natural at all times. 

I hope that Mark Schubin has more to add on this subject.
John 



----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org 
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts: