Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

CMOS and rolling shutter

P

Peter Hebert

Member
Joined
Jan 22, 2008
Messages
14
Reaction score
0
Points
0
Location
Vancouver, BC
Website
www.peterhebert.com
I keep reading that CMOS sensors all have 'rolling shutter' issues, due to the fact they are scanned from top to bottom instead of all at once. This sounds like the old vacuum tube TV cameras. Anyway, since this approach apparently has such adverse motion artifacts, why do they employ this method?

Also, I would like to know :

  • is there is such a thing as CMOS that scans all at once?
  • why not CCD for 2K-plus cameras?
  • are the rolling shutter artifacts really as bad as people say?
  • what are some of the technical pros/cons for both CMOS and CCD
 
See CHE

See CHE

Peter Hebert said:
I keep reading that CMOS sensors all have 'rolling shutter' issues, due to the fact they are scanned from top to bottom instead of all at once. This sounds like the old vacuum tube TV cameras. Anyway, since this approach apparently has such adverse motion artifacts, why do they employ this method?

Also, I would like to know :

  • is there is such a thing as CMOS that scans all at once?
  • why not CCD for 2K-plus cameras?
  • are the rolling shutter artifacts really as bad as people say?
  • what are some of the technical pros/cons for both CMOS and CCD
Click to expand...

Go see CHE, I do not think you will see the rolling shutter in most shots.

On slower chips you can see it a little if the shutter angle is too small, just as you will see "picket fensing" on a movie camera with the shutter at 5 degrees.

CMOS got a bad rep from cameras that use "automatic" exposure so that the shutter angle goes small on bright shots or too wide on dim shots, with the RED ONE you can set the angle to 180 Degrees.

Acam dII has a CCD sensor, but at lower pixel count. The resolution depends on the OLPF and compression used as well as the sensor, lens, and pixel count, and the sharpen setting in the de-Bayer (de-mosaic).

Once you get to compressed broadcast, DVD, or Blu-ray, none of these issues make much difference. Nor does the lens or 4K vs. 2K filmout, since the movie print will be projected out of focus most of the time anyway.
 
We were shooting some clips in Atlanta last Sunday while Peter Majtan was showing off his R1 to the Scarletuser SE group. There were a couple of fast pans in the clips and they show almost no observable rolling shutter artifacts on the scenes we shot. I don't think it is much of an issue with R1 and release 17 software. Very close to if not on par with film cameras.
It was my first hands on exposure to the camera and Peter gave us a good short run through of the general ops procedures. I am more impressed than ever with the capabilities of R1.
 
The shutter on the R1 rolls in an incredibly small amount of time so skew and wobble are negligible in all but the most extreme circumstances. If you're panning very quickly against buildings it may be noticeable, but on human subjects, a slight skew or wobble on a human subject is incredibly hard to detect in a motion shot because the human form is so flexible and variable that a viewer won't realize that the image is distorted unless they're examining individual frames in which the artifact occurs.
 
We just shot a music video and used a really fast shutter (1/250 I think). Some footage was shot on a stage with stage lighting and strobes, and the director and I both really liked the way the rolling shutter would only pick up parts of the frame with the strobe. Instead of the whole frame strobing, there are these awesome bands of strobe that appear randomly. Very cool effect. We are grading right now and will post the piece soon.
 
It will appear in more situations than you think...

You cannot shoot 180 degrees while using Magnetic Ballast HMIs, you must match the frequency which is 60hz, or 1/60th shutter or 144º
 
The Dalsa has (or had) a 4K CCD. Apparently one reason to go with CMOS is that it doesn't need a separate A/D converter so there is some savings in space when designing the camera, though perhaps small (the size of the A/D converter could not have accounted for the huge size of the Dalsa.)

The Genesis and F35 have a single-sensor 35mm-sized CCD with 12 million photosites.

The Arri-D21, SI-2K, and Phantom all use CMOS sensors.
 
I don't think Dalsa's Magic Pixies went on the Graeme diet. I sent the RED Magic Pixies my special dietary advice that personally lost me 50lbs. Perhaps Dalsa's Pixies were still on the USDA Food Pyramid? :-)

Graeme
 
Graeme Nattress said:
I don't think Dalsa's Magic Pixies went on the Graeme diet. I sent the RED Magic Pixies my special dietary advice that personally lost me 50lbs. Perhaps Dalsa's Pixies were still on the USDA Food Pyramid? :-)

Graeme
Click to expand...
Yeah, RED's pixies have an all-RAW diet. :)
 
They also understand about compression :-)

Graeme
 
any medicine at post?
 
You must log in or register to reply here.
Back
Top