Measuring rolling shutter: put a number on this issue!

Samuel H said:
It's on the table, close to the top, with a very nice 13.4 ms

Squig: this was a BlackMagic Micro Cinema Camera shooting 1080p24, right?
Click to expand...

Thanks very much for taking the measurements!

Does increasing frame rate or shutter speed typically have an effect on rolling shutter time? If so, would it increase or decrease it?
 
I'm curious to know how some of the popular semi-pro camcorders fair in this list. Canon XA20 and Sony X70 for instance.

... I imagine they wouldn't be better than the FS7, but you never know. The NX1's performance was a real surprise.
 
Sounds like buying from GoPro where they list a bunch of new features and may take a year to deliver (if at all). Not a good business model in my opinion.
 
Greg_E said:
Sounds like buying from GoPro where they list a bunch of new features and may take a year to deliver (if at all). Not a good business model in my opinion.
Click to expand...

I prefer they try and fail then not try. But I wouldn't plan ahead based on BM's release announcements
 
bdemenil said:
Thanks very much for taking the measurements!

Does increasing frame rate or shutter speed typically have an effect on rolling shutter time? If so, would it increase or decrease it?
Click to expand...
Either there's no change, or you get better rolling shutter... but that only happens if when increasing the frame rate the camera switches to line-skipping.
 
Samuel H said:
Either there's no change, or you get better rolling shutter... but that only happens if when increasing the frame rate the camera switches to line-skipping.
Click to expand...
are the improvements that some 4k cameras get when shooting in 1080p due to line skipping?

How do the high end cameras achieve both global shutter and high dynamic range and sensitivity?
 
Yes. For example, the a7S gets faster RS in FF mode 1080p, because it goes into line-skip mode (it's not in the table, though, because I never measured it, since there's also an APS-C mode that has both faster RS and a clean image).

High-end cameras usually have less mpix to read out of the sensor, and/or more ADC units to do it fast.
 
BM 4.6K added to the table, thanks to dia3olik :)

It's very good, at 16.3ms (about the same as the FS7, those two are really close to each other in every aspect).
 
Hi, these are the actual read out speeds for our cameras:

URSA Mini 4.6k:
4.6k – 15.17ms
4k/2160p – 12.64ms
2k/1080p – 6.32ms

Micro Cinema: 13.3ms

Pocket: 17.74ms

BMCC: 23.62ms

Thanks!
 
WOW, thanks a lot for posting that!!!

I think it validates my method: it does have some error (as expected) but I was within a couple of ms in all cases :)

4.6k – 15.17ms (I said 16.3 ms)
Micro Cinema: 13.3ms (I said 13.4 ms)
Pocket: 17.74ms (I said 17.8 ms)
BMCC: 23.62ms (I said 25 ms)

I'll update the table with your numbers. Thanks again!
 
No problem, i notice you've posted some speeds in crop modes for other cameras in your list - it might be useful for some people to add the crop modes for the 4.6K i listed too? For instance you could have a wider field of view with faster read out time on the 4.6K versus the Micro Cinema, but that might not be immediately obvious if not on the list.

Thanks!
 
Good idea. Done.

If you want my opinion, with rolling shutter this good, it's no big deal that you dropped global shutter (which had a hit on dynamic range in any case). Taking so long to ship, that may hurt you a bit more...
Killer camera, though. Can't wait to see this sensor on a DSLR-like body selling for <$3K :)
 
CaptainHook said:
Hi, these are the actual read out speeds for our cameras:

URSA Mini 4.6k:
4.6k – 15.17ms
4k/2160p – 12.64ms
2k/1080p – 6.32ms

Micro Cinema: 13.3ms

Pocket: 17.74ms

BMCC: 23.62ms

Thanks!
Click to expand...

Looking at cinema camera's for a feature and I was disappointed that you had dropped the global shutter. However posting these numbers wins a large amount of bonus points with me. Even if they weren't as great as they are, I would still appreciate you providing the information. Keep up the great work.
 
That's handheld, I can't take a measurement I'm afraid. It looks mediocre at best, that's all I can say.
 
You must log in or register to reply here.
Back
Top