Talk:Rolling shutter

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

Cmos sensor[edit]

The cmos sensor is electrical and is slow therefore the lag/rolling shutter.

This isn't clarified well enough since a lot of electronics use this chip.

The pictures in the article are Cmos chip rolling shutter artifacts.

There should be a paragraph on this issue with Cmos chips.

Yes, but what is it?[edit]

The article, as it currently stands, skirts around the question and avoids explaining how Rolling shutter is implemented. Is it a mechanical device, or a special blanking waveform applied to the sensor? (talk) 16:26, 6 January 2009 (UTC)

The real problem is that it doesn't deserve to be an article at all. Has nobody viewing this article ever used a slit-camera (or a Widelux)? Photographers today miss out on so much nifty stuff. —Preceding unsigned comment added by (talk) 20:47, 20 August 2010 (UTC)

Great Example at BoingBoing[edit]

Spblat (talk) 22:14, 20 August 2010 (UTC)

I guess that this article is wrong because it swaps the effect of CCD and CMOS sensors. In fact, CCD sensors perform charge analog-to-digital conversion on pixel at a time (the have only one ADC) and consequently show a rolling shutter effect. Conversely, in APS sensors like CMOS, the AD conversion is performed in-sensor, i.e. each pixel performs contemporaneously the conversion. See for example P4010 (talk) 08:27, 31 August 2010 (UTC)

Exposure windows for each pixel? Flash?[edit]

If I take a single exposure with a rolling shutter, how would I determine the exposure for each pixel? Suppose I have a 1/20s exposure and the rolling sweep is set to do a sweep in the same 0.0333 seconds (30fps) with 1000 rows of pixels. When I say "go", is it that the first row starts exposing, with the "exposure-begin" sweeping down, starting the second row at t=0.0333s/1000, etc. Then at t=0.05s, the first row is read, at t=0.05s+0.0444s/1000 the second row is read, etc?

Is that about right?

If so, can we use a flash as follows?: At t=0s, say "go", then at t=0.0333s (when all pixels are active), we trigger a flash for 0.001s, then at t=0.05s, the read begins. It seems like that would provide a full-frame exposure of the duration t=0.0333s to t=0.0433s. Is that right? —Ben FrantzDale (talk) 18:00, 15 November 2010 (UTC)

It looks like the answer is yes. [1] —Ben FrantzDale (talk) 18:08, 15 November 2010 (UTC)

Why does Global Shutter redirect here?[edit]

I wanted to research how global shutters work, and it directed me to this page. This article contains the line "This in contrast with global shutter in which the entire frame is exposed for the same time window." Hence, rolling shutters work differently from global shutters; therefore, they should each have their own page. I don't know how to change where it directs to, but I think at the very least, global shutters should be directed to its own page. —Preceding unsigned comment added by (talk) 04:28, 19 April 2011 (UTC)