OK. From my understanding, it should be very easy to get high
performance from this sort of content in a constant-quality mode. The
tricky part is doing this with CBR rate control constraints.

Do the CBR bounds specified in the presentation make sense (300ms bucket
allowed for deviations)? If so, I think that test case is sufficient.

On 07/23/2015 12:17 AM, Jonathan Lennox wrote:
> Hi, all —
> 
> As requested, I’m posting to the list my comment at the mic today about the 
> temporal characteristics of screensharing.
> 
> The characteristic I was describing is not universal to everything that’s 
> described as “screensharing,” but it’s notable in the common use case of 
> presentation sharing and the like.
> 
> The characteristic I’m thinking of is that the video stays static or nearly 
> static for long but unpredictable periods of time (showing the slide), and 
> then many parts of the image change all at once with little or no commonality 
> with the previous image (changing the slide). But there may also be periods 
> of time with more traditional video-like behavior (animations and the like).
> _______________________________________________
> video-codec mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/video-codec
> 

_______________________________________________
video-codec mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/video-codec

Reply via email to