Re: Re: Source for Cr overhead

On 4/18/06, Brian Paul <[email protected]> wrote:
>
> Niraj Tolia wrote:
> > On 4/18/06, *Brian Paul* <[email protected]
> > <mailto:[email protected]>> wrote:
> >
> >     Niraj Tolia wrote:
> >      > Hi Brian,
> >      >
> >      > We are using fairly recent Cr CVS code for QuakeViz (an
> earthquake
> >      > modeling app from http://www.cs.cmu.edu/~quake/>. Our testing
> right
> >      > should not be stressing a system as we rotate an small model in a
> >      > 640x480 window. However, we see a very significant difference in
> the
> >      > nubmer of frames/sec.
> >      >
> >      > Attached is a graph of a particular interactive benchmark.
> >
> >     I don't see an attachment.
> >
> >
> > My apologies. It seems I forgot to attach the PDF. Attached now.
>
> What does the vertical axis (CDF) mean?


The graph is measuring the number of FPS seen for each user input that
generates a screen update. For example, the act of moving your mouse to
rotate an object might generate 60 frames over 3 seconds, or on average, 20
FPS. As we have a very large number of user inputs in this benchmark, we
generate a CDF to figure out what fraction of events deliver 'N' frames per
second.

We ensure that the results are consistent across different scenarios by
using a replay tool for user input.

Niraj

-Brian
>
>


--
http://www.cs.cmu.edu/~ntolia