Re: [Gnash-commit] [SCM] Gnash branch, noise, updated. release_0_8_9_final-1233-g311abd4

[Rob Savoye]

On 10/30/11 08:45, Sandro Santilli wrote:
> On Sun, Oct 30, 2011 at 08:32:42AM -0600, Rob Savoye wrote:

> Btw, the 10ms "magic number" value comes from tests performed
> by Zou Lunkai about setTimer/setInterval capacity.
> You couldn't make AS actions run at faster frequency than 10ms
> per run. 

  To truly investigate performance issues, we have to assume we may have
some wrong assumptions... Sometimes I think we rely too much on our
simple test cases, and not enough on the real world. Users only have
real world results to judge Gnash's performance by.

  Since these tweaks are in a public branch, you can try for yourself.
So far this particular experiment is interesting... Right now 'noise' is
master with just a few changes, so I'm using it as my default player.

        - rob -