[squeak-dev] Re: Debugging slow browsers

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

[squeak-dev] Re: Debugging slow browsers

Jerome Peace
Debugging slow browsers
from thread:

[squeak-dev] Squeak 3.9 alpha image wanted

Hi Matthias,

With your benchmark I started a search among my collection of 3.9 alphas.
The slowdown started at update 6685..6686 and can be seen in the 6686 image.
It is absent in the 6684 image. Details are on mantis:

2084  Each overlaping browser slows down screen updates considerably.
Caching the image would speed things up.
http://bugs.squeak.org/view.php?id=2084

Your guess seems likely to be correct.
The translucent gradient was intoduced at this point.
It probably requires everything below the browser to be redrawn.

The pane-splitters were also introduced here.
They proably add to the complexity of redrawing.

Good luck in hunting down the solution.

Yours in curiosity and service, --Jerome Peace





      ____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

Reply | Threaded
Open this post in threaded view
|

Re: [squeak-dev] Re: Debugging slow browsers

Matthias Berth-2
Hi Jerome,


that narrows it down considerably, thanks a lot. Could you make the
6684 image available somehow?

Cheers

Matthias

On Fri, May 2, 2008 at 2:10 AM, Jerome Peace
<[hidden email]> wrote:

> Debugging slow browsers
>  from thread:
>
>  [squeak-dev] Squeak 3.9 alpha image wanted
>
>  Hi Matthias,
>
>  With your benchmark I started a search among my collection of 3.9 alphas.
>  The slowdown started at update 6685..6686 and can be seen in the 6686 image.
>  It is absent in the 6684 image. Details are on mantis:
>
>  2084  Each overlaping browser slows down screen updates considerably.
>  Caching the image would speed things up.
>  http://bugs.squeak.org/view.php?id=2084
>
>  Your guess seems likely to be correct.
>  The translucent gradient was intoduced at this point.
>  It probably requires everything below the browser to be redrawn.
>
>  The pane-splitters were also introduced here.
>  They proably add to the complexity of redrawing.
>
>  Good luck in hunting down the solution.
>
>  Yours in curiosity and service, --Jerome Peace
>
>
>
>
>
>
>       ____________________________________________________________________________________
>  Be a better friend, newshound, and
>  know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ
>
>