Strange speed differences on composite

[Adam Jackson]

On Wednesday 15 February 2006 12:40, Tomasz Torcz wrote:
>   Now, this is something that surprised me. I lived under impression
> that mga actually accelerates everything. Few years back I bought this
> costly (ati and nvidia cards were 2-3 times cheaper) card _because_ it
> had open source driver including 3D part. And it was sufficient to play
> quake3. Somehow I thought 3d == full acceleration. I was wrong all those
> years :(

At the time, it effectively did accelerate everything, because we weren't 
using Render as heavily so not having it accelerated wasn't as noticable.  I 
don't think XAA even grew its (anemic) Render acceleration hooks until like 
4.2.

Also, it _is_ true that being capable enough to run GL in hardware generally 
means being capable enough to accelerate Render in hardware.  The matrox 
driver in kdrive does have Render acceleration, and so did the mga driver 
until fairly recently (disabled once for being broken on dual head, and now 
effectively entirely due to XAA's Render accel hooks changing incompatibly).  
It would certainly be possible to make mga(4) accelerate Render pretty 
completely, and it should be a pretty straightforward port from the kdrive 
driver to do so.  This would make an excellent project for someone looking to 
get involved in hacking X.

- ajax
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://lists.x.org/archives/xorg/attachments/20060218/d71777d2/attachment.pgp>