Announcement

Collapse
No announcement yet.

The State, Direction Of The PSCNV Nouveau Fork

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • pingufunkybeat
    replied
    I really don't understand the point of this fork.

    I'm sure that they're not stupid, and that there may be technical reasons for reinventing the wheel, but they must be aware that this sort of insular development isolated from most of the OSS community is usually doomed from the outset.

    Leave a comment:


  • V!NCENT
    replied
    They probably looked at Fermi and wondered how much this card could perform and Gallium3D probably wasn't helping them in their ultimate vision.

    If only they could stop staring at their car navigation and realise that the world does not revolve around them <_<'

    Leave a comment:


  • dfx.
    replied
    looks like radeonhd not taught them that incompatible user-alienating ununified community-dividing graphic driver forks are shortliving, useless and disrupting.
    gallium maybe not a saviour of all humanity but it's a way most developers and users want things done, and in current state of nouveau, fork is the last thing it needs :\

    it also looks like they just want not to have troubles porting GNU/Linux driver awesomeness on those other OS's and reinventing bicycle again. of course they will not have skill or resources to do it single-handedly and no one going to help them. good fucking luck.

    Leave a comment:


  • sturmflut
    replied
    Originally posted by Zajec View Post
    They don't like Gallium, don't like TTM (GEM too I guess) and are interested in *BSD/OpenSolaris(/ReactOS?). Sound a little weird to me. It sounds they already created own memory manager (couldn't TTM be improved any way?), hope they won't create new Mesa framework.
    This fork smells like Reiser4 all over: PathScale may come up with something which performs on a Fermi GPU, but their efforts do not seem to be very well integrated with the rest of the open source crowd. Nobody will want to include it later since it is incompatible with the current state of kernel and X.Org, and nobody will want to support it when PathScale decides to no longer work on it one day. And they already started insulting people ("Gallium has very low quality" etc.). GREAT move, PathScale! Did Hans Reiser not teach you anything?

    Since all the things they point out in TTM seem fixable to me, and Nouveau already supports quite some stuff on NV50, why don't they just branch both projects and work with the community? Would probably take a bit longer, but if they really come up with something working until the end of the year it will probably already be included in the next bigger distributions (Ubuntu 10.04, Fedora 15). Nobody will want to include a fork, since PSCNV does not support older hardware distributions would even have to deliver Nouveau AND the fork, and make sure they don't block each other.

    Leave a comment:


  • [Knuckles]
    replied
    Originally posted by Zajec View Post
    They don't like Gallium, don't like TTM (GEM too I guess) and are interested in *BSD/OpenSolaris(/ReactOS?). Sound a little weird to me. It sounds they already created own memory manager (couldn't TTM be improved any way?), hope they won't create new Mesa framework.
    My thinking exactly.

    I hope they don't go around re-doing everything because they think "it sucks" and their work just gets lost on it all, or gets very hard to use because they patch things everywhere.

    Leave a comment:


  • Zajec
    replied
    They don't like Gallium, don't like TTM (GEM too I guess) and are interested in *BSD/OpenSolaris(/ReactOS?). Sound a little weird to me. It sounds they already created own memory manager (couldn't TTM be improved any way?), hope they won't create new Mesa framework.

    Leave a comment:


  • Xake
    replied
    Originally posted by ethana2 View Post
    How about they support something new and interesting, like Poulsbo?
    How can Poulsbo be newer and more intresting then Fermi?
    And even more how for a company working on a Nvidia-driver and not a PowerVR/Intel-driver?

    Leave a comment:


  • ethana2
    replied
    Uh...

    Look, I'm willing to lose 20% of my OpenGL performance if it means I get Direct3D, video decoding, OpenCL, and whatever else, open sourced, in a timely manner.

    How about they support something new and interesting, like Poulsbo?

    Leave a comment:


  • phoronix
    started a topic The State, Direction Of The PSCNV Nouveau Fork

    The State, Direction Of The PSCNV Nouveau Fork

    Phoronix: The State, Direction Of The PSCNV Nouveau Forkhttp://www.phoronix.com/vr.php?view=ODcyNg
Working...
X