[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: KDE4 graphics tweaks page
From:       "koos vriezen" <koos.vriezen () gmail ! com>
Date:       2008-07-14 19:08:45
Message-ID: d4e708d60807141208w198bac06qb84f0e5fe31f2f83 () mail ! gmail ! com
[Download RAW message or body]

2008/7/14 Aaron J. Seigo <aseigo@kde.org>:
> On Monday 14 July 2008, koos vriezen wrote:
>> Somehow this makes me sad. I mean if we know that certain version of
>> xservers are known to perform badly on some operations, shouldn't this
>> be in some blacklist, and simply disable the operation in question (or
>> at least warned upon enabling such a feature).
>
> i'd prefer it if the vendors improved their drivers. the only way that will
> happen is if there are real world apps stressing these issues out there in
> production.
>
> the fact that there wasn't up until now is why the drivers have all these
> weaknesses still.
>
> the driver writers have shown pretty conclusively that unless there is a good
> reason, e.g. applications in use, to provide support for they won't do the
> work necessary to improve those features. and that sort of makes sense: why
> invest in something nobody uses?
>
> we already do things like ask the driver if it supports argb and if it does we
> trust it.
>
> if we try and blacklist based on card-and-driver combinations, we'll give
> driver writers an excuse not to work on these features (which we really do
> need to provide the best possible experience) and we'll still miss certain
> combinations.

Yes I completely agree with your argumentations. However, that's why I
mentioned versions, nvidia-glx 173.14.09 will never work properly if
our software doesn't adapt for it.
Note, the point is to give an out-of-the-box good experience. Letting
users chase driver tweaks might be fun for some, but also give
negative feedback. Whether that will encourage nvidia to hurry up
their work is speculating.
The issue is known, why give them a bad feeling.

Koos

> intel's drivers work well, so it can be done.
>
> nvidia has shown that when an issue appears in a real world application, they
> will try and fix things. they've already fixed a few things for plasma
> specifically.
>
> i hope the same is true for amd/ati, though i don't have quite as good
> contacts there to know for sure.
>
> --
> Aaron J. Seigo
> humru othro a kohnu se
> GPG Fingerprint: 8B8B 2209 0C6F 7C47 B1EA  EE75 D6B7 2EB1 A7F1 DB43
>
> KDE core developer sponsored by Trolltech
>
>
>
>>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
>
>
 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]