[forum] Re: "Drivers? We don't need no stinking..."

[James Arthur]

On Thursday 27 March 2003 10:07, Sven Luther wrote:
> > No offence, but one Radeon 9000 isn't going to make much difference to
> > ATI.
>
> Where would be the threshold ? How many users would make a difference ?

I don't know. If one in every hundred users bought their graphics card on the 
strength of open source drivers, then it might. I'm guessing (I could be 
wrong) that the ratio is far smaller though, and probably not enough for the 
manufacturers. I'm guessing this because otherwise every card manufacturer 
would open up their drivers.

> > Having said that, I bought a Radeon 7500 for precisely the same reason,
> > and
>
> You see, already two boards.

Not really:

> > regretted it ever since.

Besides, being fanatically one-sided is, amongst other petty things, a waste 
of energy.

> You are aware that using proprietary drivers is not a guarantee of "it
> just works", at least not from the feedback i see on the debian-x
> mailing list, most problems come from mixed up mga_hal things but many
> also have problems with the nvidia drivers.

This is true. In fact, on some graphics cards I've used the drivers maintained 
by the open source community are wonderful, especially for older cards.

Personally, I've not had a problem with any of my nvidia cards under Debian.

> Well, it is not a moral stance, open source drivers often work better,
> sure you may have to use somewhat older versions,

Which is an issue for a lot of people (including me). Ideally we'd be able to 
pay people to write open source drivers.

> You are aware that DRI release (weekly or so) trunk snapshots ? And that
> Michel even has dri trunk debian packages ?

Yes, I find that the packages tend to break things, and I don't have the time 
to work out why or how I can fix it.

(Incidentally, the problem is that I can't get 3D and 2D acceleration on the 
Radeon 7500 at the same time. Enabling DRI disables 2D acceleration; 
disabling DRI disables 3D acceleration.)

--jaa