Date: 05 Aug 2002 02:34:49 -0700 From: Eric Anholt <eta@lclark.edu> To: Andrew Kolchoogin <andrew@snark.rinet.ru> Cc: current@FreeBSD.ORG Subject: Re: About 5.0 and Nvidia drivers Message-ID: <1028540090.483.412.camel@anholt.dyndns.org> In-Reply-To: <20020805085951.GA13606@snark.rinet.ru> References: <20020805055547.69959.qmail@web20708.mail.yahoo.com> <1028535362.483.311.camel@anholt.dyndns.org> <20020805085951.GA13606@snark.rinet.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 2002-08-05 at 01:59, Andrew Kolchoogin wrote: > Guys, > > yet another point. We should consider that many people already HAS some > kind of NVidia hardware and doesn't plan to change their well-tested video > card. Sure. Some people in this thread were saying "there's no ATI support / the ATI support is very bad" when it wasn't correct. So I wrote a little about the current ATI support. > I think that developing DRI-compatible (or even compatible with itself > throught wall outlet) driver for NVidia video cards is very good initiative. > > Although reverse engeneering (thing that this thread has begun from) is > very complicated and time-consuming process, if NVidia doesn't wish to > collaborate with software developers, it should be done. There's the old, slow, DMAless utah drivers (GL HW acceleration for X 3.3) for anyone who wants to port them to the XFree 4 with DRI. I remember like 17fps in quake3 on some low settings with a TNT2U 32MB/400Mhz Celeron. We even had some code released by nvidia that did sample dma (iirc, this was quite a while ago), but nobody got it working with utah. If open-source nvidia support didn't get done then when there was no nvidia binary driver, it will *definitely* not get done now. Folks with FreeBSD and NVidias probably will just wait for the binary drivers to work, which I understand will be in the not-too-distant future. Those are just my thoughts. If someone managed to make a working open-source nv dri driver that had even reasonable speeds, I would be very excited and happy. However, probably the place to start wouldn't be reverse-engineering if the target is the DRI, as PIO-based source already exists for at least some of the target cards. -- Eric Anholt <eta@lclark.edu> http://people.freebsd.org/~anholt/dri/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1028540090.483.412.camel>