*BSD News Article 75954


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!news.wildstar.net!cancer.vividnet.com!hunter.premier.net!news.mathworks.com!uunet!in2.uu.net!pell.pell.chi.il.us!there.is.no.cabal
From: orc@pell.chi.il.us (Orc)
Newsgroups: comp.os.linux.development.apps,comp.os.linux.development.system,comp.os.linux.x,comp.os.linux.hardware,comp.os.linux.setup,comp.unix.bsd.386bsd.misc,comp.unix.bsd.bsdi.misc,comp.unix.bsd.netbsd.misc,comp.unix.bsd.freebsd.misc,misc.consumers
Subject: Re: Why not buy Matrox Millennium
Followup-To: comp.os.linux.x
Date: 11 Aug 1996 23:08:24 -0700
Organization: By the Holy Claws of Klortho the Magnificent, this IS a fine morning!
Lines: 36
Message-ID: <4umhoo$ofp@pell.pell.chi.il.us>
References: <4j21ph$crr@slappy.cs.utexas.edu> <87zq43vvaj.fsf@xeno.xinside.com> <4ujl2j$2k6@pell.pell.chi.il.us> <873f1tjxf1.fsf@xeno.xinside.com>
NNTP-Posting-Host: pell.pell.chi.il.us
Xref: euryale.cc.adfa.oz.au comp.os.linux.development.apps:20242 comp.os.linux.development.system:29620 comp.os.linux.x:37796 comp.os.linux.hardware:47304 comp.os.linux.setup:68300 comp.unix.bsd.386bsd.misc:1031 comp.unix.bsd.bsdi.misc:4584 comp.unix.bsd.netbsd.misc:4363 comp.unix.bsd.freebsd.misc:25349 misc.consumers:94218

[followups narrowed substantially]

In article <873f1tjxf1.fsf@xeno.xinside.com>,
Thomas Roell <roell@xinside.com> wrote:
[I dislike X Inside resetting the console to 80x25x16 scanlines]

>Maybe I was not clear in previous postings. So let's get down to the
>technicalities again. It is not possible to read back the state of ALL
>graphics chip related registers. What XF86 does is majorly guesswork
>on registers that cannot be read back. Our experiments have shown that
>trying to guess about those registers leads in a high number of cases
>(read non-standard cases) to screwedup systems and total lockups.

I would suggest that this may not be the case for every chipset out
there.  I've got a small pile of workstations with various sorts of S3
and ATI cards in them, and the lot of them will, with XFree, go into the
graphics mode de jour and then return to the mode I set them to.  And if
I find a card where it doesn't work (like a ATI Mach 64 in a P6 I'm
using, where launching XFree leads to a dead console), well, shoot, I
certainly didn't pay anything for the privilege.

I can understand that there may be some cards out there that won't let
you read control registers, and that you can't get programming
information about so you can't deduce how they work, but it seems
foolish to penalise everyone because of these rarities.  It may be more
programming to set up your server to restore screen settings when it
terminates or you console-flip, but counterbalance that against the
sales you'll lose because you're mangling the console; I don't know
where the breakover point is, but I've already made an order conditional
on a version that doesn't mangle the console, and it seems that I am
not the only person who has brought this documented bug up, so I may
not be the only lost sale.

                 ____
   david parsons \bi/ orc@pell.chi.il.us
                  \/