*BSD News Article 25427


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!spool.mu.edu!agate!dog.ee.lbl.gov!hellgate.utah.edu!fcom.cc.utah.edu!u.cc.utah.edu!cs.weber.edu!terry
From: terry@cs.weber.edu (A Wizard of Earth C)
Newsgroups: comp.os.386bsd.misc
Subject: Re: FreeBSD 1.0R, XFree-86 2.0, and Diamond Speedstar
Date: 30 Dec 1993 02:26:59 GMT
Organization: Weber State University, Ogden, UT
Lines: 61
Message-ID: <2fte9j$59s@u.cc.utah.edu>
References: <2fqcdg$isp@ulowell.uml.edu> <1993Dec29.041309.17997@pool.info.sunyit.edu> <CIt0rC.F49@usenet.ucs.indiana.edu>
NNTP-Posting-Host: cs.weber.edu

In article <CIt0rC.F49@usenet.ucs.indiana.edu> pitts@mimosa.astro.indiana.edu (Jim Pitts) writes:
>>: >I am having trouble with getting XFree-86 2.0 running on my FreeBSD 1.0R
>>: >system. I have a Diamond Speedstar ET4000 card. No Plus, no II, no 24, just
>>: >Speedstar. The monitor is a ADI MicroScan 3E
>>
>>: Whoops...correction...it IS a Speedstar Plus...
>>
>>Diamond video cards suck!

[ ... ]

>Diamond video cards do not suck.  They just don't support the software you
>want the way you want it supported.  As far as video cards go they are
>actually very nice.  Your problem is with a lack of free support for the
>product, not a technical problem with the card itself.

Actually, there is a minor technical problem that is *why* the programming
information is not generally available -- Diamond wants to be able to
change the PALs, and that implies a change to the BIOS (or the program
that is trying to directly program the clocks) at the same time.  Diamond
can only guarantee a match between the PALs and the BIOS clock setting
code; the *CAN'T* guarantee a match between the PALs and your (free and
probably older than both the PALs and the BIOS) clock setting code and/or
seperate program.

Of course, the problem is biting or going to bite the user in NT and OS/2
and probably Chicago, as well... so supporting the alteration of PALs by
making the BIOS/PAL interaction hidden is NOT a good thing, unless they
go on to make their BIOS work in both protected and non-protected mode
(there are some ethernet cards which do this, for instance).  This is
not likely to happen, since the far instead of near references and the
base register reloading and unloading will give a performance hit to
their DOS users -- their primary market.

I don't know what Diamond plans to do with NT and OS/2, other than either
not support them or ship their own drivers and make sure they also match
the PALs on the particular card.


Ideally, from a free software standpoint, you wouldn't want to let them
change their PALs, so that the same clock programming code would work
from card to card (same goes for "ideally" from the NT and OS/2
perspectives as well).


>Of course as an owner of Diamond products I am hurt by this messy situation,
>but those are the breaks.  I chose to buy Diamond.  I hope that in the
>future Diamond will see that guarding the programming of their clock chips
>is not necessary.  When this happens I also hope the XFree86 team will be
>willing to admit when they have won a battle.  I am not holding my breath.

Definitely don't; even if they provide the information in the future as
to *how*, the specific *values* will vary from PAL rev to PAL rev -- a
card probe or BIOS is the only consistant way to deal with it.


					Terry Lambert
					terry@cs.weber.edu
---
Any opinions in this posting are my own and not those of my present
or previous employers.