home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!europa.asd.contel.com!gatech!swrinde!zaphod.mps.ohio-state.edu!darwin.sura.net!jvnc.net!primerd.prime.com!hammer!cummings
- From: cummings@hammer.prime.com (Kevin Cummings)
- Subject: Re: Need working Xconfig for Diamond Stealth
- Message-ID: <1992Dec15.170923.4087@primerd.prime.com>
- Sender: cummings@hammer (Kevin Cummings)
- Organization: Prime Computer, Inc.
- References: <crh.723882424@regent.e-technik.tu-muenchen.de> <SVERKERW.92Dec11051515@meryl.csd.uu.se>
- Date: Tue, 15 Dec 1992 17:09:23 GMT
- Lines: 78
-
- In article <SVERKERW.92Dec11051515@meryl.csd.uu.se>, sverkerw@meryl.csd.uu.se (Sverker Wiberg) writes:
- > crh@regent.e-technik.tu-muenchen.dbp.de (Christian Huebner) writes:
- > My Diamond Stealth works fine with 640x480x256, but I am unable to
- > get a higher resolution out of it.
- >
- > I dont have any technical reference of the card (poor manual), so I
- > am unable to get the clock chip reference right. Experimenting has done
- > me no good so far.
- >
- > Uhoh, the poor wretch is S***t Out of Luck.
-
- Not entirely.
-
- > The trick of getting the clock freq. right on Diamond Stealth is by
- > doing some heavy, trade-secret, wizardry that the manufacturer won't
- > talk about. This means that Linux won't support Diamond Sealth.
-
- Not quite true. Yes, the trick is getting the clocks set up right. The
- Diamond Stealth uses the new Diamond proprietary PLL clock synthesizer
- to create the clocks it needs on the fly (except for the 25 and 28 MHz clocks).
- Code exists that claims to be capable of programming the PLL so that the
- clocks can be set. The XS3 people don't want to integrate this code
- into the X386-S3 driver. The same PLL is being used by Diamond on their
- Speedstar, Speedstar-24, and Speedstar-24X boards as well. The XFree86
- people don't want to integrate the clock programming code either! It has
- to do with Diamond's "proprietary and secret" clause that has them miffed.
- Diamond will provide the code, but at the cost of a Non-Disclosure Agreement
- that won;t allow them to release the source code necessary to build it.
- No-one wants to be responsible for compiling lots of binaries into the
- X distribution (can't say that I blame them), and noether do they want to
- try and use some-else's code that has been re-engineered (disassembled)
- from the VBIOS on the SS24X card.
-
- However, you are not yet SOL if you own a Diamond Stealth, but you may not
- be happy with what you have to do in order to run X in the resolution you
- want to.
-
- The following is known to work (by me, I do it). It was posted a while back
- by someone else (sorry, I've forgotten his/her name).
-
- Boot MS-DOS on your machine, and run Windows-3.x in the resoltution you
- want to run X in. While in Windows, give it the infamous three-fingered
- salute (Ctrl-Alt-DEL, twice if necessary to re-boot your machine). This
- time, boot into LINUX. Your extended clock should still be programmed
- for the resolution you were just running in MS-Windows. login to linux,
- and start up X using the X386-S3 server. If your Xconfig file has the right
- specification for the THIRD clock, it should work fine. (It does for me!)
-
- > I don't remember if the Diamond Stealth has made its way into the
- > FAQL, byt it's most definitely a FAQ.
- >
- > At the moment I have resorted to put a Trident 8900 (Yaaawwwnnnnn!!!)
- > into my 486 which works correctly in every mode.
- >
- > I hope somebody out there can solve my problem by either mailing
- > me the Xconfig or posting it on comp.os.linux.
- >
- > Sell the Diamond (to a Windows-nerd?) and buy yourself a ET4000-card.
- > They are fast.
-
- Yes, but not as fast as the S3 at doing bitblits (window moves). The S3
- driver also makes use of the S3 line drawing speedups as well! And according
- to the Xstones just recently posted, even the Diamond Stealth should have
- faster screen capabilities than a SPARC-1! How about that!
-
- Now if only I can convince the XS3 or XFree86 team to program the PLL for us!
-
- =================================================================
- Kevin J. Cummings ComputerVision Services
- 20 Briarwood Road 500 Old Connecticut Path
- Framingham, Mass. Framingham, Mass.
-
- Home: cummings@kjc386.framingham.ma.us Work: cummings@primerd.Prime.COM
-
- Std. Disclaimer: "Mr. McKittrick, after careful consideration,
- I've come to the conclusion that your new
- defense system SUCKS..." -- War Games
- =================================================================
-