Newsgroups: comp.windows.x
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!snorkelwacker.mit.edu!stanford.edu!leland.Stanford.EDU!leland.Stanford.EDU!fangchin
From: fangchin@leland.Stanford.EDU (Chin Fang)
Subject: Re: Roell 386/vga Server
Message-ID: <1991Jun25.225419.8734@leland.Stanford.EDU>
Sender: news@leland.Stanford.EDU (Mr News)
Organization: AIR, Stanford University, CA 94305 USA
References:  <1991Jun25.132628@wsl.dec.com>
Date: Tue, 25 Jun 91 22:54:19 GMT
Lines: 31

In article <1991Jun25.132628@wsl.dec.com>, skk@wsl.dec.com (Stuart Kreitman) writes:
|> 
|> Can anyone comment on the performance and quality of the Roell
|> Server for X.V11R4 386 [extended] vga?  I'm interested in how it stacks
|> up against ISC's vga.  Is ISC shipping R4 yet?
|> 
|> By the way, I have Roell's own announcement of February 11 which includes some
|> benchmarking data.
|> 
|> vrooom!

Your own last description is VERY accurate :-).  IT is THE FAST/BEST X11R4
server for 386 UNIX boxes using SVGA boards (and 8514 compatibles).  It's so
fast on a say 33Mhz 386 box the responses feels just like a SPARC 1+, assuming
both have enough memory (386 8megs, SUN 16 megs :-(

You might be able to beat it by paying tons of money to some vendors whose 
products support some obscure but high power accelerators.

NO UNIX vendor's X performs so well on SVGAs. (unless they steal Thomas' codes :-)
Remember I said SVGAs.  Please don't tell me some guy's X is faster on a TI board
etc as I am using a 50Mhz TI board now on an IBM RISC 6000.

Conclusion: for a freely available 386 Unix server, it's the KING.

Sincerely,

Chin Fang
Mechanical Engineering Department
Stanford University
fangchin@leland.stanford.edu