Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!bloom-beacon!mit-vax!jim
From: jim@expo.lcs.mit.edu (Jim Fulton)
Newsgroups: comp.unix.ultrix
Subject: Re: DEC windows/X windows compatibility
Summary: MIT server not mothballed
Message-ID: <5947@mit-vax.LCS.MIT.EDU>
Date: 12 Apr 89 13:36:16 GMT
References: <3104@stpstn.UUCP> <2628@decuac.DEC.COM> <10871@paris.ics.uci.edu> <28757@ucbvax.BERKELEY.EDU>
Sender: jim@mit-vax.LCS.MIT.EDU
Organization: X Consortium, MIT Laboratory for Computer Science
Lines: 22


>                                   And my guess was that with DECwindows
> available, the MIT people would no longer feel the need to improve on their
> original sample server.  Anyone know if this would be a correct assumption?

Yes I know, and no it would not be.  However, we probably attach a slightly
different meaning to "improve"....

We continue to work on making the generic framebuffer code (mi, mfb, cfb) as
portable and as correct (i.e. conforming to the X Protocol) as possible.  While
we don't try to speed up individual platforms, we do try to incorporate
optimizations done by others whenever we can verify that they will work across
a wide range of platforms. 


						Jim Fulton
						MIT X Consortium


p.s.  You can speed up the MIT server by using the GNU C compiler and the
Purdue speedups done by Gene Spafford at Purdue and Martin Friedmann at
CITI/UofMich.