Path: utzoo!mnetor!uunet!lll-winken!lll-tis!ames!ncar!gatech!bbn!diamond.bbn.com!mlandau
From: mlandau@bbn.com (Matt Landau)
Newsgroups: comp.windows.x
Subject: Sun X Server Performance [WAS Re: X11.2 win entry, exit events problem]
Message-ID: <10804@jade.BBN.COM>
Date: 7 Apr 88 02:48:38 GMT
References: <3335@fluke.COM>
Reply-To: mlandau@bbn.com (Matt Landau)
Organization: BBN Laboratories Incorporated, Cambridge, MA
Lines: 19
Keywords: X,events,bugs

In comp.windows.x (<3335@fluke.COM>), toma@tc.fluke.COM (Tom Anderson) writes:
>p.s. - While I have your attention, does anyone know when Sun is going to get
>around to improving the performance of the X server for Suns? I know they have 
>a bad case of NIH (Not-Invented-Here), and I also know they are planning to 
>support X via NeWS, but how hard can it be to support a real X server? 

Not to start any SunView vs X11 vs NeWS flamage or anything, but from what 
I've heard and seen, the merged NeWS/X11 server will BE a "real X server."  
The plan is in some sense to layer the NeWS and X11 protocol interpreters 
on top of a common window "forest" and a common kernel of imaging code.

Given that the underlying event handling and imaging stuff should be the 
same for the NeWS and X sides of the server, the performance for NeWS and
X11 programs should be comparable, except for the additional overhead that
may be imposed by having to ship lots of X events over the server/client
connection instead of loading some PostScript into the server and having 
more of the application handled directly for you.
--
 Matt Landau			Waiting for a flash of enlightenment
 mlandau@bbn.com			  in all this blood and thunder