brian@radio.astro.utoronto.ca (Brian Glendenning) (02/10/89)
I recently built X11R3 on our colour Sun 3/180 (cgtwo - no monochrome plane). While I don't expect it to be as fast as a monochrome system, it seems to be too slow - certainly slower than SunView. An xterm window or emacs window doesn't look much slower than my 2400 baud modem logins from home. I know about the comment in the sun README that the performance is "untuned". Is this in an indication that I've installed something incorrectly, or is X really that slow on colour only suns? (It takes lots of cpu as well). If I made no mistakes, I compiled everything with gcc and fixes 1 to 8 and the latest Purdue speedups (2.1). If colour suns really are this slow when will things improve? Is Sun selling X yet? Presumably it would be faster :-) I want to do some development for an astronomical imaging application, but it is very painful with the current X. Thanks. -- Brian Glendenning - Radio astronomy, University of Toronto brian@radio.astro.utoronto.ca uunet!utai!radio!brian glendenn@utorphys.bitnet
bob@tinman.cis.ohio-state.edu (Bob Sutterfield) (02/17/89)
In article <1989Feb10.105927.20572@radio.astro.utoronto.ca> brian@radio.astro.utoronto.ca (Brian Glendenning) writes:
If colour suns really are this slow when will things improve? Is
Sun selling X yet? Presumably it would be faster :-)
Yes, it really is that slow. No, they aren't shipping it yet -
they're projecting a July ship date, last I heard (Miami Beach SUG
meeting in December). Teitelman shared some benchmarks of their
progress tuning their server, and showed that it's already much faster
than the sample server, which isn't too hard to believe.jim@athsys.uucp (Jim Becker) (02/18/89)
From article <BOB.89Feb16120545@tinman.cis.ohio-state.edu>, by bob@tinman.cis.ohio-state.edu (Bob Sutterfield): > In article <1989Feb10.105927.20572@radio.astro.utoronto.ca> brian@radio.astro.utoronto.ca (Brian Glendenning) writes: > If colour suns really are this slow when will things improve? Is > Sun selling X yet? Presumably it would be faster :-) > > Yes, it really is that slow. No, they aren't shipping it yet - > they're projecting a July ship date, last I heard (Miami Beach SUG > meeting in December). Teitelman shared some benchmarks of their > progress tuning their server, and showed that it's already much faster > than the sample server, which isn't too hard to believe. At the local X User Group someone in the audience said that they were using an alpha version of the Sun server. They said that the monochrome version of the server is three times faster than the MIT sample server and the color version is five times faster than the MIT server. Sun is committed to having the very best server out there. -Jim Becker ...!sun!athsys!jim
jim@athsys.uucp (Jim Becker) (02/21/89)
From article <245@tityus.UUCP>, by jim@athsys.uucp (Jim Becker): > From article <BOB.89Feb16120545@tinman.cis.ohio-state.edu>, by bob@tinman.cis.ohio-state.edu (Bob Sutterfield): >> In article <1989Feb10.105927.20572@radio.astro.utoronto.ca> brian@radio.astro.utoronto.ca (Brian Glendenning) writes: >> If colour suns really are this slow when will things improve? Is >> Sun selling X yet? Presumably it would be faster :-) >> > At the local X User Group someone in the audience said that > they were using an alpha version of the Sun server. They said that the > monochrome version of the server is three times faster than the MIT > sample server and the color version is five times faster than the MIT > server. Sun is committed to having the very best server out there. > > -Jim Becker ...!sun!athsys!jim I was corrected by a representative of Sun. These are _stated_ performance goals for FCS, but may not represent the current numbers. I believe that this statement was made by a Daisy employee, not someone speaking for Sun Microsystems (but I may be wrong here again). This was at the Bay Area X meeting, at the Daisy cafeteria... How do I keep managing to put my foot in my mouth so bad? I will be moving from Athena to Sun, to help work on their X11/News effort. Then I can just shut up and help make the numbers fact... (oh happy day, oh happy day...) -Jim Becker ...!sun!athsys!jim
turner@daisy.UUCP (Jim Turner) (02/22/89)
The numbers that were quoted were most likely for our proprietary version
of the Xsun server. (actually the numbers sound a little low, I think
that we did better) Anyway Sun made a comitment to get X/NeWS to run at least
as fast as our server. To this time I have heard nothing about a pure X server.
However I am meeting with Sun today and maybe I'll have some fresh news for
y'all
--
Don't take life too seriously, it ain't nohow permanent - Pogo
...{decwrl|ucbvax}!imagen!atari!daisy!turner (James M. Turner)