lkml.org 
[lkml]   [1997]   [Aug]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectComments (was: Re: PentiumII's.)
Date
Vadim E. Kogan writes:
> XFree86 works pretty good.

True.. Xaccel run a _lot_ faster (I tried a demo the other day)
however it suffers the same problem all commercial X servers for Linux
seem to have - bad hardware support. With both Xaccel and MetroX I
can only get two buttons on my mouse as it's a dual-protocol thing
that needs the DTR line cleared to get into 3-button mode. XFree86
has supported this as far back as I can remember, the commercial ones
still don't. X with only 2 buttons? I don't think so...
Also it's difficult/impossible to edit the keymap - I have a 104-key
keyboard and want the "penguin" key instead of the Alt key to be Meta.
XFree86 does this no problems. Bad luck if you have a commercial server.

Anyway, enough with off-topic ramble...

> I think that was a story about buying "ALL LATEST". While Linux can make
> old & cheap stuff work sometimes better. For speed need something other

hear hear.. for AU$60 we scored a 8Mb 386dx40 and a 120Mb Quantum
drive which is now running as a print server and DNS. Funnily enough,
it appears to run a lot better than a 24Mb 486dx2/66 that used to be
doing the same jobs (it's now just router/squid proxy). Can't quite
figure that one out... but the response time is definately a lot
better on the 386.

Now, my comments on 2.1.4x - I started at .45 and am now running the
second pre-patch 2.1.48. .47 had a nice interactive performace boost
(my shell startup scripts ran instantaneously, even my huge zsh
compctl stuff) .48 wasn't quite as fast - but anyway with all of them
after around 20 hours or more of uptime they begin to bog
down... large programs like XEmacs and Netscape start killing the
system off, eg take 30 seconds or so to draw the windows, swap
workspaces, even changing focus from one to another app like rxvt
takes what seems like forever, and the hard disk swaps like crazy.
Is this an expected behaviour for 2.1.4x kernels? Has it something to
do with the new caching system not removing buffers? (I remember
reading a post by Linus Torvalds about something like that a while ago)

Curious,

-- Matt [matt@blitzen.canberra.edu.au]
"I believe there comes a time when everything just falls in line, We
live and learn from our mistakes; the deepest cuts are healed by faith."
-- Pat Benetar, "All Fired Up"


\
 
 \ /
  Last update: 2005-03-22 13:40    [W:0.163 / U:0.448 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site