home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.19950329-19950528
/
000454_news@columbia.edu_Thu May 25 02:54:46 1995.msg
< prev
next >
Wrap
Internet Message Format
|
1995-07-31
|
3KB
Received: from apakabar.cc.columbia.edu by watsun.cc.columbia.edu with SMTP id AA15496
(5.65c+CU/IDA-1.4.4/HLK for <kermit.misc@watsun.cc.columbia.edu>); Thu, 25 May 1995 12:06:34 -0400
Received: by apakabar.cc.columbia.edu id AA14669
(5.65c+CU/IDA-1.4.4/HLK for kermit.misc@watsun); Thu, 25 May 1995 12:06:32 -0400
Path: news.columbia.edu!sol.ctr.columbia.edu!howland.reston.ans.net!gatech!news.mathworks.com!news.kei.com!nntp.et.byu.edu!news.provo.novell.com!news.cs.utah.edu!cc.usu.edu!jrd
From: jrd@cc.usu.edu (Joe Doupnik)
Newsgroups: comp.protocols.kermit.misc
Subject: Re: Terminal emulation incorrect?
Message-Id: <1995May25.085446.52214@cc.usu.edu>
Date: 25 May 95 08:54:46 MDT
References: <3n3koo$ref@highway.leidenuniv.nl> <3nrtfs$aoh@highway.LeidenUniv.nl> <1995May1.170516.121386@daniel>
Organization: Utah State University
Lines: 32
Apparently-To: kermit.misc@watsun.cc.columbia.edu
In article <1995May1.170516.121386@daniel>, pcoen@daniel.drew.edu (Paul Coen) writes:
> In article <3nrtfs$aoh@highway.LeidenUniv.nl>, noordam@rulcri.LeidenUniv.nl (STEPHAN NOORDAM) writes:
>> It looks like the problem has been solved. Not loading the
>> Compaq device 'power.exe', which came with my laptop,
>> seems to do the trick. (If I load it low, instead of high,
>> things get even worse).
>>
>
> We had some similar problems with DEC notebook computers here. POWER.EXE
> caused no end of problems, including dropped characters. Removing it
> fixed the problem.
>
> Part of me really wants to know what power.exe is doing, and part of me
> would really, really rather not know at all.
-----------
Lacking a laptop computer and all the fun associated with them
I must restrict myself to guesses. But if I were writing a Power.exe
program I'd hook interrupts left, right, and center. Then I'd grab the
timer tick interrupt (Int 8). Each "interesting" interrupt would pass
through my code, I'd get the time of day (slow) and update a time of last
use table, and then pass on the interrupt to the previous owner. The
keyboard would be in that chain too, and DOS itself polls the heck out
of the keyboard (to which we've added our overhead now). Each of these
interceptions is at interrupt level and typically has interrupt recognition
turned off to prevent total disaster. Int 8 ticks 18.2 times/sec. Windows
uses interesting interrupts at an incredible rate. Excellent programmers
can pull off these stunts with little overhead; other folks plod along.
Add screen saver busywork, add DOS PRINT spooler cpu hogging, add
SMARTDRIVE cpu hogging, MOUSE driver polling, delays changing between
real and protected mode (above 1MB stuff) and here we are: a 486 working
at XT 8088 speeds.
Joe D.