Path: utzoo!mnetor!uunet!husc6!tut.cis.ohio-state.edu!bloom-beacon!gatech!hubcap!ncrcae!ncr-sd!hp-sdd!hplabs!otter!kers
From: kers@otter.hple.hp.com (Christopher Dollin)
Newsgroups: comp.arch
Subject: Re: Architectural analysis of RPM-40 for general usage [very long]
Message-ID: <780008@otter.hple.hp.com>
Date: 22 Mar 88 10:50:39 GMT
References: <1840@winchester.mips.COM>
Organization: Hewlett-Packard Laboratories, Bristol, UK.
Lines: 35

"davidsen@steinmetz.steinmetz.UUCP (William E. Davidsen)" said both:

|As nearly as I can determine, workstations are used for graphics,
|software development, word processing, ...

and:

| ... generally the programs run are less than 1 min cpu, less than 
| 2MB memory ...

I'm doing software development on a workstation. There are two or three
Poplog windows (mail, X-window playthings, project work); each of these is
at least 1.5Mb [overkill for mail and X, but there it is]. Compiling the source
of the program[*1] (Lisp -> loaded binary) takes about 5min - of course that's
only when I want to start afresh - and running the program on a reasonably 
large piece of data has, on occasion, taken as long as four hours; a rapid 
bout of optimisation followed! The program can eat VM like C eats sanity. A
quick check: popmemused=523244 longwords; that's 2Mb heap in use, not counting
the Poplog kernel, the Common Lisp susbsystem, and the loaded code of the
program, all of which are locked down and don't count as active heap.

Does using LaTeX count as word processing? If so, I have a document here that
takes about 5min elapsed (last time I counted) to format. That's probably
over 1min CPU - at least it's not far off.

I wouldn't say that this was "really large" stuff, but it beats the 1min 2Mb
limit. 

[*1] Sorry to be vague, but you know how it is.




Regards,
Kers                                    | "Why Lisp if you can talk Poperly?"