[rescue] Perverse Question

Jonathan C. Patschke jp at celestrion.net
Mon Jun 9 17:02:39 CDT 2003


On Mon, 9 Jun 2003, Frank Van Damme wrote:

> On Monday 09 June 2003 19:55, Jonathan C. Patschke wrote:
> > ...don't do that.  Send it to someone who'll at least use it to
> > profile/benchmark his code on a relatively slow system.
>
> That kinda programmers still exist? >:-|

Yeah, me.  I think Dave McGuire falls into that category, as well.
It's a necessity for him, since he does embedded stuff.  For me,
it's a matter of pride.  Code that takes more CPU cycles or memory
pages than necessary is code with a bug, IMO.

That, and outright disgust at the fact that the average desktop computer
has more computing power that we used to put men on the moon, but runs
like ass when you put the newest Microsoft OS or X11 "Desktop
Environment" or Mozilla build on it.  Someone has to draw a line in the
sand and say "ENOUGH!  We WILL profile our code, thank you very fucking
much!"

Like Dave, George, and I have said repeatedly, 64MB of 1996 RAM is 64MB
of 2003 RAM.  Email is still, by-and-large, 7-bit ASCII.  The web is
still, by and large, HTML and JPEGs.  Wordprocessing hasn't changed
much.  Spreadsheets aren't much bigger.  Aside from video editing and
audio editing (which most people -don't- do, full-time), the things we
expect out of our computers haven't fundamentally changed in the last
seven years.  So, why was 64MB of 1996 RAM "living large", but 256MB of
2003 RAM "just getting by"?  It's sure not because the software is 3 or
4 times -better- than it was seven years ago.

-- 
Jonathan Patschke   )  "Leave your lawsuits at home.  I have guns."
Elgin, TX          (                      --"Kountry" Mike Lundgren



More information about the rescue mailing list