Keelan Lightfoot wrote:
>
> >On Tue, Dec 28, 1999 at 10:51:38PM -0700, Keelan Lightfoot wrote:
> >> My brother is in the desktop publishing business, and *trust me*, a faster
> >> computer IS needed :)
> >
> >He's probably keeping up with the latest slow software then.
>
> If he doesn't keep up with 'the latest slow software' then he will have
> problems with files that clients are sending to him. This is really only a
> problem with Xpress...
>
> The software isn't slow. When you start working with a full page 11x17 600
> DPI image in photoshop, the extra processor clout sure comes in handy.
>
> >Coders who are given fast computers generate slow code.
>
> I disagree. I do all my programming work on a fast computer, then every so
> often, compile and send the file over to a slower computer (this 100 MHz
> thing, for example, or a 16 MHz SE/30) and test the software to see how I
> could speed it up.
>
> - Keelan Lightfoot

Good enough for me <G>

The point in "Coders who are given fast computers generate slow code."
being that, if you only have a "super-machine", you'd not be motivated -
or ABLE - to realistically speed test the code.  It's an excuse to not
Beowulf all my machines, I think <G>

I once wrote a Windows 3.x app that was designed to run on a 386sx33+,
4+mb Ram, etc.;  My boss got a 386sx16 laptop with 2Mb RAM given to him,
and wanted to see if my app would even load.  It did, no crashes,
connected to the embedded controller simulator properly (we were
off-site), and ran rather well, I thought.  Yep, it was SLOW, paged the
heck out of the HDD too.  To me, it was pretty rewarding to see that
thing run at ALL, though.  Especially compared to how unstable and
cruddy the code had been when I inherited it <G>  If I'd only run &
tested it on one machine, it wouldn't have done that.

(Learned this habit from writing code that can eat the FAT / file
system, but it works well) - I always Compile on a faster Development
machine, and test on a slower "testbed" machine, and keep another
machine or two around for various things like note-taking, formatting a
Floppy while testing new "patch" code, for example - then when the patch
is done I can pop the code onto that floppy across the LAN here & be
done in less time.  And, no notes CAN get lost, as they're on the aux
machines' HDD.  Some of those developers who only use one P3-750 (or
whatever) machine get to thinking that everyone else has 256Mb RAM in
their machine too, and they're flat wrong.  (Others who just use one
machine, know better, I'm just "hooked" on multi-machine here.)

You can use a monitor sharing device, if you want, that's fair <G>

However you code, whatever resources you use - It's great to see others
dedicated to the art of software <G>

  Mark

--
I re-ship for small US & overseas businesses, world-wide.
(For private individuals at cost; ask.)