Re: [Gnu-arch-users] [OT] Architectural renovation

[Andrew Suffield]

On Thu, Aug 28, 2003 at 10:04:51PM +0900, Stephen J. Turnbull wrote:
>     >> Starting from today's XEmacs, three man-years.  The features
>     >> you mention are there, and performance is not a problem on
>     >> today's 512MB RAM, 1 GHz machines.
> 
>     Andrew> *cough* XEmacs is sluggish as hell on my
>     Andrew> 512Mb/1GHz/nforce2 box.
> 
> File a bug report.  Shouldn't be that way.

Easy to say, but I can't file a _useful_ bug report because I don't
know where the latency comes from, most of the time (font-lock-mode is
an extreme case; there's also something fishy with mule, or at least
japanese input/display, but I still haven't been able to identify
whether that's even xemacs at fault). I'd have to either spend about a
week dissecting all the elisp on my system until the problems stopped,
or learn enough about elisp in order to profile the damn thing
properly. I don't really have time or inclination for either.

I'm pretty sure I'm looking at a handful of bugs - but the nature of
emacs makes them difficult to even report, let alone locate and fix.

> You've picked the worst performance problem in XEmacs

Yup. Not a coincidence >:)

It's a good example of why performance is still a problem, and
probably always will be, regardless of how fast systems get. Sure, in
theory it could be fixed without making any huge changes, but in
reality it *isn't* fixed. There are various reasons for this, and I
find they're usually much more significant than the "underlying
architecture", in any program.

Or, more succinctly: what use is there in worrying about the
architecture if it can't solve the real-world problems?

-- 
  .''`.  ** Debian GNU/Linux ** | Andrew Suffield
 : :' :  http://www.debian.org/ |
 `. `'                          |
   `-             -><-          |

pgpu2BkVQAIHB.pgp
Description: PGP signature