scsh-users
[Top] [All Lists]

Re: I/O timings

To: scsh@martigny.ai.mit.edu
Subject: Re: I/O timings
From: shriram@mahasamatman.cs.rice.edu (Shriram Krishnamurthi)
Date: 21 Apr 1996 01:52:25 -0500
Organization: Computer Science, Rice University
Reply-to: shriram@cs.rice.edu
sjenkins@iastate.edu (Steven L Jenkins) wrote:

> The performance of SoftScheme is much worse.

Soft Scheme has a _lot_ of code.  The thing that really wins under
Chez is quite simply that the code is compiled.  In Scm, you have to
even go through the _parsing_ phase each time (and there are lots of
really hairy macros, such as match, to make that process slow).  I
haven't tried it, but my guess is it would load pretty quickly under
MzScheme, which can avoid this step each time; I know Cormac's system,
which is probably comparable, loads quickly.

>                                               My reference to the DrScheme/
> mzscheme family is because that system is being designed to include lots
> of useful development tools, and SoftScheme could be a candidate for
> inclusion in there.

There's a "substitute", ie, Cormac Flanagan's SBA analyzer.  If you
get DrScheme (http://www.cs.rice.edu/CS/PLT/packages/drscheme/),
there's an Analysis button at the top.  Click and enjoy.

> Yes, I guess I understand how it is meant to be used, but to me,
> it would be quite nice to have a SoftScheme-like tool available
> in a Scheme development system

Right.  That's precisely what we already have.  Of course, we mask the
load time for Cormac's analyzer by bundling it with the cost of
loading the rest of DrScheme.  It doesn't seem that unreasonable for
the latter to take several seconds.  Actually, we're not cheating;
Cormac's is only a small portion of the load-time.

Thanks for your comments.  I've redirected future discussion to
comp.lang.scheme, since this is off-topic for comp.lang.scheme.scsh.

'shriram

<Prev in Thread] Current Thread [Next in Thread>