[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: saving a numeric table to a disk file
- To: hmadorf@eso.org
- Subject: Re: saving a numeric table to a disk file
- From: mmeehan@eesof.com (Mike Meehan)
- Date: Tue, 16 Jun 92 10:32:56 PDT
- Cc: info-mcl@cambridge.apple.com
- In-reply-to: Mail from 'hmadorf@eso.org' dated Tue, 16 Jun 92 08:50:40 +0200
Hello Hans,
In performing non-classical optimization to systems with *large* number of designable
(and usually random) variables, I too have efficiency and resource requirements similar
to your own.
While I love the simplicity and power of lisp, I am coming to realize that applications
like ours are why they invented the FFI. If lisp itself has a weakness, I would say that
your observation hits the nail on the head... There is a compromise here, you dont get
something for nothing! Lisp makes it easy to implement and is efficient *most* of the
time.
I think that the way you are supposed to save your data is to (save-application ...).
My solution has been to use the FFI and write the number cruncher and heavy duty file i/o
activities in c. The good news is that you can make all that look like lisp! In my
opinion, the best set-up is to use think-c and Cartier`s MCL <-> TC interface.
(Especially since my upgrade to system 7 makes my mpw refuse to boot!)
Since it sounds like we are both new to lisp, I would very much like to hear the
comments of the more experienced lispers...
ttfn, MM
Mike Meehan
Simulation Optimization Technology Group
EEsof Inc., R&D Department
5601 Lindero Canyon Road
Westlake Village CA. 91362
FAX/VOICE: 818.879.6374
mmeehan@eesof.com