[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: hmadorf on Lisp vs Fortran - declarations

On Oct 29, 12:04pm, Peter Paine wrote:
> Many a major Lisp application runs with negligable consing or GC.

An example: G2, a commercial tool with capabilities for _REAL TIME_
simulation and control, inference, and object-oriented user interface
construction, was (until just recently) written in lisp.  The vendor
claimed that they were able to achieve real-time performance in lisp by
writing code which did _no_ consing (using explicit resource management,
I'm sure).

[My apologies in advance if this looks like an advertisement; that was
not my intent.]

Footnote: G2 has recently been reimplemented in C (I think).  The
main justification (that I'm aware of) was to reduce the memory
requirements, which it did by a large factor (I don't recall the
numbers).  I would be curious to know if there was any significant
performance increase, other than what could be attributed to reduced

> So, to my point, why don't Lisps issue an automatic compiler warning
> when declarations are being ignored. 

Yes!  At least as an option.

George Williams            BCS Huntsville Artificial Intelligence Center
Boeing Computer Services   Internet: george@hsvaic.boeing.com
POBox 240002, M/S JY-58    UUCP: ...!uw-beaver!bcsaic!hsvaic!george
Huntsville AL 35824-6402   Phone: 205+464-4968 FAX: 205+464-4930