Path: utzoo!mnetor!uunet!husc6!cmcl2!phri!dasys1!raylau
From: raylau@dasys1.UUCP (Raymond Lau)
Newsgroups: comp.sys.mac
Subject: Re: C and large data model
Message-ID: <2256@dasys1.UUCP>
Date: 18 Dec 87 01:36:00 GMT
References: <8712152132.AA10832@decwrl.dec.com> <3566@husc6.harvard.edu> <36709@sun.uucp>
Organization: The Big Electric Cat
Lines: 38
Summary: LightspeedC has my vote


Personally, I am an Aztec to Lightspeed convert.  I don't regret the decision.
I find that the speed and performance of LS more than outweighs its disadvantag
es.

It is a LOT easier to quit, make a change and press cmd-R (run) esp under MF,
where transferring is instant... than quitting the application, running qued/m
...running cc...running ln... under Aztec.  I've found that the turnaround
time in making little changes is the majority of the development time.

Although Aztec has it too, LSC's inline assembly feature is nice.  (As a matter
of fact, I didn't really use lsc for anything before it)  Some things are just
better done in .asm.


On the negative side, I miss an intergrated debugger, although I don't make
use of them much anyway.  I hate the 32k data limit.  With a hash table in
the order of 200+k, the 32k limit seems to have been exceeded easily.  I
am not too thrilled about the generated code size, although Aztec's is only
marginally better.  And a host of other little things.


One last thing...I don't know}i about 020/881 support in LSC.  I guess that
that might be important soon (read that as "as soon as I get a machine w/such").


Overall, my vote is in for LSC.  And that's after shelling out the $ for Aztec
initially.

--
Raymond Lau                       GEnie: RayLau
100-04 70 Ave.                    CIS: 76174,2617
Forest Hills, NY 11375-5133       Delphi: RaymondLau
United States of America          MacNET: RayLau
uucp: raylau@dasys1.UUCP (..{phri,cucard,bc-cis,mstan}!dasys1!raylau})

s
much use of debuggers anyway.  I hate the