Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!zaphod.mps.ohio-state.edu!samsung!uunet!mstan!rna!kc
From: kc@rna.UUCP (Kaare Christian)
Newsgroups: comp.sys.ibm.pc.misc
Subject: Re: Microsoft And Friends
Summary: More about 64k limit
Message-ID: <1048@rna.UUCP>
Date: 6 Dec 90 14:19:16 GMT
References: <4624@gara.une.oz.au> <1990Nov9.193424.1196@amd.com> <1990Nov25.003037.5814@xrtll.uucp>
Organization: Rockefeller University - Neurobiology
Lines: 26

In article <1990Nov25.003037.5814@xrtll.uucp>, silver@xrtll.uucp (Hi Ho Silver) writes:
> In article <2938@gmuvax2.gmu.edu> 2113av@gmuvax2.UUCP (John Porter) writes:
> $As has been mentioned too (painfully) frequently, X and other large
> $programs are a long ways off for Coherent, since the compiler currently
> $only compiles small model.  They managed emacs and kermit in 64k...
> 
>    The reason why Coherent is restricted to 64k segments is that it's a
> 286 operating system, not a 386 one, and the 286 doesn't allow segments
> to be larger than 64k.  A future version for the 386 is promised ...

The original poster is more on target -- coherent only suports small model.
Other compilers support multiple (small, medium, compact,
large) models, and thus support much larger programs.  What I wanted to know,
and what the marketing types I talked to couldn't tell me,
is how they did a full 16 bit compress with a small model compiler. (It
might be asm hacking.)
> 
>    BTW, for those interested in some critical reviews of Coherent, BYTE
> and PC Magazine have both discussed it recently; in each case, the
> conclusion was (to put it politely) that it's not for everyone.
> -- 
I'm glad *someone* read my PC mag first look on coherent. It hasn't
generated much comment from readers. Your summary is just right -- it's
not for everyone.

Kaare Christian