Path: utzoo!utgpu!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!bloom-beacon!oberon!orion.cf.uci.edu!elroy!cit-vax!tybalt.caltech.edu!palmer
From: palmer@tybalt.caltech.edu (David Palmer)
Newsgroups: comp.sys.mac
Subject: Re: Status of MS Word 4.0
Message-ID: <8985@cit-vax.Caltech.Edu>
Date: 28 Dec 88 00:41:56 GMT
References: <456@tekn01.chalmers.se> <482@nikhefk.UUCP>
Sender: news@cit-vax.Caltech.Edu
Reply-To: palmer@tybalt.caltech.edu.UUCP (David Palmer)
Organization: California Institute of Technology
Lines: 28

In article <482@nikhefk.UUCP> paulm@nikhefk.UUCP (Paul Molenaar) writes:
>I've had a pre-release of Word 4.0 which was fairly full of bugs
>and immensely slow due to the fact the all their P-Machine debugging
>code was still in the program. MS-Word 4.0 will, or so it seems, still
>be running on a P-Code interpreter. I wonder how they will speed up
>the program compared to Word 3.0x (which was interpreted P-Code as well).

I worked at Microsoft on Macintosh Word V 1.0 (way back in '84) on
program speed-ups.  (It gives me a good feeling to know that in one
summer I probably saved several lifetimes of useful work for all
the people using that program.)

The c-compiler Microsoft uses for its products generates P-Code, but
any block may instead be compiled to native 68000 code.  This means that
the routines which do things which may be done slowly (does it really matter
if it takes 10 milliseconds to change the cursor to a right-arrow instead
of 10 microseconds?) can use compact P-Code, while things that need to
be done quickly (e.g. screen redisplay) may take up more memory but
run as fast as possible.

The version you used might have been all P-code.  Perhaps they were
also using the principle 'Make it work, THEN make it fast'.  (Perhaps
they learned something from Word 3.0, 3.01,...)

		David Palmer
		palmer@tybalt.caltech.edu
		...rutgers!cit-vax!tybalt.caltech.edu!palmer
	"I was sad that I had no shirt, until I met a man with no torso"