Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!usc!wuarchive!uunet!orca.dsd.es.com!javelin.es.com!bgeer
From: bgeer@javelin.es.com (Bob Geer)
Newsgroups: comp.os.msdos.programmer
Subject: Re: DOS and the 640K limit
Message-ID: <1990Dec21.162409.363@javelin.es.com>
Date: 21 Dec 90 16:24:09 GMT
References: <921@VAX1.CC.UAKRON.EDU> <111240018@hpcvlx.cv.hp.com>
Reply-To: bgeer%javelin@dsd.es.com
Organization: Evans & Sutherland Computer Corp., Salt Lake City, Utah
Lines: 23

bill@hpcvlx.cv.hp.com (Bill Frolik) writes:
>...  For some reason they picked A000:0000
>as the address beyond which all this system stuff lives, which gives
>us the 640K "limit".

When the first IBMPC was designed, 640k was 10 times the amount of
memory in the then-common Z80/6502/6809-based pc's, and "who'd ever
use *that* much memory?"  Also, 16k-bit chips were the norm (during
the design cycle -- 64k-bit chips were just getting reliable as the
IBMPC was finally available) & were at least as expensive in late-70's
dollars as 1meg chips are today.

While processors have gotten faster, memory chips are more dense &
cheaper, disks are bigger, modems are faster, we're still living with
the restrictions of a late-70's design of bios.  This bios is the
reason the 286 chip didn't achieve its full potential, & the 386 has
such a great advantage because it's design was driven in part to
compensate for the bios design shortcomings.
-- 
<> Bob `Bear' Geer <> bgeer%javelin@dsd.es.com...dsd.es.com!javelin!bgeer <>
<>      Alta-holic <>   speaking only for myself, one of my many tricks   <>
<> Salt Lake City, <>    "We must strive to be more than we are, Lal."    <>
<>          Ootah  <>           -- Cmdr. Data, learning schmaltz          <>