NNTP-Posting-Date: Tue, 31 Dec 2002 09:12:31 -0600
Date: Tue, 31 Dec 2002 09:15:04 -0600
From: gary drummond
X-Mailer: Mozilla 4.78 [en] (Win98; U)
Subject: Re: Which basic compiler to buy?
References: <email@example.com> <firstname.lastname@example.org> <email@example.com> <3E036EE0.DE68AFD7@sbcglobal.net> <firstname.lastname@example.org> <3E03DBA2.ED29B7D0@mfi.net>
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly
"Michael A. Terrell" wrote:
> Tom MacIntyre wrote:
> > On Fri, 20 Dec 2002 13:26:24 -0600, gary drummond
> > wrote:
> > >Tom Del Rosso wrote:
> > >>
> > >> "Tom MacIntyre" wrote in
> > >> message news:email@example.com...
> > >> > >
> > >> > >To be fair, IBM's hardware set the memory limit.
> > >> > >
> > >> >
> > >> > But wasn't DOS still incapable of exceeding the 640k long after the
> > >> > hardware support was there?
> > >>
> > >> That was mostly because of Intel's hardware goof. Their segmented
> > >> memory made it necessary for any program of more than 64k to do a lot of
> > >> address arithmetic, and that made it hard to migrate to a different
> > >> system with more segments. DOS couldn't really have been designed with
> > >> enough foresight to have been portable to the 286 memory model. Intel's
> > >> data on the 8086 didn't say a word about planning for the segments
> > >> becoming virtual, so the natural assumption was that consecutively
> > >> numbered segments were consecutive. In fact Intel encouraged that
> > >> assumption when they explicitly stated that segments and offsets could
> > >> be added different ways, for example the way 0040:017A = 0000:057A.
> > >>
> > >> Getting back to IBM's goofs, the original DOS made by a small company
> > >> which sold it to MS, had 1 meg of RAM, because they used a little 8086
> > >> system that could switch out the ROM! I don't know who made that
> > >> system.
> > >
> > >I think it was an 8088 at first...
> > >
> > >Gary
> > Memory may let me down again, but the 8086 was first, and they cut it
> > back on the external bus in the 8088. Anyone?
> > Tom
> Yes, the 8086 was first. The 8088 was a 8 bit buss version to allow
> the 8086 core to be used in cheaper designs. The original PC design was
> a close copy of the example design in the early Intel books, modified to
> use existing 8 bit cards.
> Michael A. Terrell
> Central Florida
I agree, the 8086 was manufactured and released first.
I was addressing the original PC/XT processors.
I had to find my old Byte mags...
Guide to the IBM Personal Computer (Fall 1984)
PC 4.77Mhz 8088 w/8087 socket, 256K Ram, 360K 5.25" Floppy.
XT 4.77Mhz 8088 w/8087 socket, 256K RAM, 10MB fixed disk, 360K 5.25"
QDOS was developed by Seattle Computing, and purchased by M$.
I think Seattle had an S-100 8086 system they sold...