Path: ...!news.mixmin.net!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Thomas Koenig <tkoenig@netcologne.de>
Newsgroups: comp.arch
Subject: Re: Address bits again, Article on new mainframe use
Date: Mon, 2 Sep 2024 10:46:36 -0000 (UTC)
Organization: A noiseless patient Spider
Lines: 38
Message-ID: <vb452c$2qcil$1@dont-email.me>
References: <v9iqko$h7vd$1@dont-email.me> <vb060e$162j5$3@dont-email.me>
 <2377cad7ee947ad71c9c3a8afbcdc26e@www.novabbs.org>
 <vb3bdk$1r1t9$5@dont-email.me> <vb4314$jja$1@gal.iecc.com>
Injection-Date: Mon, 02 Sep 2024 12:46:37 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="8931b5407bde990b59eec3041f2a60a5";
	logging-data="2962005"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX19zQDWbHvwksvuKx2usIlh9TUcDx8mppKg="
User-Agent: slrn/1.0.3 (Linux)
Cancel-Lock: sha1:RMmgQanqnvgEbVSbcNAZr8vmdWY=
Bytes: 2697

John Levine <johnl@taugh.com> schrieb:

> It is true that the -11 died for lack of address space, but nobody I
> know has ever come up with a good design where the address size is
> bigger than the word size.  You end up with segments as on the 286
> or bank switching which is what later -11's did.

What about the 6502?  That was very much a 8-bit CPU, but
cleverly used its zero page as index registers.  Or all the other
8-bit CPUs, which usually had 16-bit address registers.

People could have built CPUs with 16-bit accumulators and 32-bit
addresses (or 8-bit accumulators and 32-bit addresses).  It is
debatable if these would have counted as a good design, though.

> VAX stood for Virtual Address Extension.  The key improvement was the
> 32 bit addresses.  Everything else was a detail.  Some of those details
> were unfortunate but that's a different argument.
>
> Also don't forget that back in that era everyone who had disks used
> overlays. The IBM mainframe linkers had complicated ways to build
> overlays and squeeze programs into 64K or whatever.

Overuse of overlays almost sank OS/360.  Brooks recounts that,
on a high-end machine, the FORTRAN compiler would have compiled
only a few lines per minute.  It was caught in time by simulation.

But the MVS linker was slooooooooooow.  I remember waiting for fifteen
to 30 minutes of wall time for a link step to complete, on a much
later model.

> Even though the
> address space was 16MB it was a long time before machines had that much
> RAM and by then they'd added paging to make the physical memory size less
> relevant.

Browsing through Wikipedia, it seems the biggest memory on a /360
was 6 MB.