Xref: utzoo comp.misc:2948 comp.arch:5857
Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!nrl-cmf!cmcl2!rutgers!bellcore!tness7!tness1!sugar!ficc!karl
From: karl@ficc.UUCP (karl lehenbauer#)
Newsgroups: comp.misc,comp.arch
Subject: Re: History of PCs (also kind of long)
Keywords: history, pc, workstation
Message-ID: <1193@ficc.UUCP>
Date: 2 Aug 88 14:19:14 GMT
References: <5946@venera.isi.edu> <5458@ecsvax.uncecs.edu> <1876@looking.UUCP> <753@applix.UUCP>
Organization: Ferranti International Controls
Lines: 24

In article <753@applix.UUCP>, scott@applix.UUCP (Scott Evernden) writes:
> My Cosmac-clone (anyone remember the BYTE ads for the Quest ELF?) was
> one of the funnest toys I ever had.  16 absolutely general-purpose
> 16bit registers, etc. etc...

As I recall, the RCA 1802 has eight 16-bit registers.  Although they were 
"general purpose" as there were no architectural restrictions on what 
registers could be used for what operations, as Intel parts are known for, 
with its 8-bit accumulator, no multiply or divide instructions and no 
subroutine call, it was kind of hosed.  Moving a register to a register was 
four instructions (e.g. GLO 4, PLO 5, GHI 4, PHI 5), register-indirect to 
register and register to register-indirect were about eight.  The polyForth 
environment was awesome considering the part.  We sometimes ran two users
developing code on a 24 KB system; polyForth was multitasking/multiuser.
The 24K system could cross-compile a 48K target.

The 1802 achieved a lot of its success by being one of the earliest, if not
the earliest, CMOS processor.  It's low power consumption made it idea for
solar powered applications like satellites and offshore platforms.  RCA 
later came out with an 1804 which, as I understand it (I never saw the part), 
fixed some of the problems with the chip, but it didn't catch on as it 
wasn't pin compatible.
-- 
-- +1 713 274 5184, karl@sugar.uu.net (uunet!sugar!karl)
-- Ferranti International Controls, 12808 W. Airport Blvd., Sugar Land, TX 77478