Forgot your password?
Cyhist Feb 03 1997 B
Document Actions
Date: Mon, 3 Feb 1997 08:06:55 -0500
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
From: "Paul B. Schneck" <schneck@mitretek.org>
Subject: CM Re: Emulators; Word-Parallel Bus
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
Mike Williams wrote:
>However, it has always been my stand that what people REALLY mean by "von Neumann machine" is not just a stored program computer but A STORED PROGRAM COMPUTER WITH PARALLEL DATA TRANSFER.
>
>Anyone got any comments on that?
>
I don't believe that "parallel data transfer" is one of the criteria used to distinguish a vN machine from a calculator or other machine. For example, early IBM 360 systems (20) implemented only an 8-bit wide bus. Other systems (40) had a 16-bit wide bus. Larger systems (50, 65) had a 32-bit wide bus. and the largest (75, 85, 9x) had a 64-bit wide bus. They all have some degree of parallelism, but the "architecture" is vN, while the implementation is targeted at market (speed, price). I use architecture to mean the instruction set functionality as viewed by the programmer. The programmer could not discover the bus width by running a program. Paul Schneck
______________________________________________________________________
Reply-To: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
Sender: "CYHIST Community Memory: Discussion list on the History of
Cyberspace" <CYHIST@SJUVM.STJOHNS.EDU>
From: "Paul B. Schneck" <schneck@mitretek.org>
Subject: CM Re: Emulators; Word-Parallel Bus
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
Mike Williams wrote:
>However, it has always been my stand that what people REALLY mean by "von Neumann machine" is not just a stored program computer but A STORED PROGRAM COMPUTER WITH PARALLEL DATA TRANSFER.
>
>Anyone got any comments on that?
>
I don't believe that "parallel data transfer" is one of the criteria used to distinguish a vN machine from a calculator or other machine. For example, early IBM 360 systems (20) implemented only an 8-bit wide bus. Other systems (40) had a 16-bit wide bus. Larger systems (50, 65) had a 32-bit wide bus. and the largest (75, 85, 9x) had a 64-bit wide bus. They all have some degree of parallelism, but the "architecture" is vN, while the implementation is targeted at market (speed, price). I use architecture to mean the instruction set functionality as viewed by the programmer. The programmer could not discover the bus width by running a program. Paul Schneck
______________________________________________________________________
