Forgot your password?
Cyhist Jan 24 1997 R
Document Actions
Date: Fri, 24 Jan 1997 22:14:10 -0800
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: Jay Hosler <jhosler@cisco.com>
Subject: Re: CYHIST Digest - 23 Jan 1997 to 24 Jan 1997
In-Reply-To: <199701250531.VAA00301@beasley.cisco.com> from "Automatic digest
processor" at Jan 25, 97 00:30:06 am
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
>Date: Fri, 24 Jan 1997 02:34:17 EST
>From: "Peter G. Capek" <capek@watson.ibm.com> Subject: Emulation
>
>______________________________________________________________________
>Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
>
>
>I'm a bit surprised at some of the comments about the invention of emulation which have been made here recently. The IBM System/360 line was one of the most well known, if not one of the first, uses of emulation in a commercial system: different models of System/360 emulated various of the earlier IBM systems, and I believe that in at least one case a later version of 360 emulated a competitive machine. All but one of the originally announced models of System/360 (30, 40, 50, 60; the model 70 wasn't) were in fact microcoded and included emulators for other architectures. For example, the model 30 could emulate the 1401 (and perhaps other machines), and the model 60 (later named the model 65 with slight modifications) could emulate the 7090/94.
>
>However, at least in IBM, the story is much older than that. There were emulators for Stretch on earlier machines, although due to memory constra= ints,
>I doubt they were very useful for any serious work. Someone commented on a modification to ENIAC which effectively caused it to emulate a differen= t
>architecture to make it easier to program. I believe the place where thi= s is
>described, albeit briefly, is in Goldstine's The Computer From Pascal to Von Neumann. This is one of the earliest pieces of work that I know of w= hich
>intentionally trades off performance for ease of programming.
>
>Peter Capek
>
During 1972, as a CSC employee, I was involved in a data-processing development done by CSC for Travelers insurance. At Travelers, from a long-tenure Travelers data-processing staff member, I heard an astonishing emulation story. Travelers first undertook to automate its master file in the mid-fifties; this development was done on the RCA 501. Two successive generations of RCA systems where implemented before Travelers inevitably switched to IBM. In the early phases of the IBM-based implementation, in the early sixties, 1401's were used to emulate production RCA code. In the mid-sixties, when 360's were installed at Travelers, they in turn emulated 1401 code, some of which was still emulating RCA code. I know of no earlier commercial use of two levels of emulation.
Jay Hosler
______________________________________________________________________
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: Jay Hosler <jhosler@cisco.com>
Subject: Re: CYHIST Digest - 23 Jan 1997 to 24 Jan 1997
In-Reply-To: <199701250531.VAA00301@beasley.cisco.com> from "Automatic digest
processor" at Jan 25, 97 00:30:06 am
______________________________________________________________________
Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
>Date: Fri, 24 Jan 1997 02:34:17 EST
>From: "Peter G. Capek" <capek@watson.ibm.com> Subject: Emulation
>
>______________________________________________________________________
>Community Memory: Discussion List on the History of Cyberspace ______________________________________________________________________
>
>
>I'm a bit surprised at some of the comments about the invention of emulation which have been made here recently. The IBM System/360 line was one of the most well known, if not one of the first, uses of emulation in a commercial system: different models of System/360 emulated various of the earlier IBM systems, and I believe that in at least one case a later version of 360 emulated a competitive machine. All but one of the originally announced models of System/360 (30, 40, 50, 60; the model 70 wasn't) were in fact microcoded and included emulators for other architectures. For example, the model 30 could emulate the 1401 (and perhaps other machines), and the model 60 (later named the model 65 with slight modifications) could emulate the 7090/94.
>
>However, at least in IBM, the story is much older than that. There were emulators for Stretch on earlier machines, although due to memory constra= ints,
>I doubt they were very useful for any serious work. Someone commented on a modification to ENIAC which effectively caused it to emulate a differen= t
>architecture to make it easier to program. I believe the place where thi= s is
>described, albeit briefly, is in Goldstine's The Computer From Pascal to Von Neumann. This is one of the earliest pieces of work that I know of w= hich
>intentionally trades off performance for ease of programming.
>
>Peter Capek
>
During 1972, as a CSC employee, I was involved in a data-processing development done by CSC for Travelers insurance. At Travelers, from a long-tenure Travelers data-processing staff member, I heard an astonishing emulation story. Travelers first undertook to automate its master file in the mid-fifties; this development was done on the RCA 501. Two successive generations of RCA systems where implemented before Travelers inevitably switched to IBM. In the early phases of the IBM-based implementation, in the early sixties, 1401's were used to emulate production RCA code. In the mid-sixties, when 360's were installed at Travelers, they in turn emulated 1401 code, some of which was still emulating RCA code. I know of no earlier commercial use of two levels of emulation.
Jay Hosler
______________________________________________________________________
