home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!news.centerline.com!jimf
- From: jimf@centerline.com (Jim Frost)
- Newsgroups: alt.folklore.computers
- Subject: Re: ASSEMBLY LANGUAGE IBM
- Date: 23 Nov 1992 19:11:09 GMT
- Organization: CenterLine Software, Inc.
- Lines: 16
- Message-ID: <1eracdINN728@armory.centerline.com>
- References: <BxtpIv.AMD@mentor.cc.purdue.edu> <STEVEV.92Nov17104240@miser.uoregon.edu> <NICKEL.92Nov17235221@desaster.cs.tu-berlin.de> <1992Nov18.104521.9036@news.columbia.edu>
- NNTP-Posting-Host: 140.239.3.202
-
- lasner@watsun.cc.columbia.edu (Charles Lasner) writes:
- >In any case, this is talking about control-store architecture, not machine
- >architecture. The implemented instruction set is the same as what was defined
- >on paper circa 1960 and simulated on 7090-class machines. As someone posted
- >earlier, the binary of any program from the beginning will run on each and
- >every model. This assumes that the implied O/S functions are present, as
- >many of these programs issue SVC instructions that have to be handled.
-
- Almost, but not quite. The model with instruction pipelining (the /65
- I think) couldn't run FP applications reliably because exceptions
- weren't precise.
-
- That sucker died fast though. :-)
-
- jim frost
- jimf@centerline.com
-