home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.apollo
- Path: sparky!uunet!utcsri!helios.physics.utoronto.ca!alchemy.chem.utoronto.ca!system
- From: system@alchemy.chem.utoronto.ca (System Admin (Mike Peterson))
- Subject: Re: Just what is in a Apollo 9000/730?
- Message-ID: <1992Nov22.162046.7561@alchemy.chem.utoronto.ca>
- Organization: University of Toronto Chemistry Department
- References: <1992Nov13.222325.9880@ccu1.aukuni.ac.nz> <9330037@hpfcso.FC.HP.COM> <1992Nov20.184357@helix.nih.gov>
- Date: Sun, 22 Nov 1992 16:20:46 GMT
- Lines: 17
-
- In article <1992Nov20.184357@helix.nih.gov> rvenable@helix.nih.gov (Richard M. Venable) writes:
- >In article <9330037@hpfcso.FC.HP.COM>, wlw@hpfcso.FC.HP.COM (Will >Walker) writes:
- >|> In comp.sys.apollo, rvenable@helix.nih.gov (Richard M. Venable) writes:
- >|> The "small number" is two. They are FMPYADD and FMPYSUB.
- >Only these two? And not a multiply as well? Seems a little trivial
- >compared to way it's described in Workstation (Feb91, p.44; Apr91, p. 1).
-
- The multiply-add combination instructions are probably one of the big
- reasons that the IBM RS/6000 workstations can leave the 7x0 series
- systems in the dirt on specific applications, despite otherwise
- similar performance on "general" applications. This instruction should
- also be helpful for veclib routines - programs that can use the IBM
- blas/vector routines get huge speed-ups, even over the same routines on
- our 7x0 systems.
- --
- What are the chances that any HP computer system will ever "work" properly?
- ... and Slim just left town. -*- Mike Peterson, SysAdmin, U/Toronto Chemistry
-