home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.intel
- Path: sparky!uunet!charon.amdahl.com!pacbell.com!mips!darwin.sura.net!Sirius.dfn.de!zrz.tu-berlin.de!cs.tu-berlin.de!news.netmbx.de!Germany.EU.net!mcsun!sunic!ugle.unit.no!nhl-1.nhl.sintef.no!nro
- From: nro@nhl-1.nhl.sintef.no (Nils Reidar Boe Olsen)
- Subject: Speed of P5
- Message-ID: <1992Aug19.170552.23411@ugle.unit.no>
- Sender: news@ugle.unit.no (NetNews Administrator)
- Organization: The Norwegian Hydrotechnical Laboratory
- Date: Wed, 19 Aug 92 17:05:52 GMT
- Lines: 13
-
- Rumours from this news-group says that the P5 processor
- will have a FPU which is an upgrade from the one that is
- used in the i860 processor. This should mean that the P5
- should run floating-point intensive programs faster than
- the i860. I am presently running a 3D CFD code on my PC,
- which has a 25 MHz 486 and a MicroWay Number Smasher i860
- card. The code runs 26 times faster on the i860 than on
- the 486 (no swapping to disk). So my question is: will
- the P5 really be at least 20-30 times faster than the
- 486 on floating-point intensive programs ?
-
- Nils Reidar Olsen
-
-