home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!uwm.edu!wupost!darwin.sura.net!jvnc.net!nuscc!eletanjm
- From: eletanjm@nuscc.nus.sg (TAN JIN MENG)
- Newsgroups: comp.arch
- Subject: Re: CACHE MISS PENALTY FOR 386/486??
- Message-ID: <1992Aug20.015043.7280@nuscc.nus.sg>
- Date: 20 Aug 92 01:50:43 GMT
- Article-I.D.: nuscc.1992Aug20.015043.7280
- References: <1992Aug19.200257.24361@nic.umass.edu>
- Organization: National University of Singapore
- Lines: 14
-
- Let me try to *narrow* down the options.
-
- I'm interested in order of magnitude only - I realise how difficult it
- might be.
-
- Lets define a "typical AT clone" as a non memory interleave system, main
- memory DRAM access with 1 wait state - no other special memory access
- scheme and the external memory cache is implemented in a standard manner
- regards INTEL recommendations.
-
- I've seen discussions on this topic re. RISC systems but have not seen
- one regarding x86 or 68xxx.
-
- jin meng
-