home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.next.hardware
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!uwm.edu!rpi!batcomputer!munnari.oz.au!mel.dit.csiro.au!mineng.dmpe.CSIRO.AU!dmssyd.syd.dms.CSIRO.AU!metro!basser.cs.su.oz.au!orthanc!jimu
- From: jimu@orthanc.cs.su.oz.au (James Uther)
- Subject: Re: Acorn 610 as processor/coprocessor?
- Reply-To: jimu@orthanc.cs.su.oz.au (James Uther)
- Organization: Basser Dept of Computer Science, University of Sydney, Australia
- Date: Sun, 15 Nov 1992 23:36:44 GMT
- Message-ID: <1992Nov15.233644.22719@cs.su.oz.au>
- References: <1992Nov8.130614.19372@wam.umd.edu> <BxL7t4.3tE@iat.holonet.net> <1992Nov12.134146.631@wam.umd.edu>
- Sender: news@cs.su.oz.au (News)
- Lines: 18
-
- In article <1992Nov12.134146.631@wam.umd.edu>, rsrodger@wam.umd.edu (Yamanari) writes:
- |> In article <BxL7t4.3tE@iat.holonet.net> bwilliam@iat.holonet.net (Bill Williams) writes:
- |> >It has only 32000 transistors yet can magically support 4096 bytes of
- |> >cache RAM in the 32000 transistors... hmmm lets see... lets pretend that
- |> I was referring to the Byte magazine article some months
- |> back. 32k transistors, 4k cache and no FPU.
- |>
- |> Can anyone cite a claim from Acorn themselves??
- no, but
- from what I hear, the 610 is just a subcomponent, or module. The 32k
- trannies represent the cpu, but an actual 610 uprocessor tends to be built
- with other things added. For example, the 610 for the newton started life
- by grabbing the 610 cpu module and sticking it in the corner of the silicon.
- Then a large object-based mmu with hardware garbage collection was designed
- and added. All in all far more than 32k trannies. The 610 is simply a core
- around which final uprocessors are designed.
-
- james
-