home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!noc.near.net!hri.com!enterpoop.mit.edu!eru.mt.luth.se!lunic!sunic!aun.uninett.no!nuug!fredrik.atari.no!jornmoe
- From: jornmoe@fredrik.atari.no (Joern Moe)
- Newsgroups: comp.sys.m68k
- Subject: Busperformance of the 68030?
- Message-ID: <H.Qg_VCBlXLZo@fredrik.atari.no>
- Date: 13 Dec 92 21:17:36 GMT
- Reply-To: jornmoe@fredrik.atari.no
- Organization: None
- Lines: 20
- X-Software: HERMES GUS 1.03 Rev. Apr 14 1992
-
- The 68030 have a synchronous mode in which it does bus accesses in 2 clock
- cycles. Does this mode have overcapacity? I.e.: Can it fetch instructions
- and red/write data faster from/to the bus faster than the CPU can process
- them? (I know this will depend on the instructions but 'in general')?
-
- As far as I can see from the instruction-timing tabels in the manual I think
- the bus have overcapacity so that a bus designed to be zero waitstate
- (and synchronous) would cause some CPU-space's to be inserted.
- Now to the main question: Is there some 'rule' or 'expirienced factor' by
- which one can reduce bandwith by, and still gain ~100% CPU performance.
- I.e. how much of the bandwith of a zero-waitstate design can be used by
- dma devices (or other processors) without loosing CPU performance?
-
- Anny (qualified) opinion is appriciated!
-
- --
- _______________________________________________________
- / Joern F. Moe / All above is my own personal opinion! \
- / Oslo, Norway / Any lack of opinion above is also mine! \
- -----------------------------------------------------------
-