home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.atari.st
- Path: sparky!uunet!tcsi.com!iat.holonet.net!news.cerf.net!usc!cs.utexas.edu!hermes.chpc.utexas.edu!news.utdallas.edu!corpgate!crchh327!bnr.ca!grier
- From: grier@bnr.ca (Brian Grier)
- Subject: Re: 030 sync cycles (...)
- Sender: news@news.rich.bnr.ca (news server)
- Message-ID: <C1GzxH.9L3@news.rich.bnr.ca>
- Date: Tue, 26 Jan 1993 16:46:29 GMT
- References: <1993Jan22.135725.4674@ugle.unit.no> <1983@enst.enst.fr>
- Nntp-Posting-Host: 131.253.206.80
- Organization: Bell Northern Research
- Lines: 27
-
- elendir@inf.enst.fr writes:
-
- > Stig Vidar Hovland writes
- >
- > ] Mr. Joern Moe has probably circuit diagrams for the Falcon and in that case
- > ] he should know what he is talking aboat. With additional hardware, it is
- > possible
- > ] to terminate an access to 16 bit memory with STERM.
-
- > Of course with additionnal hardware you can do everything. But I remember
- > something in the developer documentation (v 0.3, July 92) speaking of wait
- > states...
- > Anyhow, a 32 bit synchronous access in 4 cycles has 2 wait states in it. And
- > that, you cannot discuss it :-)
-
- Well if you are making "a 32 bit synchronous access in 4 cycles" on a 16 bit
- bus there can be NO WAIT STATES. To get 32 bits you must make two 16 bit
- accesses. Since each access requires 2 clock cycles, two accesses require
- 4 clock cycles.
-
- I think what you meant is that 4 clock cycles, to access 32bits, on a
- 16 bit bus is the equivalent of 2 wait states on a 32 bit bus.
-
-
- Brian, WS1S (ST/TT user)
- Bell Northern Research
- Research Triangle Park, NC
-