home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!pipex!warwick!uknet!gdt!uwe-bristol!slave.uwe.ac.uk!root
- From: root@slave.uwe.ac.uk (Operator (Phil/Dylan))
- Newsgroups: comp.os.linux
- Subject: Re: real time
- Message-ID: <1993Jan9.135712.29550@csd.uwe.ac.uk>
- Date: 9 Jan 93 13:57:12 GMT
- References: <1993Jan3.054159.11008@umr.edu> <1993Jan4.191229.22825@news.stolaf.edu> <C0Dt5v.6G@gamble.uucp> <1993Jan5.173235.6490@news.stolaf.edu>
- Sender: usenet@csd.uwe.ac.uk (uwe nntp usenet poster)
- Reply-To: p_copela@csd.uwe.ac.uk
- Organization: The New University in the West of England, Bristol (WEB)
- Lines: 17
- Nntp-Posting-Host: usenet@csd.uwe.ac.uk (uwe nntp usenet poster)
-
-
- Of no real value to the debate, there is yet another solution but it's
- a bit of a cheat. As I work in the transputer labs, I have axcess to
- several b004/b008 transputer cards for the PC which I can quite happily
- set running RT applications (sig processing mainly) running on the
- transputers,.. data can be cashed or transfered back and forth from the
- i[3/4]86 to the t800's such that by writing the code with some
- 'intelligence' it's possible to maintain a RT system outside of the
- kernel by getting a foreign processor to do the work for you,... mind
- you the cost is a bit on the heavy side + the transputer development
- tools available are errrm 'primitive' but that doesn't stop cross
- compilations from a sun being mish mashed with a common FS (NFS springs
- to mind) but all of this is academic, the main problem is in the
- specification for your RTS.
-
- Phil
- =--=
-