home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!darwin.sura.net!jvnc.net!nuscc!eletanjm
- From: eletanjm@nuscc.nus.sg (TAN JIN MENG)
- Subject: BUSES
- Message-ID: <1992Jul23.092211.18462@nuscc.nus.sg>
- Organization: National University of Singapore
- Date: Thu, 23 Jul 1992 09:22:11 GMT
- Lines: 29
-
- I'm a tyro in this group. So please be gentle ;-).
-
- I'm just wondering...
-
- With all this talk about current and future buses for desktop machines
- (EISA, TurboChannel etc), a lot of concern seems to be on the speed of
- the bus (as opposed to for example interrupt/device management or the
- ability to use different speed devices efficiently).
-
- It seems to me that many if not most applications do not need (relatively
- speaking) such a high bandwidth. Even with an application such as FDDI,
- the bandwidth is shared by many machines anyway and also software bottleneck
- becomes more significant at those speeds.
-
- Software controls the rate at which you exercise the bus at these high speeds.
- Also, if I get my file within 0.5 sec as opposed to 1.0 sec, will I
- notice enough to pay for the hardware?
-
- Aren't there other aspects of system design (including software) that we
- should look at to get optimum cost/benefit?
-
- Faster is always sexier I know.
-
- Anyone care to have a go?
-
- jin meng
-
- PS. I admit that there will be always a need for high speed buses for
- special purposes. I am talking more about general desktop machines.
-