Slow transfer with 21140 card
Robert G. Brown
Thu Jul 23 11:03:13 1998
On Wed, 22 Jul 1998, John Fulmer wrote:
> Keith Dart wrote:
> > On Wed, 22 Jul 1998, Robert G. Brown wrote:
> > > On Wed, 22 Jul 1998, Hannu Krosing wrote:
> > >
> > > > You are probably right. But are the switches _always_ full-duplex ?
> > No, they don't have to be. Most modern switches can be either full or half
> > duplex, and this is auto-negotiated via a protocol first developed by
> > N-Way. Alas, interoperability has been poor with this feature, and it
> > doesn't always work properly.
> However, I have seen some switches that do auto-negotiation that forces
> full-duplex as the default. Or at least some integrators that do it....
I spent ten minutes or so on:
(Charles Spurgeon's MARVELOUS ethernet site) and didn't find anything
that prohibits hubs from being full duplex, BUT "switches are
described in chapter 15" and unfortunately the website only goes
through 14:-( Guess I'll have to buy the book...
BTW, this website is THE online reference for ethernet for all of you
who really want to learn about MII and N-way and autonegotiation and
collision domains and how to actually CALCULATE whether or not your
network layout will work from first principles...
> + John Fulmer | "UNIX was not designed to stop +
> + Secure Network System | you from doing stupid things, +
> + Lawrence, Kansas | because that would also stop you +
> + | from doing clever things." +
> + firstname.lastname@example.org | +
> + http://www.iegroup.com | -- Larry Wall +
> + "The opinions contained above may not reflect those of SNS" +
Robert G. Brown http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567 Fax: 919-660-2525 email:email@example.com