Ethernet names...

Warner Losh imp at bsdimp.com
Thu Oct 4 12:43:30 CDT 2018


On Thu, Oct 4, 2018 at 11:39 AM Paul Koning via cctalk <
cctalk at classiccmp.org> wrote:

>
>
> > On Oct 4, 2018, at 1:30 PM, Grant Taylor via cctalk <
> cctalk at classiccmp.org> wrote:
> >
> > On 10/04/2018 11:26 AM, Paul Koning via cctalk wrote:
> >> That's sort of accurate.  A quick look shows some key differences: V2
> adds the "collision presence test" -- verifying the collision detect signal
> is working.  There is also the "jabber timer" -- a watchdog timeout that
> stops excessively long frames.  And V2 introduces the loopback protocol
> (protocol type 90-00).
> >
> > That mostly sounds like the frame formats are the same on the wire and
> that the differences are in the protocols that use said frame.
> >
> >> The collision presence test is somewhat of an interoperability issue:
> if you attach a V1 transceiver to a V2 NIC, the NIC would complain on every
> transmit that it didn't get the collision test signal.
> >
> > That makes me wonder about the "heartbeat" switch that I see on older
> AUI transceivers.
>
> Yes, I think that's the collision test.  So in the OFF position you have a
> V1 compatible transceiver, ON is needed for V2.
>

Is that the same as SQE? I always had to turn that bad-boy off or risk
collision storms :(

Warenr


More information about the cctalk mailing list