situation.
AlwaysWrong is wrong again (surprise, surprise).
BTW, how were "they" (whomever "they" is) baited, DimBulb? Seems you're the dead fish around here.
Wrong again, on many levels.
No thanks. You're not my type, DimBulb. Be careful or you'll make Roy jealous.
situation.
No. They are seldom happening at the same time, and their internal timeouts between tests and switch-outs differ from maker to maker.
This means that one WILL eventually achieve a handshake, and then both will discontinue making switch-outs.
situation.
And if they're two identical NICs?
Sylvia.
That's exactly the situation we do have. Pin 2 is transmit on the DTE and receive on the DCE.
Sylvia.
My reference states From DTE: pin 2 transmit, pin 4, request to send, pin 20 data terminal ready (ie output); To DTE: pin 3 received data, pin
5 clear to send, pin 6 data set ready. I left out DCD and secondary / reverse channel info. For a non-cross, one would need to have pin 2 as transmit on one end and receive on the other, etc. Not so nice.
situation.
The time in which they perform the test differ as there is no communication between them in said regard. So one will always win, and then a success in the handshake will cause the other to stop trying.
situation.
The Ethernet protocol assumes that packets will collide and become garbled. Each NIC will back off for a random amount of time.
EXACTLY! Which is why, when one needs to hook up a computer that expects to see a modem to another computer, a reversal is needed on those two lines so that the two devices *think* they are communicating with a modem device. Hence "null modem" cable.
Exactly. Far cheaper as well.
Computer talks to modems. When computer needs to talk to another computer via the same ties directly, it is obvious that a conflict will be present. The 'null modem cable' was part of the spec, even before one was ever used.
situation.
situation.
Yep. All you have to do is establish communication first, so that you can decide whose turn it is.
Sylvia.
situation.
Even when "communication"fails, as in no handshake acknowledged, the exchange will eventually match, with 'eventually' being a fairly quick process. In dumb systems handshake simply fails.
In a smart system, it senses the presence of the comm cable, but the failed handshake sets off the pair swap retry.
situation.
That is how Token Ring works but not Ethernet. Ethernet uses a "cocktail party" protocol where you wait for a pause in the conversation before speaking and, if after you start, someone else is speaking you back off until there is another pause. It's usually efficient but if the group is too large and a few speakers are too verbose there will be no communication. Token Ring works by passing the microphone around the group so that everyone who has something to say gets a chance to say it. It can't catastrophically fail but, for normal conversation, it can be extremely slow.
on.
The world would be better with more hermaphroditic connectors.
Ho hum!
Probably. IIRC she refused to post a photograph.
That would explain "her" difficulty in understanding gozinta/gozouta ;-)
...Jim Thompson
Have something to add? Share your thoughts — no account required.
Ask the community — no account required