0
Is there any reason they couldn't have been made to use the same pins as NICs and routers for transmitting and receiving? I had trouble finding any information about this through Google, so I hope you can help.
user304805
Posted 2014-03-03T23:18:43.380
Reputation: 3
1
If hubs used the same pins as NICs and routers, then when you connected a NIC or router to a hub or switch, you'd be connecting transmit to transmit and receive to receive, and that wouldn't work very well. The idea was to allow you to use straight-through cables to connect end devices to network infrastructure and have that connect send on one end to receive on the other. It's irrelevant today because pretty much every device supports auto-MDI/X.
David Schwartz
Posted 2014-03-03T23:18:43.380
Reputation: 58 310
Though seldom seen these days, coaxial ethernets didn't discriminate between switches (hubs) and hosts. However, they had to rely on the Carrier Sense Multiple Access w/ Collision Detection feature, which can't support full duplex. This meant that only one host could transmit at any time, that the theoretical max of 10 Mbit/sec was for all hosts on the segment. If two hosts transmitted at the same time, the additional voltage drop across the coax would indicate to all other NICs that there was a collision, and to discard the current frame; it would be corrupt. It didn't work very well. – Nevin Williams – 2014-03-11T03:24:43.057
Read up on DTE/DCE. http://en.wikipedia.org/wiki/Data_terminal_equipment http://en.wikipedia.org/wiki/Data_circuit-terminating_equipment
– Zoredache – 2014-03-04T01:59:09.310