On Sun, 03 Jun 2018 17:10:28 -0400, Clare Snyder
wrote:

On Sun, 03 Jun 2018 03:36:52 -0400, J. Clarke
wrote:

SNIPP
Then you don't get what noisy lines are all about, and what can create
them. Interference. Also just pulling too hard on Ethernet cabling can
stretch the wire messing things up.

So tells us of a documented case in which "pulling too hard on the
cable" or "interference" reduced the transfer rate for 1000BaseTX to 2
mb/sec. You're saying "can happen". So show us when it _did_ happen
or you're just spreading FUD.

If you pull too hard on a gigabit cable and damage the cable it will
NOT autonegotiate down.

Maybe it will, maybe it won't. Depends on the damage. I just
replaced a network cable that was consistently resulting in
autonegotiation from 1 gig to 100 meg and then sometimes failing the
100 meg. It had a lot more problems than "being pulled too hard"
though.

I have seen it slow the network down SIGNIFICANTLY due to dropped
packets and retries - to well below a 2mb equivalent. Any one
conductor suffering damage in Gb ethernet WILL slow the network down.
Or kill it DEAD (at least the one segment)

Ok, so you have numbers to present that show this? Please show the
actual numbers.

A kinked Cat5E cable will fail the quality test for GB ethernet.
Running the cables parallel to a high current AC conductor will do the
same.

I have tried experiments in this regard and Fluke does not seem to be
able to detect the high current conductor.

It won't slow down the bit-rate - but it will definitely cause
deterioration in the service via lost packets and retries, which
translates to a slower EFFECTIVE bit-rate.
SNIPPPPP

How much slower, based on actual evidence?

When talking 10BT /100BT, the troublesome auto-negotiation protocols
COULD downgrade a 100BT to 10BT, and often provided better throughput
on a reliable 10BT connection than on a flakey 100BT - but 10BT is "so
nineties"

So you say that the autonegotiation protocols for 100baseT can reduce
to 10 but somehow the autonegotiation protocols for gigabit can't
reduce to 100. Why would that be?