Weird Pipe Found Buried in Yard

That was specific to WideBand Corps implementation of 1000Bt networking. Not everyone's implementation was as robust!!!

You want to try in a 300 ft long building with various itterations of cat5/cat5e cabling running both bundled and helter-skelter, some in troughs, some not - running both over and beside flourescent lamp troughs, down walls behind service panels, under floors, over suspended ceilings, with "noisy" led lighting, you name it - with servers and switches in 2 different areas of the building, with 2 separate networks interlinked (one data, one voip) and other challenges. The voip was POE 10/100 - the data mostly 1000BT.

There is a reason IT techs like myself keep our hair cut short!!!!

They HAD to have "sniffed" the internet when and where I was connecting to the EZWeb interface either when Iset it up (on my wired internet connection), or when I accessed it (once on wired, and once on secured home wifi) Both are behind a NAT firewall.

Less than miniscule? OK, it's less than miniscule. So what?

So that's still miniscule bandwidth.

It *IS* small. Something that CAN be ignored on GBE.

But it was NOT the audio that was causing the issue. Good grief!

Yes, I have.

Boy HOWDY! Some people have it and many don't.

Nonsense.

IEEE 802.3-2015 section 40.5.1 Support for Auto-Negotiation:

"All 1000BASE-T PHYs shall provide support for Auto-Negotiation"

IEEE 802.3-2015 section 55.6.1 Support for Auto-Negotiation:

"All 10GBASE-T PHYs shall provide support for Auto-Negotiation"

So how much current was that sub-panel handling vs how much a residential garage handles?

So that should be down to 100 mb/sec.

There does seem to be some confusion on this point. Initial autonegotiation is supposed to be per Clause 28, which, among other things, specifies (28.1.4.3) "Provision has been made within Auto-Negotiation to limit the resulting link configuration in situations where the cabling may not support the highest common capability of the two end points.".

Once Clause 28 autonegotiation has determined that all the pieces are in place to run gigabit, then Clause 40 takes over and works out the details of the gigabit connection.

By the way, don't know if you're aware of it but 802.3 is a free download from IEEE after you tell them your life story.

I'm not going to try to dig out all the details--802.3 is over 4000 pages and Clause 28 alone is more than 50 pages.

Made me remember there were occasional times I'd get incensed about guys stringing cables near fluorescent lamps, especially when they were clearly instructed to that very morning. And one of the worst offenders were variable frequency drives of A/C equipment, pumps, etc. We did a lot of automation/energy saving jobs.

Or running cables right near a water pump under the flooring in a server center.

The thing is though, that's not the situation we're discussing, we're discussing running a line from somebody's house to somebody's garage. Unless it's the Gates Mansion or Mar A Lago or some such that shouldn't be even coming close to the span limits.

Note, I work in such a building--I should send you a photo of the tunnel sometime. But fortunately I don't have to maintain the network--by some weird quirk of fate I am now a finance guy.

Of course not, that is why I mentioned Total System Load.

Have you ever logged traffic on a network? Including collisions because of it? I have, and that is why I am aware of the variables and take them all into account. I don't just blow things off as inconsequential until after inspection.

More of your outdated knowledge. Collisions haven't been part of Ethernet for a very long time.

About 300 amps per sub panel IIRC

Effective, yes - but not negotiated.

I know it is VERY complex, and not all equipment 100% implements the entire specification. What doesn't appear to be clear is what happens after Clause 40 does it's negotiation and conditiopns change?

I was told the protocol doesn't change - it just resends lost packets and retries untill it succedes - which slows the effective speed, without actually changing the connection protocol.

For the situation being discussed, an old 10B5 co-ax would likely do the job - - - - Regardless, for a cabled connection I WOULD bury it - but I'd be pretty tempted to run a secured wifi if I was not doing anything particularly sensitive (ie - no internet banking).

Early in my carreer I had a contract with a software developer locally with 2 buildings connected by an IR link - back in the co-ax days. What a total pain in the ass that link was!!!! Same place that had such a slow network it was painfull. A whole bunch of engineers and they couldn't figure out if you only needed a 10 ft co-ax drop, a 10 foot drop was infinitely better than a 50 foot coil of co-ax bundled behind the desk.. I must have taken close to 1000 ft of cable out of that building - and miraculously the network kicked into over-drive!!! Didn't help the remote link though - went down every time it rained or snowed, and in the fall when the leaves started to blow around - - - and it was only less than 200 feet - - -

Can be with half duplex. That's 10-100 ( not always - but more often than you would think) or giga over 4 pairs. Not state of the art by any stretch, but still SURPRISINGLY common -

- - -

That only shows your lack of experience or knowledge.

HAND

If you're running half-duplex today there's something wrong.

Wrong, of course. It shows I know something about the design of these things.

But the fact that there was audio involved is irrelevant. IOW, you brought it in as a red herring.

Collisions? In a switched network? Really? Are you sure the last time you did this stuff, it wasn't 10base2?

Moonshine bootlegging? ;~)