[LINK] ERRATA - NBN retail cost and 12 year technology bell curve

[Tom Koltai]

> -----Original Message-----
> From: link-bounces at mailman.anu.edu.au 
> [mailto:link-bounces at mailman.anu.edu.au] On Behalf Of Tom Koltai
> Sent: Sunday, 17 April 2011 10:45 AM
> To: 'Tom Worthington'; 'Richard Chirgwin'
> Cc: link at mailman.anu.edu.au
> Subject: Re: [LINK] NBN retail cost and 12 year technology bell curve
>
It's been pointed out to me, offlist, that I might be mistaken with the
following statements:
>
> 
> 150 M was selected as being 50% of the maximum ethernet cable length.
> 
> Statistically, I believe that would meet the requirements for 
> Fibre connection. Technically the downside is 70 ms switching 
> induced latency versus 35 ms with fibre to the CPE.
>
>
I was informed that :
>
>CAT5e has a maximum length of 100 metres. Check your facts.
>
Which of course is correct. 
Unfortunately my addled brain cells confused modern Ethernet cat5 with
rolling out coax ICL Oslan networks which of course is totally
different.

(I never actually installed any cat5... By the time that was introduced
I was an executive chair warmer.)

As for the switching, I quoted milliseconds when I should have quoted
nanoseconds. DUH!!!

As an example of switch latency:

Quote/
There are two ways of measuring the switch latency. HP reports latency
of their
S&F switches on a LIFO basis (standard way of latency measurement for
S&F
switches). LIFO latency is the time elapsed between the end of the last
bit of the packet going into the switch to the beginning of the first
bit of that
packet emerging from the switch. Cut-through switch latency is however
measured
on a FIFO basis --- time measured from the beginning of the first bit of
the
packet going in the switch and beginning of the first bit of the
outbound packet
coming out of the switch. To convert LIFO figure into FIFO, one must add
the
time spent on the wire:

  FIFO = LIFO+(packet length(bytes)*0.08 usec)

That doesn't make much difference for very short packets, but it adds 80
usec
for packets 1kB long.
/Quote

Apologies to linkers who were relying on the aforementioned data being
correct.

TomK