🚀 go-pugleaf

In article <y$4ObJAcKIG3EwQE@ch-w.demon.co.uk>, John Chapple <jc@ch-
w.demon.co.uk> writes

>Does it *really* take an engineer something like 15 minutes to terminate
>an optical cable, 

Properly, yes.  There do exist 'quick terminations' for fibre optic, but
they don't give as good a result and are not as reliable.

>(I understand it's quite a slow process, sticking the
>end in some sort of resin then smoothing it off with many? different
>grades of some sort of abrasive (apparently theres no quick solution))?

AIUI, the important thing is to cleanly cut the fibre, so that the end
is cut perfectly flat and square, so as to maximise light transfer into
the receiver and reduce reflections back into the fibre.  A microscope
is necessary to examine the fibre end after it's been cut.

>Not really related to subject, but, I understand that on an ethernet
>network if utilization goes above 25% this is regarded as 'a bad thing'

yes (usually quoted at 30% but the difference doesn't matter)

>(whatever that means) my 3com hub has peaked at 80% whilst installing a
>sp over the network on 2machines, just using netbeui BTW. Would this be
>a matter of concern on a real network?

If it were at a *permanent* utilisation level of 80%, that would
indicate a severe overload or a fault (you would see the collision light
on pretty solid) and some form of segmentation/move to switching would
be a good idea.

If, however, it's a temporary situation on a small network (such as for
you installing a couple of SPs) it's no cause for concern.

Ethernet's design was based around the concept of transferring lots of
small packets very quickly.  I don't think it's optimised for large file
transfers (though am willing to be corrected on this point) and the
multi-megabyte bloatware files being chucked around the LAN by Billy$hit
don't help.

-- 
Mike Tomlinson
See the sci.electronics.repair FAQ mirror at  
http://www.jasper.demon.co.uk/repairfaq/