[lug] Cat 5e cable color codes

[D. Stimits]

John Karns wrote:
> On Sat, 10 Aug 2002, D. Stimits said:
> 
> 
>>The particular model number is not shown on that page, but the color
>>scheme matches it exactly. Suggestions on importance of following some
>>scheme (other than just being consistent) would be useful.
> 
> 
> I don't think that it's enough to merely be consistent.  There are two
> different (IEEE?) standards which are used, and "A" and a "B".  AFAIK
> these utilize the pairings in the cable to maximize signal to noise ratio,
> and should be observed.  Probably more important for 100 mbit than 10 mbit
> systems.
> 
> I really don't know what the differences are between the two schemes in
> terms of when one should choose one over the other, and would also be
> interested to read anything about it.
> 
> And a question of my own: does one have to consider hub / switch / NIC
> internal wiring specs in choosing between the two schemes?

With all the replies, and based on web site info, it seems that the 100 
MBit NIC systems are better off with the "A", though I'm not sure if it 
really is entirely a technical aspect (consensus seems to be that in 
ordinary 100 MBit ethernet, the "A" layout is more common). One of the 
web sites indicates that quality can be effected adversely by as little 
as an extra half-inch of untwisting of what is otherwise a twisted 
segment of wire. The particular layout may be some scheme to get the 
best capacitance/inductance qualities, or it might not matter a whole 
lot. The 5e cable is supposed to be able to run up to and including 
gigabit, and I imagine at 100 Mbit there probably is not much difference 
between "A" and "B" (on technical grounds) until you get to maximum 
cable lengths. On standards grounds, it just seems that the "A" is more 
acceptable for being more common on new equipment. Take all of that with 
the proverbial grain of salt, I chose "A" and did not have any 
comparisons (and it works quite well, no packet losses or errors over 
about 50 feet per spoke out of the switch).

D. Stimits, stimits AT idcomm.com