@Medezark said:
@blakeyrat said:
I was working with Cat5E at 100baseT speeds; if you're trying to run gigabit over Cat6, I'm guessing that you won't get jack shit unless you get your pairs correct, not for more than a few meters. Even Cat5E at a tenth the speed would crap-out after 40-50 meters if you got it wrong.
So, you're saying the order of the wiring in the last 1/2" of wire is going to affetct the signal over 100 meters? Also, why is it that PIN 1 and 2 (transmit + and - ) Are perfectly fine to be together, but pin 3 and 6 (Recieve + and -) have top be seperated by two unused pairs?
I really think the whole "specific coloring"/"patterning" is so as to retain the "reserve for telephony" wires. In the past (1995?) I've run standard CAT5 over 300ft wired with OW~O~BW~B~GW~G~BRW~BR on both ends with no loss of signal or speed (Full Duplex/ 100 MB), as long as the end crimps were nice and solid.
The only time I had issues was with poorly set crimps or if the order of the pairs was off. Even with CAT6 and Gigabit later I found no degradation of signal. Maybe I had better cables, or was lucky.
It doesn't matter what colors are used as long as differential pairs (TX+/TX-, RX+/RX-) are each twisted together. It will most definitely not work over any appreciable length (feets for gbit, dozens of feet for 100mbit) if you "split" the pairs -- if the differential pairs are not running, each, on a twisted pair of wires inside of the cable.
Oh, by "not working" I mean it won't pass any sort of cable certification tests where transmission parameters are measured. If all you have to 'test' your connection is the fact that the browser connects to tdwtf, then I guess you are TRWTF :)