Originally posted by: TwiceOver
Originally posted by: RadiclDreamer
Originally posted by: TwiceOver
It isn't as important as people think. In the example given WAY above. I use green first rather than orange. This would be the "B" standard? I don't really know, last time I checked electrons were color blind even though I have been in situations that seem they aren't.
Also, you can technically and actually have two runs on one cable. I do this with the link to my garage. 4 pins go into the house from modem to router, the other 4 pins come back from the house to a switch.
Whatever works. I'm sure some guy some where came up with the idea that if you take 1 3 5 6 you will have that much less cross talk. But then again he completely omitted 7 8. If you wanted to eliminate cross talk all together then you would probably pick one from each pair knowing you only need 4 wires.
What spidey was referring to is that this is EXTREMELY poor practice. There is a greater incidence of cross talk and interferance as a result of configurations like this. The pattern is there for a reason.
I understand 100% that it will "work" but a skateboard will technically get your 50 miles to school, but isnt the most intelligent thing to do.
This is usually fine for a small home network where noone but your wife cares if speeds and reliability arent there, but for the real world where runs can go 100 meters this is bad practice. It just causes weird shit.
Well yeah, if you are doing a 300m run (about the max of cat5) you should probably adhere to some standards. Now if you are going from one cube to the next, nobody (even electrons) will care. Also more than likely you are running to a wall plate or a structured cable run of some sort. That probably won't matter either. Now if the run from plate to the "Way Back" is f'd up then yeah... You'll run into "Weird Shit"
100 meters max for any Category rated copper (90 Meters Solid core "in the wall" with 5 meters max of stranded "jumper cordage" on either or both ends.
Solid versus stranded is important, because stranded has a much higher loss, and solid doesn't flex much before it breaks.
Connectors are important too, there are four kinds, two for flat (untwisted, like for phones) and two for round cable ... one is for stranded, the other is for solid conductor. IN the round cable connectors, there are rated and unrated. If you use an unrated connector, you have an unrated cable. The span takes on the characteristics of the lowest rated component.
It's actually MORE important to follow the spec for shorter runs, as Near-End Crosstalk (NeXT) is likely to occur and degrade the signal if not properly terminated.
Regarding "Electrons don't know color" ... no, but manufacturers of cheap cabling do. They expect the standards to be followed and engineer the cable to fit the least expensive street price. Before Cat5e (which qualifies all four pair) only the orange and green pairs were set for "high speed data" ... running it on orange or brown pair, even in the right positions, was a screwed cable, because the pair weren't conditioned for it.
5e and above qualify all four pair, but there are still some conditioning features that make the proper positioning very important.
Regarding using multiple applications over a single four-pair cable (phone & data, or data & data) : It degrades the performance significantly, and is against the spec. If the other application happens to be a phoneline, the 90-120VAC ring voltage can blow the receiver of the NIC on the other pairs, and the analog signal of a call-in-progress will screw the data on the other pair (bad crosstalk). For DSL, it'll drive your data rate into the weeds.
Very Bad Idea.
Regarding :
"Don't be an idiot and don't tell me to "chill". There was nothing argumentative or angry in my post, so there is nothing to chill over. And when have I posted in matters such as this that would elicit such a response? So why don't you chill?
"Br, WBr, G, WG, O, WO, Bl, WBl (for example)" - this was an EXAMPLE. Examples don't have to be precise to get a point across. That point is, it doesn't matter if you follow the "correct" color pattern, so long as the pins are correct. In this case, wire is wire, it doesn't care. I know those aren't in the right pin order, and in this case I don't care. "
Please, just stop posting. you're talking out of your ass. There are exactly two examples: EIA/TIA 568a and EIA/TIA 568b. That's the point, it **********=> DOES <=********* matter. so ... go away, stop ... you're totally wrong.
I'm not being argumentative or angry, I'm trying to stop you from infecting the other n00bs with your totally wrong knowledge.
I didn't say you were a bad person, I asked you to stop talking about something you know nothing about.
Cabling is the foundation of the network. If the cabling isn't right, nothing else matters. You can put the hottest Banana 5000 Uber Machines on your network, if the cabling is not right, it won't perform, and you've wasted (or encouraged other people to waste) cash and set yourself (or others) up for grief & frustration when stuff "kinda" works.
There's more to it than plastic-coated wires, there's actually some serious engineering involved. It's much more than continuity and polarity too. That's why some people get paid the big bucks, they know the difference.