Someone said in a recent thread that they've tried four pair cabling and it might not have been CAT5, but it worked fine for networking.

I've got this flip, see...

So we finally discovered the phone wiring rats' nest. All wires are four pair. The latest thing, maybe? Well, no. This wire was run inside the walls of a house that has a run of ten foot satellite dish signal and control wire, and down the hill we can see the 4" or so pipe that the big dish used to be mounted to. That tells me we should think of this house as 1980s.

So, a test. Connect the router to one of the 4 pair cables, and connect that to another one. Run speedtest.net at the router and then at the end of that monstrosity, and see what the difference is.

Our network alternative is DECA over the DirecTV RG distribution. And of course we'll be putting in a few WAPs where they will help.

Is this too crude of a tool to give me any useful information?

How long is the run? I used some 4 pair untwisted pair wire for some POE Elan keypads. It worked on the 4 runs that were less than about 75'. The 3 that were longer did not work. We had to retro CAT5e to those. That probably does not directly answer your question.

Deca is a decent alternative. Mobotix 2Wire setup would work as well....but I am not sure the bandwidth capability off the top of my head compared to deca.

Thanks for the info.
The run from the central point, to which I can easily move the router, is about a fifty foot walk, so maybe seventy feet of wire. I was going to punish the cable by adding another run in series with it. And duh, I just realized I can check its length with my TDR.

Bubby has helped me out A LOT with the DECA approach. This test will help me make up my mind, but I might end up doing both! The RG6 goes exactly where I need it, but for instance these phone lines are on walls opposite where I need internet.

A better test than comparing two Speedtest results (because it varies so much and involves equipment outside of your control) would be to grab 2 laptops and run iperf.

https://iperf.fr/

It's a little geeky to get set up, but basically you set up a computer on one end and start iperf as a Server. Then set up the 2nd computer on the other end and run it as a client. It gives you the actual bandwidth that you are getting between the two computers.

I've pushed IP cams a pretty long distance over 4 PR untwisted for the same reason. Moca works better though if that's an option.

If you use an internet speed test, it will only go to the speed of what the subscriber pays for, plus or minus a bit. If you connect a router at the head end and a computer to other ends, right click on the network icon and select 'Status'. It will open a box that shows the connection speed to the router and it shows how many bytes have been sent and received. I would also ping the router's IP address and see if the delays are excessive, how long each ping takes and if any packets were lost.

WE HAVE A WINNER!


I'm going to be on site tomorrow and Monday, and I will definitely do this! Thank you so much for the method. This is beautiful. Thank you!

Edit: uh-oh. Details.
Right click on the network icon where? In Network and sharing center? I don't see one. In Windows Explorer? There's no "Status" there. Somewhere in the router?

This procedure should work with the desktop I'm working on right now but I can't find a network icon that yields "status" as a choice after a right click.

Thanks.

Start -> Run -> enter ncpa.cpl

Or:

Control Panel -> Network & Sharing Center (or if you're in category mode, "Network and Internet: View Network status and tasks") -> Change Adapter Settings (left side)

Then:

Right Click over icon -> Select Status

This will at least give you the basic connection speed. A managed switch with statistics would help keep track of errors or flapping on the port.

No need to open a browser for this.

On a Windows computer, you should see a WiFi signal strength or wired network icon at the lower right corner. You can check the speed with a wired or WiFi network. If you want, you can open Network & Sharing Center and click on Change Adapter Settings, where you'll see the Local Area Connection and the computer's own wireless adapter(s). Right click on the one that you want to check and you'll see 'Status'- left click and you'll see the small window I referred to. If that network is active, 'Status' will be in bold type and if it's inactive, it will be faded and you won't be able to use it.

You should be able to see it in the router, but that's obviously not an option if you don't have the password.

I don't carry two computers, so I use this method. I don't remember if I stumbled upon it or if I learned it in the networking class I had, but it's about the easiest way I have seen that doesn't require added software, apps or jumping through hoops of real fire.

This can be useful, but the indicated speed is the theoretical maximum based on raw parameters extracted from the wireless card. All that is required is a completed connection between the computer and an access point. Actual transmission of user data may not be possible even though the indicated connection seems strong.

---

I've found that my laptop's wireless capability is more potent than a typical phone or pad -- particularly the older phone/pads. If I walk the area with my laptop, it may be doing very well, but the customer's phone/pad struggles. I like to use my own phone to map the area. This will more realistically allow me to estimate the effectiveness of coverage for the customer.

At some point I involve a customer's unit as a reality check.

I have also observed multi-path issues where moving the client device a fraction of an inch will be the difference between poor performance and success. In one case I was contemplating blue tape registration marks on the kitchen counter for the customer's laptop. Rotating the laptop a few degrees resulted in large performance variations.

Always a risk is interference from a neighbor because this interference can vary by time of day or week and location within the customer's site.

buzz,
how would you edit what you wrote if the idea was to get the actual performance of mystery wire in the walls, with no wifi involved at all? I ask because that's the question here.

"Evaluating non-CAT5" is still what I need to do. If there's interference from the neighbors, man, will that be a problem!

You'd need a cable certification tool, such as this: [Link: flukenetworks.com]

buzz, are you saying that nothing short of that tool would give me any useful information by which I might choose to use that cabling or not to use it?

I see different speeds, depending on the network my computer is connected to. Here, it shows 72Mb and at places with 802.11g, it shows 50Mb. Some slower 802.11n networks (due to range or building material obstructions, I see 65Mb. That doesn't matter WRT this thread, though (except, as a parallel use of the Status button) because Ernie is trying to validate the unknown wire.

A tool such as this will quickly give a clear diagnosis of the situation.

In a way this is similar to throwing an amplifier on to a test bench. Sure, I could hook up a pair of speakers to see if the amplifier makes noise, but this is a very imprecise test. On the test bench I can very quickly check everything.

It all depends on how quickly you need to work and the confidence you need to have in the robustness of the network.