dslreports logo
    All Forums Hot Topics Gallery


how-to block ads

Search Topic:
share rss forum feed



Speed test results on NanoBridge PTP

Click for full size
Forgive me for all my posts, but I am just soaking up as much as I can here.

For my PTP link, I currently have a NanoBridgeM5 installed.

I have a 50/5 cable connection currently behind it, but will soon be upgrading to a 100/100 fiber circuit. I am going to order a RocketM5 to replace this hardware, but in the mean time I was curious what these test results mean.

Running the tests from the Station side to the AP, in three different modes, I get:
(Ran 4 tests to get an average)
RX: 22.35 Mbps
TX: 18.77 Mbps
Total: 41.12 Mbps

RX: 19.04 Mbps
TX: 6.73 Mbps
Total: 25.77 Mbps

RX: 21.96 Mbps
TX: 17.90 Mbps
Total: 39.86 Mbps

RX: 19.69 Mbps
TX: 17.84 Mbps
Total: 37.53 Mbps


RX: 36.82 Mbps
TX: 0.00 Mbps
Total: 36.82 Mbps

RX: 42.65 Mbps
TX: 0.00 Mbps
Total: 42.65 Mbps

RX: 43.37 Mbps
TX: 0.00 Mbps
Total: 43.37 Mbps

RX: 44.22 Mbps
TX: 0.00 Mbps
Total: 44.22 Mbps


RX: 0.00 Mbps
TX: 41.93 Mbps
Total: 41.93 Mbps

RX: 0.00 Mbps
TX: 61.21 Mbps
Total: 61.21 Mbps

RX: 0.00 Mbps
TX: 54.04 Mbps
Total: 54.04 Mbps

RX: 0.00 Mbps
TX: 56.00 Mbps
Total: 56.00 Mbps

I know the speedtests on these units are not true numbers, but at least they give me an idea of each direction. So what one matters? duplex? receive?
Is there a reason for the fluctuation in each of them?
Can I make it better with the NanoBridge? Or just replace it with a RocketM5, and then diagnose the speeds?
Are my power levels good?
How does the noise floor look?

The link is about 5.5 miles. It does have line of sight, I am not positive on any trees hitting the Fresnel zone, there, it is mostly clear but there is one tree that could possibly be in the Fresnel perhaps?

If I had to guess, the tree is .25 miles from the AP, and 20-30ft below a direct line of sight line.

Bottom line I am looking to maximize the speed in this link. If that be done with this hardware I would like to do so, if I need something different that is fine.
If I do go with a rocket, what type of speeds could I possibly get on a link at 5.5 miles? Could I get close to the 90mbps/100mbps mark?

IMHO If it's a fresnel problem you'll see the CCQ dive when load goes up. If CCQ stays steady then you're probably clean.

Personally I never trust the built in testers... they drive CPU up as they try to increase throughput on the same device so it's counter productive.

Not much of UBNT shop here anymore but on RB411GL Dual with those kinda numbers I'd expect maybe 110 simplex and ~40x40 'duplex' with UDP. A bit less with TCP.