MikroTik

Ethernet test results published on brochure are intended one way ??

What should I expect for a full duplex comunication ?

Can I use bonding of multiple QRT5ac for a point to point link to increase bandwidth ? (are radio channels interfering for close units ?)

I was asked for a minimum guaranteed 1Gbps/1Gbps link

Thank you

Have you checked the theoretical airspeed of this 2 chain mimo ac? 866Mbits/s. So it means you would need four qrt5ac on each side to be able theoretically guarantee the requested speed.

Ethernet test means testing of the ethernet port. The test does not include Wireless communication which is usually slower than Ethernet.
The Ethernet test shows performance of the unit itself, of it's CPU capability.

Wireless test will depend on the environment, so we don't have such tests.

Not even with 4xQRT you wont be able to get 1GBPS full duplex. Maybe 600-700mbps.

Four at each side. Eight in total...

omfg

Yep. That would be something... But surely not recommendable.

If you need 1Gb just use a Mimosa B11 on 2x 80mhz channels, or for shorter links buy a unlicensed UBNT AF24-HD, or for SHORT SHORT links under 1.5 miles, use the Siklu 70/80Ghz backhaul and get up-to 5Gb FDD

Thank you for replies
I'e played with bandwidth test in this setup :

Crs125-24G-1S---------------QRT5ac<<<<<20mt>>>>>QRT5ac------------RB3011UiAS

192.168.1.254--------------192.168.1.10----------------192.168.1.20---------192.168.1.30

Test ran from Crs125-24G-1S

QRT5ac bot with reduced power to achieve about -30db rx/tx signal , wlan1 and ethernet bridged

First of all there is a huge difference between TCP and UDP bandwidth

With default udp packet and tcp connections count I have these results :

192.168.1.254 --- 192.168.1.30

TCP
RX=116Mbps
TX=65Mbps
Both=35/51 Mbps (Tx/Rx)

UDP
RX=233Mbps
TX=503Mbps
Both=175/240 Mbps (Tx/Rx)

....strange that 503Mbps on TX UDP......also strange that low TCP TX rate


Same test started from QRT1 to QRT2 :

TCP
RX=220Mbps
TX=218Mbps
Both=82/82 Mbps (Tx/Rx)

UDP
RX=504Mbps
TX=530Mbps
Both=250/295 Mbps (Tx/Rx)

Curious the last test , from 192.168.1.10 to 192.168.1.30 (QRT1 to RB3011) , UDP test are better despite a device more :

TCP
RX=199Mbps
TX=250Mbps
Both=75/75 Mbps (Tx/Rx)

UDP
RX=510Mbps
TX=585Mbps
Both=265/300 Mbps (Tx/Rx)

Radio configured 20/40/80MHz Ceee, protocol NV2
Radio rates reach 866.6Mbps both directions with Tx/Rx CCQ about 99/100%
Perhaps there are best settings (on ethernet side also) to achieve best results, but the TCP rates are not so promising....

Any further suggestion about test ?

Ethernet test means testing of the ethernet port. The test does not include Wireless communication which is usually slower than Ethernet.
The Ethernet test shows performance of the unit itself, of it's CPU capability.

Wireless test will depend on the environment, so we don't have such tests.

Ahhhhhh it gives an illusion of radio data rate ......
Is QRT5ac the fastest radio device at this time ?

Doing tests from/ to CRS125 may not be the best way, CPU is rather weak on these devices, try with PC instead CRS125. Not sure about RB3011

Those tests are useless.
Strange results are because of that CRS.

Thank you for tips
So, is it more affordable to run bandwidth client/server on two powerful PCs in place of CRS and 3011 ??

Absolutely.