802.11ac - throughput test

Two 802.11ac routers with Broadcom chipsets. Indoor test:
80Mhz channel
Radio speed connection 1170Mb

Result: 661Mbps real TCP transmission!!!

I can’t wait for new chipsets from Qualcomm :slight_smile:

1 stream/session or more?

Test was made with NTttcp software with 20 session.

can u post the screen capture?

Here it is:
Old firmware
80211ac_TCP.JPG
New firmware
test TCP w jedna strone.jpg
Spectrum:
spectrum.jpg

Here is a result od UDP test. Test performed between two RB800. 802.11ac card were connected to two antennas http://www.cyberbajt.com/en/shop_item/3324/pan_519mimo_3x3.html. Distance between antenas about 10m.
udp-test.jpg

WOOW
I managed to get more than 850Mb with UDP test

is it possible to run the test with 20MHz or 40MHz channel?
160Mb/s in 20MHz sounds little better, then 802.11n

Czolo, your link mentions Atheros 3x3 MIMO card, but topic is about 802.11ac. Could you clarify what cards you used? As far as I know, 802.11ac chips from Atheros are not available yet.

I haven’t mentioned about Atheros 3x3. I’m using Broadcom BCM4360 802.11ac cards. RB800 are only for UDP tests through eth ports (as you can see Atheros card is disabled)
Here is card specification:
http://www.broadcom.com/products/Wireless-LAN/802.11-Wireless-LAN-Solutions/BCM4360

Where is the card plugged in?

For now:

Channel 40Mhz (radio connected @405M)
333Mbps UDP

Channel 20Mhz (radio connected @175M)
105Mbps UDP

A little coorection for 20Mhz
After antenna aligment: 150-160Mbps

Into motherboard from Netgear r6300

hi!
i need more throughput on my exist links

  1. link 26km 2*2 (UBNT dishes and rocket m5 AP)
  2. link 2.6km now 22 i upgrade to 33

have questions

you make outdoor test?
how long link?

what router board you use?
or used the original product board? if that how to install os on this board?

My link was for about 1km. Hardware was Netgear R6300. In 2x2 mode results were worse.

what antennas you used for this link?
how much noisier frequency range?
what parameters achieved?
whether it is a software limitation (possibly ACK)?

By your result seems that 40 MHz 2x2 have same result of a 802.11N
Nice if u can post tcp results too

3x3 antennas from Poland. Software was really simple, so I don’t know how musz noisy it was. Only limitation that was there was channel size. But I’m waiting for new QCA98xx. It should work better.

Maybe you dont know this software. All tests above were with TCP :smiley:
So results are far better than 802.11N.