Why Airgrid 1-pol almost as fast as MT Dual-pol?

RouterOS Wireless Networking
1

On our 1 network we have RB493AH->Airgrid M5->Airgrid M5->RB433AH
Airgrid is Ubiquity ‘mimo’ (but can only be 1 pol since it’s a grid) integrated solution.
There we get 63Mbps TCP (From 493 to 433 test).
Airgrid:
VB TCP Send.PNG
Today we set up a Mikrotik link, about same distance (±8km) with great LOS, good fresnel, etc. (Massive) 32dBi dual-pol dishes, R52n cards at both ends. 1m LMR400 cable runs.
We get marginally better speed with 70Mbps! Massive antennas, tweaking configs, etc. NStreme enabled, .
Mikrotik:
PDS TCP rec.JPG
Airgrid does not even have nStream, it’s only 1 pol and it’s almost as fast and you just plug and play to get that speed. (Using 40Mhz)

What can I be missing here? Signal strength is -65, SNR 52dB.

2 R name=“ToHDP” mtu=1500 mac-address=00:0C:42:61:2C:EC arp=enabled disable-running-check=no interface-type=Atheros 11N
radio-name=“000C42612CEC” mode=ap-bridge ssid=“TWP2P12” area=“” frequency-mode=manual-txpower
country=no_country_set antenna-gain=0 frequency=5320 band=5ghz-onlyn scan-list=default rate-set=default
supported-rates-b=1Mbps,2Mbps,5.5Mbps,11Mbps
supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps,24Mbps,36Mbps,48Mbps,54Mbps basic-rates-b=1Mbps
basic-rates-a/g=6Mbps max-station-count=2007 ack-timeout=dynamic tx-power-mode=default
periodic-calibration=disabled periodic-calibration-interval=60 dfs-mode=none wds-mode=disabled
wds-default-bridge=none wds-default-cost=100 wds-cost-range=50-150 wds-ignore-ssid=no
update-stats-interval=disabled default-authentication=yes default-forwarding=yes default-ap-tx-limit=0
default-client-tx-limit=0 proprietary-extensions=post-2.9.25 wmm-support=enabled hide-ssid=no
security-profile=default disconnect-timeout=3s on-fail-retry-time=100ms preamble-mode=short compression=no
allow-sharedkey=no station-bridge-clone-mac=00:00:00:00:00:00 ht-ampdu-priorities=0,1,2,3,4,5,6,7
ht-guard-interval=any ht-extension-channel=below-control
ht-supported-mcs=mcs-0,mcs-1,mcs-2,mcs-3,mcs-4,mcs-5,mcs-6,mcs-7,mcs-8,mcs-9,mcs-10,mcs-11,mcs-12,mcs-13,mcs-14,mcs-
15
ht-basic-mcs=mcs-0,mcs-1,mcs-2,mcs-3,mcs-4,mcs-5,mcs-6,mcs-7,mcs-8,mcs-9,mcs-10,mcs-11,mcs-12,mcs-13,mcs-15
ht-txchains=0,1 ht-rxchains=0,1 ht-amsdu-limit=8192 ht-amsdu-threshold=8192 hw-retries=7 frame-lifetime=0
adaptive-noise-immunity=none hw-fragmentation-threshold=disabled hw-protection-mode=none hw-protection-threshold=0
frequency-offset=0



2 interface-type=Atheros 11N
chip-info=“mac:0x80/0x2, phy:0xd0, a5:0xc0, a2:0x0, eeprom:0x0”
pci-info=“00:12.0” capabilities=tx-power-control,ack-timeout-control,
virtual-ap,alignment-mode,scanning,
nstreme,sniffing,power-channel,ht40-a,
ht40-g,wmm,spectral-scan
default-periodic-calibration=enabled
supported-bands=2ghz-b,5ghz,2ghz-g,5ghz-10mhz,5ghz-5mhz,2ghz-10mhz,2ghz-
5mhz,5ghz-11n,2ghz-11n
2ghz-b-channels=2412:0,2417:0,2422:0,2427:0,2432:0,2437:0,2442:0,2447:0,
2452:0,2457:0,2462:0
5ghz-channels=5180:0,5185:0,5190:0,5195:0,5200:0,5205:0,5210:0,5215:0,
5220:0,5225:0,5230:0,5235:0,5240:0,5260:0,5265:0,5270:0,
5275:0,5280:0,5285:0,5290:0,5295:0,5300:0,5305:0,5310:0,
5315:0,5320:0,5745:0,5750:0,5755:0,5760:0,5765:0,5770:0,
5775:0,5780:0,5785:0,5790:0,5795:0,5800:0,5805:0,5810:0,
5815:0,5820:0,5825:0
2ghz-g-channels=2412:0,2417:0,2422:0,2427:0,2432:0,2437:0,2442:0,2447:0,
2452:0,2457:0,2462:0
5ghz-10mhz-power-channels=5175:0,5180:0,5185:0,5190:0,5195:0,5200:0,
5205:0,5210:0,5215:0,5220:0,5225:0,5230:0,
5235:0,5240:0,5245:0,5255:0,5260:0,5265:0,
5270:0,5275:0,5280:0,5285:0,5290:0,5295:0,
5300:0,5305:0,5310:0,5315:0,5320:0,5325:0,
5740:0,5745:0,5750:0,5755:0,5760:0,5765:0,
5770:0,5775:0,5780:0,5785:0,5790:0,5795:0,
5800:0,5805:0,5810:0,5815:0,5820:0,5825:0,
5830:0
5ghz-5mhz-power-channels=5175:0,5180:0,5185:0,5190:0,5195:0,5200:0,5205:0,
5210:0,5215:0,5220:0,5225:0,5230:0,5235:0,5240:0,
5245:0,5255:0,5260:0,5265:0,5270:0,5275:0,5280:0,
5285:0,5290:0,5295:0,5300:0,5305:0,5310:0,5315:0,
5320:0,5325:0,5740:0,5745:0,5750:0,5755:0,5760:0,
5765:0,5770:0,5775:0,5780:0,5785:0,5790:0,5795:0,
5800:0,5805:0,5810:0,5815:0,5820:0,5825:0,5830:0
2ghz-10mhz-power-channels=2412:0,2417:0,2422:0,2427:0,2432:0,2437:0,
2442:0,2447:0,2452:0,2457:0,2462:0
2ghz-5mhz-power-channels=2412:0,2417:0,2422:0,2427:0,2432:0,2437:0,2442:0,
2447:0,2452:0,2457:0,2462:0
5ghz-11n-channels=5180:0,5185:0,5190:0,5195:0,5200:0,5205:0,5210:0,5215:0,
5220:0,5225:0,5230:0,5235:0,5240:0,5260:0,5265:0,5270:0,
5275:0,5280:0,5285:0,5290:0,5295:0,5300:0,5305:0,5310:0,
5315:0,5320:0,5745:0,5750:0,5755:0,5760:0,5765:0,5770:0,
5775:0,5780:0,5785:0,5790:0,5795:0,5800:0,5805:0,5810:0,
5815:0,5820:0,5825:0
2ghz-11n-channels=2412:0,2417:0,2422:0,2427:0,2432:0,2437:0,2442:0,2447:0,
2452:0,2457:0,2462:0
ht-chains=0:chain0,1:chain1




2 name=“ToHDP” enable-nstreme=yes enable-polling=yes disable-csma=no framer-policy=best-fit framer-limit=3998




8 interface=ToHDP radio-name=“000C4265B693” mac-address=00:0C:42:65:B6:93 ap=no wds=no rx-rate=“243.0Mbps-HT”
tx-rate=“117.0Mbps-HT” packets=6933102,2928231 bytes=3748849318,1672769648 frames=3789490,2097049
frame-bytes=3777477724,1672871590 hw-frames=5612686,39238232 hw-frame-bytes=322965767,3466293947
tx-frames-timed-out=0 uptime=4h35m41s last-activity=0s signal-strength=-66dBm@6Mbps signal-to-noise=51dB
strength-at-rates=-66dBm@6Mbps 0s,-65dBm@9Mbps 4h35m39s980ms,-64dBm@12Mbps 4h35m39s830ms,-65dBm@18Mbps 4h35m39s350ms,-
65dBm@24Mbps 4h35m39s40ms,-66dBm@36Mbps 4h35m38s360ms,-67dBm@48Mbps 4h33m38s790ms,-69dBm@54Mbps
4h33m32s390ms,-66dBm@HT20-4 4h23m39s370ms,-65dBm@HT20-5 4h18m38s900ms,-67dBm@HT20-6 4h18m8s390ms,-
72dBm@HT20-7 4h18m8s270ms,-72dBm@HT40-4 10m24s170ms,-70dBm@HT40-5 7m20s80ms,-70dBm@HT40-6 0s,-
75dBm@HT40-7 23m51s560ms
tx-signal-strength=-66dBm tx-ccq=95% rx-ccq=97% p-throughput=90584 nstreme=yes framing-mode=best-fit
framing-limit=3998 routeros-version=“4.16” last-ip=10.1.6.26 802.1x-port-enabled=yes compression=no wmm-enabled=yes

2

Yes , My friend i want to do NS dual in mikrotik transmit the signal and the AirGrideM5 and another products from UBNT MIMO revived the signal but config in mikrotik is NS dual , if possible help me …

3

70mbps can be achieve with just a 5Ghz, 40mhz link.

4

airgrid have 70 mbps half duplex bandwidth, but nstreme dual have about 60-70 mbps full duplex bw.

when airgrid transmit 50-70 mbps, the ping times reach 20-30 ms, but
nstreme dual in same speed transmission have a stable 1-3 ms.

this is the difference.

5

He’s talking about Dual Polarity not Nstream dual. NStream dual uses twice the bandwidth.

Single chain 40mhz channel we’ve seen 8km link’s pushing 100mbit out of the box. The same link MT push’s 50-60mbit. Latency increase is from TDMA and expect to see the same with NV2

UBNT is ahead of the game with speed but MT have more features since they make router software and UBNT makes wireless link software.

Both have their place at various points in a network

6

Don’t get me wrong, I still love MT and wish my whole network could be that as you can’t compare the two, feature wise.
I was just a bit dissapointed seeing the results as setting up those massive dishes, crimping cables, experimenting with different configs, weatherproof housings, etc. etc. is so much more work & lots more money, for such a little difference.

I was hoping that I just have a config wrong somewhere since I’ve seen 90Mbps+(UDP) full duplex done before on MT, see recent post with SXT’s:
http://forum.mikrotik.com/t/802-11n/28753/909

802.11n
7

ekkas, the Ubiquity devices most likely have airmax (tdma) enabled on them. In your setup you are using Nstreme. Please upgrade to RouterOS v5.0rc10 or at least enable wireless-nv2 package in v4.16 and enable Nv2 wireless protocol.
Then you should be able to get good results, like the one from the other topic where user is using SXT and getting more than 100mbps full dubplex.

About your current Nstreme setup, you might want to lower the framer-limit back to 3200 and try out dynamic-size or exact-size framer-policy. But better to use Nv2.

Also check the resource usage on the boxes where you are generating the traffic as you need powerful routers in order to generate a lot of TCP traffic. From where to where you are running the test?

Here is a test where one of our customers demonstrates how Nv2 outperforms Airmax by nearly 50%. This was done in the Animal Farm wireless user meeting, a very popular event for WISPs in the USA. Compare page 9 results with page 10 and page 12 results with page 13:
http://www.txox.com/backhaul_bakeoff.pptx

8

But better to use Nv2.

Thanks Uldis, I’m just weary to load ROS5 (or 4.16+NV2 package) again as we had problems with far clients losing connection with RC9 and haven’t seen a fix for it in the RC10 changelog.
The RB that have that uplink also have a few far clients on it on another card.

From where to where you are running the test?

RB493AH->RB433(with R52n)->RB433(with R52n)->RB433AH
So I’m running from 493AH to 433AH through 2 RB433’s.
Ekkas

9

but you can’t compare TDMA with non-TDMA. It’s apples vs oranges.

10

Even with a PTP link?

11

of course!

12

If upgrading from an Nstreme to Nv2 should I keep Nstreme enabled or just turn it off altogether?

13

On the client I suggest to use wireless-protocol=any and on the AP specify wireless-protocol=nv2. You can leave the rest nstreme settings as it is as they will be ignored if you use wireless-protocol=nv2

14

Please explain it, what you mean. The most interesting part is “more than 100mbps full duplex”.

15

it is between the wireless. If that board would be gigabit you could pass more than 100mbps to the ethernet side as well.