¿How to fix NV2 transmission rate? / MCS / Stability and CCQ with NV2. More questions...

[jmontero]

Hi to everyone,

This is mi first post, but I’ve been reading the forum for a long time. Also I couldn’t find any “Presentation” section, so here I am.

I have been using NV2 for 6 months without any trouble. We have more than 30 Mikrotik routerboards; some of the used as PtMP and other used in PtP configuration. Sadly we got some problems in the last two weeks so I have some questions to the community.

---

1.- There is some way or parametter (different to the bandwidth) to fix the NV2 transmission rate?
My experience with NV2 shows that any parameter I tried to use to fix the transmission rate is useless. This ‘must have’ need comes when I have a unstable connection and some packet lost with a ping. Actually I’m using the bandwidth to decrease the transmission but using a low MCS would be much better; because even with low bandwidth the transmission rate is very variable.

With other protocol (802.11) the MCS is a good way to fix the transmission rate, and get a more stable connection; because at the end you are fixing the signal modulation (PSK, QAM…)

2.- The CCQ percentage is right calculated with NV2?
This seems to be a problem derived from the first question. In NV2 the transmission rate used to be very variable (from 50 Mbps to 270 Mbps) and in this situation the CCQ also changes very often. If I change the protocol to 802.11 the transmission rate is much more stable and the CCQ seems to be well calculated.

3.- NV2 use different MCS levels? What are they? What kind of signal modulation is used by NV2 protocol?
As the transmission rate varies, I want to know if NV2 is based on MCS or other kind of modulation table.

4.- There is some better setup for NV2 parameters in PtP or PtMP?
IMO the parameter tdma-period-size is the most critical in NV2, but, I think that shoud get different values for PtP and PtMP:
Point-to-point configuration: The tdma-period-size parameter could be greater to get better throughput (and worst delay)
Point-to-multi-Point configuration: This parameter shouldn’t be so big, because the delay is increased per each user connected to the master.
Am I right?

5.- What is the best transmission protocol with some folliage obstacle like a tree?
This final question comes because we are having some strange behaviour with a PtP 4KM transmissión. We get good signal levels but low CCQ and transmission rate. We think that there is a problem with some kind of tree but…It shoudn’t be detected with a low signal level?

Thanks to everyone!!

---

Anyway, I love NV2 and Mikrotik in general. Please continue with your improvement of everything in RouterOS and Hardware as you do!!

Greetings!

PD: There is some kind benefit for registring all of these routerboards?

[pukkita]

Hola!

I think you’re mixing “physical link”, (layer 1 = radio) with Layer 2.

I’m not sure what do you mean with “NV2 transmission rate”, if you meant how to fix a given MCS, that’s done indepently from NV2.

I’d bet when your NV2 shows varying CCQs, radio rates are also changing, isn’t it?

Also bear in mind that for NV2 CCQ rates to be useful, traffic must be passing thorugh the link, as it does its calculations based on actual traffic.

Regarding 5, there’s no such “protocol” per se apart from 802.11n or whatever you are using. But, both nstreme and NV2 need near to perfect link (L1) conditions to do their job; if CCQ goes below 90 in regular 802.11a/n using nstreme or NV2 on top will probably do more harm than good, as as soon as you have lost frames and need lots of retransmission TDMA will be screwed.

In such situation the first goal is to stabilize the link in L1 (radio) terms; it could be the channel used, the antenna used, the height…

[jmontero]

@pukkita

Hola!

I think you’re mixing “physical link”, (layer 1 = radio) with Layer 2.

I’m not sure what do you mean with “NV2 transmission rate”, if you meant how to fix a given MCS, that’s done indepently from NV2.

I’d bet when your NV2 shows varying CCQs, radio rates are also changing, isn’t it?

Also bear in mind that for NV2 CCQ rates to be useful, traffic must be passing thorugh the link, as it does its calculations based on actual traffic.

Regarding 5, there’s no such “protocol” per se apart from 802.11n or whatever you are using. But, both nstreme and NV2 need near to perfect link (L1) conditions to do their job; if CCQ goes below 90 in regular 802.11a/n using nstreme or NV2 on top will probably do more harm than good, as as soon as you have lost frames and need lots of retransmission TDMA will be screwed.

In such situation the first goal is to stabilize the link in L1 (radio) terms; it could be the channel used, the antenna used, the height…

Gracias por la rápida respuesta pukkita!!

First of all maeby I’m confusing some terms… What I’m trying to fix (without succesfull) with NV2 is the “tx-rate” and the “rx-rate”. I couldn’t monitor the actual MCS, so the variation of the those rates makes me think that the MCS is always varying… Isn’t it?

On the other hand; the MCS is a Layer 1 parameter so…it should be independent of the Layer 2 protocol, Am I right? Why with NV2 I can’t fix the transmission/reception rate to a low speed? (like 10 Mbps).

Below you can check the parameters of a NV2 transmissión where I couln’t fix the rates. If I set a lower parameter in MCS configurations always change… Where is a list of MCS used with NV2?

name="wlan1" mtu=1500 mac-address=XX:XX:XX:XX:XX:XX arp=enabled disable-running-check=no 
      interface-type=Atheros AR9888 radio-name="XXXX" mode=station ssid="XXXX" area="" 
      frequency-mode=superchannel country=no_country_set antenna-gain=0 frequency=5180 
      band=5ghz-a/n channel-width=20mhz scan-list=default,complete-channels wireless-protocol=nv2 
      rate-set=configured supported-rates-a/g="" basic-rates-a/g="" max-station-count=2007 
      distance=dynamic tx-power-mode=default dfs-mode=none vlan-mode=no-tag vlan-id=1 
      wds-mode=disabled wds-default-bridge=none wds-default-cost=100 wds-cost-range=50-150 
      wds-ignore-ssid=no update-stats-interval=disabled bridge-mode=enabled 
      default-authentication=yes default-forwarding=yes default-ap-tx-limit=0 
      default-client-tx-limit=0 proprietary-extensions=post-2.9.25 wmm-support=disabled 
      hide-ssid=no security-profile=default disconnect-timeout=3s on-fail-retry-time=100ms 
      preamble-mode=both compression=no allow-sharedkey=no 
      ampdu-priorities=0 guard-interval=any 
      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 
      ht-basic-mcs=mcs-0,mcs-1,mcs-2,mcs-3,mcs-4,mcs-5,mcs-6,mcs-7 
      vht-supported-mcs=mcs0-9,mcs0-9,mcs0-9 vht-basic-mcs=mcs0-7 tx-chains=0,1 rx-chains=0,1 
      amsdu-limit=8192 amsdu-threshold=8192 tdma-period-size=2 nv2-queue-count=2 nv2-qos=default 
      nv2-cell-radius=30 nv2-security=enabled  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 rate-selection=advanced 
      multicast-helper=default multicast-buffering=enabled keepalive-frames=enabled

Greetings and thanks again!

[pukkita]

Keep in mind you cannot “fix” the MCS in the stations. Only in the AP.

NV2 has nothing to do with MCS settings.

You can monitor the MCS rates in Wireless > Registration.

suerte!

[jmontero]

@pukkita

Keep in mind you cannot “fix” the MCS in the stations. Only in the AP.

NV2 has nothing to do with MCS settings.

You can monitor the MCS rates in Wireless > Registration.

suerte!

Buenas!

Hi everyone!

Thanks for the info to find the MCS; although I coulnd’t find the specific MCS, the only thing that I found is…

/interface wireless registration-table print stats
rx-rate="130Mbps-20MHz/2S" tx-rate="130Mbps-20MHz/2S/SGI"
tx-rate-set="OFDM:6-54 BW:1x SGI:1x HT:0-15"

There is any other way to check the actual MCS?
Anyway really thanks!!

So the CONCLUSION is:
“With NV2 you can not set a fixed TX/RX Rate”

From now I will use other transmission protocol for “difficult” connections. Greetings!

[pukkita]

That’s the actual negotiated MCS. 130Mbps-20MHz SGI = MCS14

You have MCS 0-15 as available.

There’s a MCS table in the Wikipedia.

Keep in mind AP will select a lower MCS if the highest one set cannot be stablished reliably.

I guess what you want to do is set a lower MCS so that connections are stablished reliably?

P.S. De dónde eres?

[jmontero]

@pukkita

That’s the actual negotiated MCS. 130Mbps-20MHz SGI = MCS14

You have MCS 0-15 as available.

There’s a MCS table in the Wikipedia.

Keep in mind AP will select a lower MCS if the highest one set cannot be stablished reliably.

I guess what you want to do is set a lower MCS so that connections are stablished reliably?

P.S. De dónde eres?

Hi again!

From the table, How do yo know that the MCS is 14. I think that it is MCS 15 because of from Mikrotik Wiki: “ht-guard-interval (any | long) [NV2 only supports 800nS long guard interval!]”

Anyway, YES, what I want to do is to set a Lower MCS / data rate / whatever to have a reliably connection … is it possible with NV2?

The problem of fixing MCS with NV2 is what made me think that the Layer 2 and 3 are connected in this way…

PS: De Córdoba, tu?

[pukkita]

@jmontero

@pukkita

That’s the actual negotiated MCS. 130Mbps-20MHz SGI = MCS14

You have MCS 0-15 as available.

There’s a MCS table in the Wikipedia.

Keep in mind AP will select a lower MCS if the highest one set cannot be stablished reliably.

I guess what you want to do is set a lower MCS so that connections are stablished reliably?

P.S. De dónde eres?

Hi again!

From the table, How do yo know that the MCS is 14. I think that it is MCS 15 because of from Mikrotik Wiki: “ht-guard-interval (any | long) [NV2 only supports 800nS long guard interval!]”

Where have you read NV2 only supports long GI? GI has nothing to do with NV2 TDMA…

@jmontero

Anyway, YES, what I want to do is to set a Lower MCS / data rate / whatever to have a reliably connection … is it possible with NV2?

The problem of fixing MCS with NV2 is what made me think that the Layer 2 and 3 are connected in this way…

PS: De Córdoba, tu?

Of course it’s possible, you can limit that on the AP HT MCS tab.

I don’t undesrtand what you mean with layer 2 and 3…

Sevilla

[Bergante]

As far as I know, (correct me if I am wrong) the decision to choose a MCS relies mainly on the S/N ratio. The problem is, in non optimal situations maybe there is a good S/N ratio but the actual CCQ is low because, for example,
multipath reflections are causing retransmissions.

In that case you may experience a “jumping” MCS issue. The units keep selecting a high MCS, going back to a lower one when retransmissions occur. It can lead to higher delays than simply setting a lower MCS.

I have one of such links, and I found out that it is stable using 16-QAM modulation but not with 64-QAM.

For now I cannot place the link in a better position. But it is far more stable if I limit the maximum MCS to the best working value.

In my case, doing this works like a charm:

ht-basic-mcs=mcs-11 ht-supported-mcs=mcs-11

When trying any higher values it almost imediately fell back to a lower value and the CCQ went down. With this, CCQ remains above 90%, generally close to 100%.

The trick is to find out the maximum stable MCS you can use and try sticking with it.

[chechito]

when using winbox
enable “advanced mode” when you are on wireless interface settings

that enable data rates and ht mcs tabs.

that allow you to select data-rates available, in many situations disabling 64 qam data-rates, or even 16 qam 2/3 too, help a lot to deal with some difficult situations to obtain stability in exchange of sacrificing some throughput

[jmontero]

Hi and thanks to everyone!

@pukkita

Where have you read NV2 only supports long GI? GI has nothing to do with NV2 TDMA…

Here you have this information, search for “ht-guard-interval”
http://wiki.mikrotik.com/wiki/802.11n_Setup_Guide

@Bergante

As far as I know, (correct me if I am wrong) the decision to choose a MCS relies mainly on the S/N ratio. The problem is, in non optimal situations maybe there is a good S/N ratio but the actual CCQ is low because, for example,
multipath reflections are causing retransmissions.

In that case you may experience a “jumping” MCS issue. The units keep selecting a high MCS, going back to a lower one when retransmissions occur. It can lead to higher delays than simply setting a lower MCS.

Thats exactly my problem

@Bergante

In my case, doing this works like a charm:

ht-basic-mcs=mcs-11 ht-supported-mcs=mcs-11

When trying any higher values it almost imediately fell back to a lower value and the CCQ went down. With this, CCQ remains above 90%, generally close to 100%.

Few minutes ago I tried the following configuration:

AP Side:

 0  R name="wlan1" [b]mode=bridge[/b]
      band=5ghz-a/n/ac channel-width=20/40mhz-Ce scan-list=default,complete-channels 
      wireless-protocol=nv2 rate-set=configured 
      supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps,24Mbps,36Mbps,48Mbps,54Mbps 
      basic-rates-a/g=6Mbps max-station-count=2007 distance=dynamic tx-power-mode=default 
      dfs-mode=none vlan-mode=no-tag vlan-id=1 wds-mode=disabled wds-default-bridge=none 
      wds-default-cost=100 wds-cost-range=50-150 wds-ignore-ssid=no update-stats-interval=disabled 
      bridge-mode=enabled default-authentication=yes default-forwarding=yes default-ap-tx-limit=0 
guard-interval=any 
      [b]ht-supported-mcs=mcs-2 ht-basic-mcs=mcs-2[/b] vht-supported-mcs=mcs0-9,mcs0-9,mcs0-9 
      vht-basic-mcs=mcs0-7 tx-chains=0,1 rx-chains=0,1 amsdu-limit=8192 amsdu-threshold=8192 
      tdma-period-size=2 nv2-queue-count=2 nv2-qos=default nv2-cell-radius=30 nv2-security=enabled

Station:

 0  R name="wlan1" mode=station
      band=5ghz-a/n/ac channel-width=20/40mhz-Ce scan-list=default,complete-channels 
      wireless-protocol=nv2 rate-set=configured 
      supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps,24Mbps,36Mbps,48Mbps,54Mbps 
      basic-rates-a/g=6Mbps max-station-count=2007 distance=dynamic tx-power-mode=default 
      dfs-mode=none vlan-mode=no-tag vlan-id=1 wds-mode=disabled wds-default-bridge=none 
      wds-default-cost=100 wds-cost-range=50-150 wds-ignore-ssid=no update-stats-interval=disabled 
      bridge-mode=enabled
     guard-interval=any 
      [b]ht-supported-mcs=mcs-2 ht-basic-mcs=mcs-2[/b] vht-supported-mcs=mcs0-9,mcs0-9,mcs0-9 
      vht-basic-mcs=mcs0-7 tx-chains=0,1 rx-chains=0,1 amsdu-limit=8192 amsdu-threshold=8192 
      tdma-period-size=2 nv2-queue-count=2 nv2-qos=default nv2-cell-radius=30 n 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 rate-selection=advanced multicast-helper=default 
      multicast-buffering=enabled keepalive-frames=enabled

Please notice that I have set the “mcs-2” …and if I run a Bandwitdh-test and monitor at both sides I get:
tx-rate: 216Mbps-40MHz/2S
rx-rate: 130Mbps-20MHz/2S/SGI
and even more…

What I’m doing wrong with configuration?

Greetings and thanks again!

[pukkita]

Ouch sorry

Guess that should be a wiki leftover from a past limitation? I do have NV2 stations connected to an AP with SGI that show up in registration table as SGI…

Forget about the stations, stations will follow whatever is set on the AP.

did you untick unwanted MCS in the HT-MCS tab??

[alisaleh]

Good day, my satuation is 4 km link , I use ap bridge in side and the other side station wds, with nv2 I had ccq less than 90 and sometimes it goes over, with 802.11 I get ccq over 90 mostly and even 100/100 sometimes.

[pukkita]

@alisaleh

Good day, my satuation is 4 km link , I use ap bridge in side and the other side station wds, with nv2 I had ccq less than 90 and sometimes it goes over, with 802.11 I get ccq over 90 mostly and even 100/100 sometimes.

As stated earlier: Both nstreme and NV2 need stable, near to perfect link (L1) conditions to do their job; if CCQ goes below 90 in regular 802.11a/n using nstreme or NV2 on top will probably do more harm than good, as as soon as you have lost frames and need lots of retransmission TDMA will be screwed.

[alisaleh]

@pukkita

@alisaleh

Good day, my satuation is 4 km link , I use ap bridge in side and the other side station wds, with nv2 I had ccq less than 90 and sometimes it goes over, with 802.11 I get ccq over 90 mostly and even 100/100 sometimes.

As stated earlier: Both nstreme and NV2 need stable, near to perfect link (L1) conditions to do their job; if CCQ goes below 90 in regular 802.11a/n using nstreme or NV2 on top will probably do more harm than good, as as soon as you have lost frames and need lots of retransmission TDMA will be screwed.

Yep , I do not know a lot about mikrotik but I am trying my best to always get better performance using mikrotik devises , my link signal is about 50/52 so isn’t it perfect? Or I need to do other things I am using now 802.11 protocol and not nv2 or nstreme even I tried to get it work, in 4 km do I need nv2 protocol ?how to let ccq over 90 if I used again nv2 ?

[Bergante]

@jmontero

Please notice that I have set the “mcs-2” …and if I run a Bandwitdh-test and monitor at both sides I get:
tx-rate: 216Mbps-40MHz/2S
rx-rate: 130Mbps-20MHz/2S/SGI
and even more…

What I’m doing wrong with configuration?

New 802.11ac units? I think that due to a limitation in the chipset you can’t limit the MCS modes used. My configuration works on old OmniTIK units (802.11n).

Y saludos de Bilbao

[jmontero]

@pukkita

did you untick unwanted MCS in the HT-MCS tab??

Yes, I tried also with the web interface.

You have an PM.

¿Is possible that the bandwidth test use all the MCS without any restriction?
Greetings

[Bergante]

¿Son equipos 802.11ac nuevos? En ese caso NO puedes fijar las velocidades. Era posible en los 802.11n. Limitaciones del chipset.

[jmontero]

@Bergante

¿Son equipos 802.11ac nuevos? En ese caso NO puedes fijar las velocidades. Era posible en los 802.11n. Limitaciones del chipset.

Gracias!! Me estaba volviendo loco!

@Bergante

@jmontero

Please notice that I have set the “mcs-2” …and if I run a Bandwitdh-test and monitor at both sides I get:
tx-rate: 216Mbps-40MHz/2S
rx-rate: 130Mbps-20MHz/2S/SGI
and even more…

What I’m doing wrong with configuration?

New 802.11ac units? I think that due to a limitation in the chipset you can’t limit the MCS modes used. My configuration works on old OmniTIK units (802.11n).

Y saludos de Bilbao

Thanks!!! I was getting crazy!!!

NOTE FOR EVERYONE: You can not set TX/RX Rates (MCS) on new 802.11ac new Router Boards!!

Please Fix that! (if it is possible!).

Greetings!

Thanks to everyone that post in this topic!

[honzam]

I think someone from MikroTik said that he could not fix. It’s a given from AC hardware …