Community discussions

MikroTik App
 
jookraw
Member Candidate
Member Candidate
Topic Author
Posts: 141
Joined: Mon Aug 19, 2019 3:06 pm

hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 1:08 pm

Hello,

I just got my hAP ax², and finally Mikrotik has a fast WiFi!
Still no 160MHz, but it is a good start! specially on a device with modest specs like the hAP ax², in some cases I've noted that 80MHz can be faster in a crowded.
Screenshot 2022-10-19 115257.png
Screenshot 2022-10-19 115252.png
You do not have the required permissions to view the files attached to this post.
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 1:46 pm

Hi,

where did you buy him? :)

Could you do more test? Some comparism vs hAP AC2 or AC3 procesor usage with cake, etc.?
 
jookraw
Member Candidate
Member Candidate
Topic Author
Posts: 141
Joined: Mon Aug 19, 2019 3:06 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 1:59 pm

moderator nite: no need to quote whole preceding post. Just use "Post Reply" button.
I saw it on stock yesterday in a local distributor, did not thinked twice... they are all gone...

I dont have it setup as a router at the moment, just as an AP.
Edit:
I've done a quick and dirty setup with cake (I don't use it everyday) set to 600M, did end up with just 230Mbps with one of the 4 cores at 100%
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 2:44 pm

from which country are you?
 
jookraw
Member Candidate
Member Candidate
Topic Author
Posts: 141
Joined: Mon Aug 19, 2019 3:06 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 3:25 pm

from which country are you?
Poland
 
User avatar
chechito
Forum Guru
Forum Guru
Posts: 2989
Joined: Sun Aug 24, 2014 3:14 am
Location: Bogota Colombia
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 9:05 pm

what Wi-Fi client device are you using for that tests??
 
jookraw
Member Candidate
Member Candidate
Topic Author
Posts: 141
Joined: Mon Aug 19, 2019 3:06 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 9:08 pm

what Wi-Fi client device are you using for that tests??
I've made the tests with a Windows PC with Intel AX200 WiFi card, I also have tested with my phone, Galaxy S21
 
anuser
Long time Member
Long time Member
Posts: 601
Joined: Sat Nov 29, 2014 7:27 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Oct 19, 2022 9:54 pm

Does current wifiwave2 driver support ATF (airtime fairness)?
 
gotsprings
Forum Guru
Forum Guru
Posts: 2087
Joined: Mon May 14, 2012 9:30 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Thu Oct 20, 2022 3:34 pm

Is WiFi1 the 2.4 radio?
 
User avatar
chechito
Forum Guru
Forum Guru
Posts: 2989
Joined: Sun Aug 24, 2014 3:14 am
Location: Bogota Colombia
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Thu Oct 20, 2022 3:38 pm

Is WiFi1 the 2.4 radio?
in hap ax2 wifi1 is 5ghz and wifi2 is 2ghz

in hap ax3 wifi1 is 2ghz and wifi2 is 5ghz


dont ask me why :lol:
 
gotsprings
Forum Guru
Forum Guru
Posts: 2087
Joined: Mon May 14, 2012 9:30 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Thu Oct 20, 2022 4:19 pm

moderator nite: no need to quote whole preceding post. Just use "Post Reply" button.
I was gonna be real surprised at a speedtest like that in 2.4Ghz.

Also i think Mikrotik did that once before on some other board.

Edit: Didn't the wireless on the RB4011 have the 5 and 2.4 reversed for wan1 and wan2 compared to everything else?
Last edited by gotsprings on Sun Oct 23, 2022 3:42 pm, edited 1 time in total.
 
Stril
Member Candidate
Member Candidate
Posts: 200
Joined: Fri Nov 12, 2010 7:18 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Thu Oct 20, 2022 10:01 pm

I can confirm: single stream performance and above UBNT Wifi 6 LR

700-800 MBit/s are possible
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Oct 21, 2022 7:23 am

moderator nite: no need to quote whole preceding post. Just use "Post Reply" button.
Could you do more test vs UBNT Wifi 6 LR? speed, WiFi range? WiFi speed through the wall?
 
gotsprings
Forum Guru
Forum Guru
Posts: 2087
Joined: Mon May 14, 2012 9:30 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Oct 21, 2022 12:22 pm

moderator nite: no need to quote whole preceding post. Just use "Post Reply" button.
Why do you care about through a wall?

Are walls suddenly standards?
Everything from materials to thickness blows that away.
 
whatever
Member
Member
Posts: 348
Joined: Thu Jun 21, 2018 9:29 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Oct 22, 2022 11:16 am

How is hap ax2 performance with wifi5 (ac wave2) clients?
 
User avatar
chechito
Forum Guru
Forum Guru
Posts: 2989
Joined: Sun Aug 24, 2014 3:14 am
Location: Bogota Colombia
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Oct 23, 2022 5:11 am

I can confirm: single stream performance and above UBNT Wifi 6 LR

700-800 MBit/s are possible
Highest 802.11ax datarate per spatial stream at 80mhz is 600mbps
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Thu Nov 03, 2022 1:47 pm

I get more than 600 mbps on hap ax2...
Last edited by BartoszP on Thu Nov 03, 2022 2:24 pm, edited 1 time in total.
Reason: removed excessive quotting of preceding post
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Thu Nov 03, 2022 2:46 pm

Highest 802.11ax datarate per spatial stream at 80mhz is 600mbps

Yup. But hAP ax2 has two spatial streams and if client also supports more than one, the theoretical datarate doubles, hence 700-800 Mbps is possible in real life.
 
DeDMorozzzz
Frequent Visitor
Frequent Visitor
Posts: 81
Joined: Tue Aug 22, 2017 7:09 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Nov 04, 2022 5:14 am

All this is great but are there any proofs that ODFMA and MU-MIMO really work?
As far as I know non of vendors make this dreams come true though this are the main changes of AX generation
AC have promised MU-MIMO but steal it's only dreams
So what is the real difference between AC\AX without that?
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Nov 04, 2022 1:13 pm

OFDMA is the easier part and doesn't bring much in case there are few concurrently active clients of same WiFi AP ... and the big difference will be mostly slightly better RTT with lower delay jitter ... the difference between OFDM and OFDMA is that the later can service multiple clients at the exactly same time, during same transmission period. Radio-wise both are the same (same modulation), it's signal processing which is more complex for OFDMA. Not something unheard of though, OFDMA is part of LTE since very beginning almost 15 years ago.

The MU-MIMO is a harder nut. The essence of MU is beam-forming and the more professional systems offering it (later releases of 4G and 5G) feature antenna arrays, e.g. 8x8 antenna array. Without antenna arrays beamforming is not effective and thus MU-MIMO can not work. Such antenna arrays are not small either, a well designed 8x8 array for 5GHz (and up) will be something like 1m x 1m panel with individual X-pol elements spaced 2 lambda apart. And add additional space for weatherproof case. Not-so-decent (and effective) indoor designs might come with 0.5m x 0.5m wall or ceiling mount antenna array. Anything smaller will have mediocre results at best.
My guess is that full-scale MU-MIMO deployment will be in point-to-multipoint deployments where the hub location will feature (large) antenna array while clients will keep using traditional (MIMO 2x2) directional antennae.
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Nov 04, 2022 1:31 pm

As to difference between ac and ax if you take MU-MIMO out of equation: in ideal radio conditions ax will run 25% faster (the difference between 256QAM in ac and 1024QAM in ax is 8 vs. 10 bits per symbol or 25%). And the possibility of having cleaner spectrum on ax at 6GHz (clean spectrum is paramount for using high modulations) - until everybody and their dogs switch over to 6GHz that is.

When talking about massive MIMO (4x4 or 8x8) one has to keep in mind that total Tx power is limited and that limit is then divided between all those Tx streams. 8x8 MIMO thusly means 6dB lower signal strength than 2x2 MIMO.
Beam forming in MU-MIMO doesn't help on Tx front either, limit is EIRP which includes beam forming gain. Beam forming does help with Rx though.
 
DeDMorozzzz
Frequent Visitor
Frequent Visitor
Posts: 81
Joined: Tue Aug 22, 2017 7:09 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Nov 04, 2022 6:08 pm

OFDMA is the easier part and doesn't bring much in case there are few concurrently active clients of same WiFi AP ... and the big difference will be mostly slightly better RTT with lower delay jitter ... the difference between OFDM and OFDMA is that the later can service multiple clients at the exactly same time, during same transmission period. Radio-wise both are the same (same modulation), it's signal processing which is more complex for OFDMA. Not something unheard of though, OFDMA is part of LTE since very beginning almost 15 years ago.

The MU-MIMO is a harder nut. The essence of MU is beam-forming and the more professional systems offering it (later releases of 4G and 5G) feature antenna arrays, e.g. 8x8 antenna array. Without antenna arrays beamforming is not effective and thus MU-MIMO can not work. Such antenna arrays are not small either, a well designed 8x8 array for 5GHz (and up) will be something like 1m x 1m panel with individual X-pol elements spaced 2 lambda apart. And add additional space for weatherproof case. Not-so-decent (and effective) indoor designs might come with 0.5m x 0.5m wall or ceiling mount antenna array. Anything smaller will have mediocre results at best.
My guess is that full-scale MU-MIMO deployment will be in point-to-multipoint deployments where the hub location will feature (large) antenna array while clients will keep using traditional (MIMO 2x2) directional antennae.
The advantages of OFDMA is not only amount of active clients. Also it should provide RU to help avoid interference. If AP knows that a client has interference in its upper part of the channel AP doesn't use that part.
MU-MIMO was announced long time ago and nothing is done. All the commercials says the AX have this but in reality it is not. So why they talk about something so long if nothing is really done?
As far as I get that MU-MIMO doesn't really need any beamforming. SU-MIMO can use lots of channels what is the difference? MIMO uses multipath.
And what is the point of talking about LTE in this thread?
Fake technologies
Mikrotik is a WISP manufacturer so I think OFDMA will bring a lot. At home I use N and wires, Its really works fine for me.
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Nov 04, 2022 8:55 pm

As far as I get that MU-MIMO doesn't really need any beamforming.
You get it wrong. Go and study what's MIMO and what's the MU part.

And what is the point of talking about LTE in this thread?
Fake technologies
I'm talking about LTE in this thread because both technologies that are supposed to make difference between ax and ac are already available and used in LTE, they were not newly invented. And I kind of know what they are all about, I worked 17 years as senior radio engineer for a major MNO.
 
DeDMorozzzz
Frequent Visitor
Frequent Visitor
Posts: 81
Joined: Tue Aug 22, 2017 7:09 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 4:18 am

As far as I get that MU-MIMO doesn't really need any beamforming.
You get it wrong. Go and study what's MIMO and what's the MU part.

And what is the point of talking about LTE in this thread?
Fake technologies
I'm talking about LTE in this thread because both technologies that are supposed to make difference between ax and ac are already available and used in LTE, they were not newly invented. And I kind of know what they are all about, I worked 17 years as senior radio engineer for a major MNO.
17 years hmmm... If I have worked that long I would mentioned al least much more subcarriers of AX that is now really works as far as I know.
Are you sure that MU-MIMO really needs a beam-forming? All media says that a 4*4 router can (at least is able) to transmit to 4 1*1 user at a time, isn't it?
I think you shouldn't have talked about that LTE-thing. This is far from AX.
If Wi-Fi 7 will be presented supporting a code-modulation and I'll ask - "WTF? why it doesn't support it?" Your answer will be - "Chill bro, even 3g supports it"? I don't give a damn about LTE Im talking about AX and tons of presentations made several years ago. No one writes on the box of a brand new router "Most of the technologies you have seen in the presentation are missing".
PS: As far as I get that. All MIMO types use a multipath and phase-shifting. TxBF (which also uses phase-shifting) can help but it is not necessary. If it is necessary than a 4*4 router can transmit only to 2 1*1 (or 1 2*2) clients at a time
 
User avatar
chechito
Forum Guru
Forum Guru
Posts: 2989
Joined: Sun Aug 24, 2014 3:14 am
Location: Bogota Colombia
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 5:05 am

i see this situation with each new Wi-Fi generation, behind the technical specifics the most important factor is pushing the industry and the consumer Hype

there are improvements? yes

but not so dramatic unless you have an ideal scenario, which is beyond the reality of most users
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 10:52 am

I would mentioned al least much more subcarriers of AX that is now really works as far as I know.
Yes, but since they use narrower frequency bands (4 times the number of subcarriers inside same width of channel) single symbol lasts 4 times longer, hence the same bitrate. So from users perspective nothing changes on this front.

I think you shouldn't have talked about that LTE-thing. This is far from AX.
I admit that I'm not an expert on WiFi (not nearly as much as I'm an expert for LTE), but from what I read it's exactly tge same technological details. And it's quite frequent that technology which got successfull in one application (LTE) is then used in another - similar - application (WiFi).

I can fully understand your frustration about lack of implementation of the new features which should make ax shine compared to ac. But lack of implementation doesn't mean that technology is fake, as I explained some of it is hard to implement properly. And used LTE as example where technology works (in certain conditions) ... and remember, LTE base stations come with pretty much higher price tag than average ax AP.
At the same time marketing doesn't want to point out the missing bits. At the same time all marketing depts (also in mobile markets) were notorious for their BS statements, so why would they become altruistic with ax? Do you still remember the speed ratings of ac APs, everybody with basic knowledge about how WiFi works knows the limits but that doesn't stop marketing to put unachievable numbers in big letters all over the boxes (and propaganda material) only to get many users angry when they discover the truth.

The sooner you start accepting the reality, the sooner you'll loose your anger :wink:
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 10:57 am

but not so dramatic unless you have an ideal scenario, which is beyond the reality of most users

Indeed. I'll bring up another parallel from "pro" world: 5G doesn't bring any improvement compared to 4G in low frequency bands (e.g. lower than 1.5GHz) apart from some new relatively narrow frequency bands ... because new features (such as higher order MIMO, etc.) are almost impossible to implement. It doesn't mean that there aren't cases where 5G will bring much better performance, but it will be only in certain use cases.
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 10:59 am

Hi,
could we stick to the hap ax2? I do not care in this topic about 5G...
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 11:57 am

Hi,

please check my stress test....

https://ibb.co/BZPS9Y2

at one time:
downloading torrents from NAS app
downloading thorough WIFI 6 file from NAS
bendwith test thorough Zerotier both
profil
CAKE QOS
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 4:32 pm

Bandwidth test in the test soup only serves to consume CPU resources and thus slowing the rest.
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sat Nov 05, 2022 8:52 pm

Why do you think that I have used for stress test exactly Bandwidth test...?To test CPU...
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Nov 06, 2022 3:54 pm

Why do you think ...

I've no idea what you tried to show with your test. You asked explicitly "please check my stress test" without explaining objectives.

As it seems your way of communication on this forum is not very compatible with mine, I'll refrain from commenting or answering your posts from this time forward.
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Nov 06, 2022 5:23 pm

A stress test analyzes the performance capabilities of your computer by subjecting it to maximum usage over a prolonged period of time. By pushing your processing power to the limit, CPU load tests help you evaluate how much your hardware can handle.
 
theprojectgroup
Frequent Visitor
Frequent Visitor
Posts: 99
Joined: Tue Feb 21, 2017 11:40 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Thu Nov 17, 2022 11:48 pm

Can you please post your wifi config of hap ax2?
Thanks.
 
whatever
Member
Member
Posts: 348
Joined: Thu Jun 21, 2018 9:29 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Fri Nov 18, 2022 4:05 pm

How is hap ax2 performance with wifi5 (ac wave2) clients?
I can now answer my own question: Speedtest.net mobile app on 1x1 clients shows ~300 Mbps download speed, 2x2 client gets up to ~550 Mbps. 80MHz channel width on 5GHz.
 
tritron777
just joined
Posts: 8
Joined: Sun Nov 22, 2015 7:14 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Nov 27, 2022 2:09 am

Where did you get ax2 any place to buy ax2 ax3 in Poland or around
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Tue Feb 21, 2023 1:43 pm

Hi,

how are you satisfied with the AX2 ?
 
User avatar
anav
Forum Guru
Forum Guru
Posts: 18958
Joined: Sun Feb 18, 2018 11:28 pm
Location: Nova Scotia, Canada
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Tue Feb 21, 2023 5:48 pm

gotsprings:
Why do you care about through a wall?

[/quote]
Well a wall at least implies one layer of gyprock maybe two and there could be some electric cabling in the mix and maybe even insulation.
So there is some basic level of interference in the signal. Not all walls are made of translucent rice paper.

Course, you may live in hippie dippie house and all the rooms have beads for doors :-) and wall paper instead of walls. ;-)
you did say california right....
 
holvoetn
Forum Guru
Forum Guru
Posts: 5317
Joined: Tue Apr 13, 2021 2:14 am
Location: Belgium

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Tue Feb 21, 2023 5:52 pm

Well a wall at least implies one layer of gyprock maybe two and there could be some electric cabling in the mix and maybe even insulation.
So there is some basic level of interference in the signal. Not all walls are made of translucent rice paper.
There are places were walls are made of brick and mortar. Some even concrete with steel reinforcements.
Sweet stuff for whatever radio signal you're trying to get through ...
 
User avatar
anav
Forum Guru
Forum Guru
Posts: 18958
Joined: Sun Feb 18, 2018 11:28 pm
Location: Nova Scotia, Canada
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Tue Feb 21, 2023 6:28 pm

No argument there hoelve!.

As for this AX2/3 discussion, unless your devices are magically wifi6 compatible whatever wifi6 really is???? I mean 6E is the standard that actually adds another spectrum 6Ghz!!! and WIFI 7, who knows havent cracked that nut yet but tis coming in 2024 LOL.

What the OP has discovered is simply MT doing WIFI 5 properly, thank god, and what most have discovered IMHO. In other words WAVE2 = proper wifi5 performance!

I was sitting down yesterday getting over 500Mbps off a TPLINK EAP245 wifi5 product but only 300-400 on a TPLINK eap615 Wall plate ( a wifi 6 product )
(wifi 6 = AX). The difference is that the 615 has lower power output max 21 vice 24 dbs and less antennas with slightly lower gain.
(Both have mu-mimo and ofdm)
245 - Internal Omni
• 2.4 GHz: 3× 3.5 dBi
• 5 GHz: 3× 4 dBi
24 dBm (2.4 GHz)
24 dBm (5 GHz)

615 - 2 Dual-Band Antennas
• 2.4 GHz: 2× 3 dBi
• 5 GHz: 2× 4 dBi
• FCC:
< 21 dBm (2.4 GHz)
< 21 dBm (5 GHz)

In conclusion, one will only do better than good wifi5, if utilizing a device that has ax technology as an endpoint device.
I fully expect that before even using AX, the AX3 should outperform the AX2 on wifi 5, due to better antenna gain and slightly more output power.
(no idea on the number of antennas though on ax2, and it appears only two on ax3?)

ax2
2.4 GHz (4.5 dBi), 5 GHz (4 dBi)
Up to 24db on 2.4ghz
Up to 22db on 5ghz
ax3
2.4 GHz (3.3 dBi), 5.5 GHz (4 dBi)
Up to 29db on 2.4ghz
Up to 28db on 5ghz
????? not sure what to make of those TX specs though, seems to break regulations so take those with a grain of salt.

Certainly makes things difficult when diff vendors put out diff specs where one cannot easily compare at any level software or hardware. :-((
Will be fun to compare ax3 outputs hopefully this weekend...........
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Feb 22, 2023 7:36 am

sorry guys but I do not understand to measure antenna gain....you select country and there is limitation on total power...it means antena gain+power from the device...for examlpe in CZ is 20dbi for 2,4GHz and max 30dbi for 5,4 GHz. If I can get 750Mbit from AX2 I do not need 6E or wifi 7 when cable do not handle more than 1000Mbit.

check Mikrotik website.....¨
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Feb 22, 2023 9:14 am

Radio basics: signal strength budget is

S = Tx power + Tx antenna gain - path loss + Rx antenna gain

which then has to be greater than receiver sensitivity (for link to work at all). And SINR is ratio between signal (S) and noise (which includes thermal noise, receiver noise figure, unrelated noise and interference), SINR defines achievable speeds, hence one wants SINR as high as possible. And one of means of getting SINR as high as possible is to have S high.
Also note that most realistic antennae don't have same gain in all directions, thus importance to direct antenna towards link peer (in case of PtP installation) ...

Now, sum "Tx power + Tx antenna gain" (a.k.a. EIRP) is limited by country regulations. However Rx antenna gain is not limited at all (neither is gain of Tx antenna). And since WiFi uses exactly same frequencies for both Tx and Rx, this means that Rx antenna gain on some device is same as Tx antenna gain on the that device. So while antenna gain doesn't help achieve higher radiated power<*> (due to country limitations), it does help to make signal stronger on receiving side. It's not entirely black and white as high-gain antenna will amplify also interference (depending on antenna gain pattern and angle between link peer and interferers), but with some luck (or knowledge) high-gain antenna will improve link quality.

<*> This is not the whole story though. Realistic WiFi transmitters can't transmit at same maximum power for all different radio modes, most of them experience significant Tx power backoff for higher throughputs. Publicly available data for hAP ax3 says that absolute maximum Tx power is 28dBm, but maximum Tx power for highest speeds (MCS11) is only 20dBm (numbers are for 5GHz radio). Calculation is as follows: EIRP limit for certain country and certain channel might be 30dBm and antenna gain is 5.5dBi. Meaning that max Tx power is limited to 24.5dBm. So for lower speeds, it's EIRP which limits Tx power (device max Tx power is much higher than that ... for hAP ax2 it's not so much, it's absolute max Tx power is mere 24dBm), but for higher speeds, it's device radio output (20dBm) which restricts transmitted power. So for hAP ax3, actual EIRP will range between 30dBm for low speeds (country limitation) and 25.5dBm for highest speeds (transmitter limit). If antenna gain was mediocre (e.g. 3dBi which was usual figure for previous generations of APs), actual EIRP would be between 30dBm and 23dBm (with EIRP curve dropping below 30dBm at much lower rates than it does with high gain antennae).

Also consider Rx sensitivity, which also usually drops with higher rates (-96dBm for lowest speeds v.s. -67dBm for highest rates, again for hAP ax3 and 5GHz band). If you add 0dBi (antenna with no gain) or even negative number (if using shitty antenna, unfit for certain frequency band ... happens a lot for WWAN due to wiiide frequency range used, i.e. from 600MHz up to 5GHz, which is realy hard on antennae), meaning that receiver is pretty deaf for higher rates (and then higher power of transmitter really helps). Using good Rx antennae (which are also Tx antennae) helps with reception a lot.


Also important: often mobile clients (e.g. phones, etc) aren't capable of high Tx powers, could be they can only work up to 20dBm (or even lower). As the same equation applies also for uplink, it's obvious that high gain AP antennae help very much in the uplink direction.


Ah, and BTW, antenna gain is physical property of particular antenna design ... it's not something that can be adjusted by software. ROS only takes information about it into account when calculating Tx power to be used to adhere to EIRP country regulations at all times.
 
Rox169
Member
Member
Posts: 432
Joined: Sat Sep 04, 2021 1:47 am

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Feb 22, 2023 10:31 am

Thank you for explanation. This is theory but I quickly tested AX3 vs AX2 and the signal and speed was almost the same. Maybe anyone could do proper real test AX2 vs AX3. Do not forget include test thorough wall...
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Feb 22, 2023 2:28 pm

As I wrote, differences will be seen in mediocre conditions. Which start to happen when signal strength drops to slightly above receiver sensitivity (and one should take into account also receiver sensitivity of client), where device Tx power capabilities make difference. If signal strength of AP is around -50dB (which is very realistic figure if in same room as AP), then those few dBm difference in Tx power won't make any difference. If signal strength is around -75dB, which is around or even below receiver sensitivity for highest rates, then AP with higher Tx power will manage higher throughput.

And also measure uplink speed where antenna gain difference should jump out (together with possible difference in receiver sensitivity, both hAP ax3 and ax2 have nominally the same receiver sensitivity). But again in mediocre to bad conditions where those few dB of difference in antenna gain might translate into 50% of difference in throughput. When AP signal strength is in range of 40-50dB, nothing can make much of a difference between two devices.

Mind that signal strength (as measured and shown by wireless client) is measured on beacon frames, which are transmitted at lowest supported rates and thus country regulation of EIRP kicks in. Meaning that it's very reasonable to see exactly the same signal strength from both devices. The only case where one could see difference in signal strength is due to directivity of antenna. Ideal antenna with higher gain actually trades area (or rather: volume) of coverage to signal level inside covered area (volume). Most notably this happens with directional antennae (e.g. 65° horizontal beam width, 50° vertical beam width) but also with "omnidirectional" antennae. Which are not really omnidirectional (which would mean spherically symmetrical pattern), they are more like a doughnut (toroidal symmetry) with "hole in coverage" in vertical direction (i.e. along the axis of "stick" antenna), they are (more or less) omnidirectional only in horizontal directions. So omni antennae with gain higher than 2.15 dBi (ideal dipole antenna with length of half wavelength) get that gain by reducing vertical beam width ... So if using AP with high gain antennae it is possible that signal strength is actually lower than with "plain" antennae ... in some directions. Or the other way around: it is possible to find areas with better signal strength when using AP with antennae with lower gain.

And believe me, this is not only theory. Speaking with 17 years of experience as senior radio engineer for mobile network operator. The difference between MNO equipment and (all!) WiFI APs is that WiFi gear is shitty compared to MNO equipment, starting from Tx amplifiers (linearity sucks) to Rx sensitivity to (and not limited to) antennae supplied with APs (with rare exceptions). But the principles are the same.
 
User avatar
anav
Forum Guru
Forum Guru
Posts: 18958
Joined: Sun Feb 18, 2018 11:28 pm
Location: Nova Scotia, Canada
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Feb 22, 2023 4:58 pm

Speaking with 17 years of experience as senior radio engineer for mobile network operator.
Assuming you were a junior engineer for sometime, and this 17 years is an old job......how old are you gramps??? :-)
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Wed Feb 22, 2023 9:22 pm

I was born as senior ... you grumpy old fart :-P
 
User avatar
anav
Forum Guru
Forum Guru
Posts: 18958
Joined: Sun Feb 18, 2018 11:28 pm
Location: Nova Scotia, Canada
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 4:50 pm

Can I combine the 2.4 and 5ghz into one configuration................ so the router selects what the device will connect two..........
Can do this on tplink called band steering.
 
holvoetn
Forum Guru
Forum Guru
Posts: 5317
Joined: Tue Apr 13, 2021 2:14 am
Location: Belgium

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 5:03 pm

No need to post the same question in multiple threads :lol:

Yes, perfectly possible.
 
User avatar
anav
Forum Guru
Forum Guru
Posts: 18958
Joined: Sun Feb 18, 2018 11:28 pm
Location: Nova Scotia, Canada
Contact:

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 8:05 pm

I was hoping for a more intelligent answer in this thread with different players! ;-PPP
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 8:41 pm

The wifiwave2 configuration tree (in CLI, dunno about GUIs) contains branch called security. There you define a profile. When provisioning wireless interfaces (either master or virtual), refer to the profile instead of setting things directly. This way you can be sure that all interfaces (which use that security profile) actually have identical security settings, ensuring as smooth client roaming as it gets. The only remaining thing is to make sure that datapath (can be configured as profile as well) settings make those wireless interfaces member of same L2 broadcast domain (i.e. (V)LAN).

But when it comes to router selecting wireless interface to which client will connect ... well not in 802.11 networks, here we have democracy and clients decide what they'll do.
 
holvoetn
Forum Guru
Forum Guru
Posts: 5317
Joined: Tue Apr 13, 2021 2:14 am
Location: Belgium

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 8:49 pm

If you have experience using capsman in the past, wifiwave2 uses a very similar structure for setting it up.
Both CLI as GUI.
 
User avatar
mkx
Forum Guru
Forum Guru
Posts: 11381
Joined: Thu Mar 03, 2016 10:23 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 9:10 pm

He successfully avoided capsman so far ... karma bit him :wink:
 
Joe1vm
newbie
Posts: 28
Joined: Sat Apr 06, 2013 4:07 pm

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 10:14 pm

If you have experience using capsman in the past, wifiwave2 uses a very similar structure for setting it up.
Both CLI as GUI.
Hi.
GUI for CAPSMAN with wifiwave2? Where could I find it?
 
holvoetn
Forum Guru
Forum Guru
Posts: 5317
Joined: Tue Apr 13, 2021 2:14 am
Location: Belgium

Re: hAP ax² WiFi6 (802.11ax) Initial Tests, good news!

Sun Feb 26, 2023 10:20 pm

Hold your horses.
The old capsman. For now.

Who is online

Users browsing this forum: No registered users and 22 guests