I've been testing mikrotik outdoor ac devices at home (inside the house) and came up with some nice speeds that were quite stable.
on screenshot is 802.11ac 80mhz link that is very steady at 420-430mbit.
with nv2 i was able to get that a bit higher (450-460mbit peaking at 500mbit).

this was done in ideal situation , antennas 3 m apart, output power on 0dBm.

can such a result (or at least 250-300mbit) be expected outdoor on a short link (about 500m) on clean spectrum?

local 5ghz is not too crowded and we have more than 200mhz of continous free spectrum, although there're a lot of foreign beacons coming from the nearby country (by ssid and radio name i pinpointed the aps to be about 120-150km away from me, mounted on high mountains) but the received signal here (sea level urban area) is about -87 to -90 dBm.
on my old AN links this foreign noise is not causing any harm.

It's hard to get these speeds in the field, but it's not impossible.
What are devices?
What is the ROS version?

tested with sxt ac, but for real outdoor link will use disc 5 ac. minimum price increase for 5dBi better antenna.
ros last long term stable.

can such a result (or at least 250-300mbit) be expected outdoor on a short link (about 500m) on clean spectrum?

Why not? But no one can predict exactly what will happen.
Just one example: How clear will your fresnel zone be?
It should be very open, and not even touching the ground (!)
https://www.everythingrf.com/rf-calcula ... calculator The remote 802.11 transmitters at -87 or -90 dBm might make your own transmitter wait. It is still decodable strength! Setting "Adaptive Noise Immunity" might help enough to ignore that level. Manual setting of the noise "noise-floor-threshold" is not possible with these newer 802.11ac transmitters AFAIK.
However nv2 will just ignore that, as it is not a "clear-channel-listener waiting it's chances before talking". nv2 will just talk (at least that is what I read about nv2 [*]) , and will have to cope with interference, but that should be no problem with those low levels.

[*] from the wiki: "As Nv2 does not use CSMA technology it may disturb any other network in the same channel. In the same way other networks may disturb Nv2 network, because every other signal is considered noise."
I don't know what "nv2-noise-floor-offset" does (CLI). Probably the same story as "noise-floor-threshold".

PS: there is more information on the connection in the "registration table" than in the statistics and status page

However nv2 will just ignore that, as it is not a "clear-channel-listener waiting it's chances before talking". nv2 will just talk (at least that is what I read about nv2 [*]) , and will have to cope with interference, but that should be no problem with those low levels. [*] from the wiki: "As Nv2 does not use CSMA technology it may disturb any other network in the same channel. In the same way other networks may disturb Nv2 network, because every other signal is considered noise."

In view of this, should we still be using NV2 on 802.11ac, especially for such short links without the risk of any hidden-node issue?
Ref. this post from @Normis in 2019 viewtopic.php?t=136002&start=500#p756033:

Nv2 was designed for older chipsets, to solve problems that were there at the time. New chipsets and 802.11ac do not have the same issues, so there is no real benefit to use Nv2 now.

Or I may be misinterpreting/misunderstood that recommendation?

There were serious problems in 2019 with ARM and also with NV2. Those have clearly been solved. Saying that one of the protocols is better than the other today is not always correct. viewtopic.php?mode=quote&f=7&p=838112

My current experience ( 5 SXT sa5 ac and 15 SXTsq ac) on one 1km2 area, 100+ client devices, have made me switch all of the SXT AP's to NV2 for best performance. 40MHz channels, driving 20 hAP ac2 and 15 wAP ac. Surrounded by quite some interfering long-range AP's. It''s by experimenting over a year, and I'm not alone in this forum in 2021 with similar outcome.

Nv2 saved this setup, but any other area might give different results. Test and compare.

Okay, I'll try out the different protocols, and assess performance.
Many thanks for sharing your experience.

One of the differentiators with nv2 is: can you handle all 802.11 traffic in your channel just as noise, or do you need to co-exist (co-channel interference, wait for channel clear) with them?

If other transmitted signals are weak enough ( eg 40 dB lower ???) you can just handle them always as noise with nv2. Is actually my case, and seems also to be in OP's case.

Channel(s) selection is critical in this.

Hi miko385!

I have some trouble to reach that kind of speed in a ptp (viewtopic.php?f=7&t=177916&p=875846#p875846), can you post your configuration? which models are you using?

These speeds UDP or TCP?

Hi miko385!

I have some trouble to reach that kind of speed in a ptp (viewtopic.php?f=7&t=177916&p=875846#p875846), can you post your configuration? which models are you using?

straightforward , no config at all.
reset config, add bridge, bridge ports, and setup a wireless bridge (bridge + station bridge setup) with classic wpa2 aes encryption.

everything else is default.