Wauw! Finally! When can I order?

Waiting for this for long time.... Can't wait to test use it....


So, how long do we now have to wait for similar but a/c?

What can be the possible scenarios of use of this routerboard?

SFP to tower, no losses, no interference.
High speed CPU for fast 3 chain links, the only other high power device for wireless is the RB800, and that is big and expensive.

Does anyone know of any proven 3x3 MIMO 5GHz antennas? Ideally for PtP links

SFP to tower, no losses, no interference.
High speed CPU for fast 3 chain links, the only other high power device for wireless is the RB800, and that is big and expensive.

Or setup on tower with 3 single chain sector antennas in full sync! I've done so with dual chain, now you got one chain more, and more power as a bonus!

And yes, the rb800 is nice but too big for most boxes. Finally a new powerful board that fit in my standard boxes.

Does anyone know of any proven 3x3 MIMO 5GHz antennas? Ideally for PtP links

hit your search engine and find some on the market. But "proven", I presume that is to find out by yourself. Because the performance of any antenna system is a product of the used technology, spectrum, spectral circumstances (interferences, obstructions, fresnell), usage etc. etc. etc.
What might work for one is totally un-acceptable for another.

I made myself a 12km triple chain link with 2x2 Jirious dome and one diagonal mesh antenna on top, but probably due to interferences and/or reflection of a chimney can't get no more than about 120Mb aggregated tcp traffic over the link. I can switch on or off one chain (any chain) without loosing any throughput, so obviously something is wrong...

I am planning some new links for high throughputs so I am definitely going to use these new boards and thinking of what antenna's exactly to use.

But the main thing in mho is that you need better than 25dBi antennas to get good performances on medium to long backhauls (+10km) in 'noisy' spectrum. Since I haven't found any triple chain dish antenna (yet?) I have to play with an extra antenna.

It might be that 802.11ac dual chain setups might bypass my high bandwidth demand wishes even before good high gain triple chain antenna's will hit the market....

So Normis, next newsletter with some info on ac? (Higher power needed than the present available range of radio's.)
Why not this 952GS-5Hnt delivered with 802.11ac? That would shake the competition!
And make me happy!!!

I made myself a 12km triple chain link with 2x2 Jirious dome and one diagonal mesh antenna on top, but probably due to interferences and/or reflection of a chimney can't get no more than about 120Mb aggregated tcp traffic over the link. I can switch on or off one chain (any chain) without loosing any throughput, so obviously something is wrong...

I am planning some new links for high throughputs so I am definitely going to use these new boards and thinking of what antenna's exactly to use.

But the main thing in mho is that you need better than 25dBi antennas to get good performances on medium to long backhauls (+10km) in 'noisy' spectrum. Since I haven't found any triple chain dish antenna (yet?) I have to play with an extra antenna.

From searching the net it looks like experimentation like you're doing is the way to go.
I came across a study of various angular spacings of polarizations: http://alog.ipduh.com/2013/09/5ghz-3x3- ... eeder.html but it looks like the conclusion is that physical spacing is going to work much better than different polarizations. So much for the 0°/120°/240° 3x3 MIMO dish.

It's pretty noise-free where I'm based so when my preferred supplier gets some cards in I'll set up a test-bed with spaced antennas.

MCS-23 with 30/60MHz channels tops out at 675Mbps? A 2x2 / MCS-14 802.11ac (highest 64-QAM) would be 650Mbps by trading the extra antenna for larger bandwidth (80MHz). You can't get a MikroTik 802.11ac card right now though...

Plus 802.11n allows up to 4x4 / MCS-31 which would be 900Mbps and this is higher than 2x2 / MCS-18 80MHz 802.11ac at 867Mbps (just). Probably more reliable as well I reckon because surely 256-QAM is more sensitive to distortions than 64-QAM? Two 2x2 MIMO antennas spaced apart and rotated through 45°?

[edit: note that MCS numbers are different between 802.11n and 802.11ac]

I can switch on or off one chain (any chain) without loosing any throughput, so obviously something is wrong...

There is nothing wrong with 3rd chain, you simply can't use it to aggregate more throughput because of a
low chain-to-chain isolation if you use all 3 of them in p2p configuration.

Is there an enclosure available designed for this device?

Ther is something like this:
http://www.cyberbajt.pl/produkt/4332/gi ... o-3x3.html

I can switch on or off one chain (any chain) without loosing any throughput, so obviously something is wrong...

There is nothing wrong with 3rd chain, you simply can't use it to aggregate more throughput because of a
low chain-to-chain isolation if you use all 3 of them in p2p configuration.

Its an old topic and an old remark; Do you still stand by this?

I mean, if you can't use triple chain on p2p what is the use of triple chain than anyway? In a p2mp environment? How? You still need a 3x3 station to connect to the 3x3 ap and then we have the same issue.
Or are you saying that in such case the benefit to a 2 chain or even 1 chain station might be there will always be one chain from the ap giving a better signal to the station than one of the two others?

@WirelessRudy
Well, let's put it this way - has anyone ever seen any professional grade backhaul p2p link with
more than 2 polarizations at the same carrier frequency (more than 2x2)?
Ceragon (Nera), Radwin, Saf, Altai, Exalt, Cambium....
Nope, there is 0 (zero) 3x3 or 4x4 p2p products from those companies.
If something is too good to be true it probably isn't.

Considering frequency allocation is a major limit on 5G, especially if you're going to use 80MHz 802.11ac channels, then adding more chains is a big plus.

I'd find it way easier to add more antennas with physical separation than finding more frequency allocation.

4x4 MIMO 802.11ac or 802.11n miniPCIe cards please!