Wi‑Fi 7 / 802.11be

Just a nice to have…

With WiFi 6 / ax having barely been introduced into Mikrotik harware lineup as of this general message. I read Wi-Fi 7 is not far off.
Further reading I see Mikrotik’s favorite supplier Qualcomm is now producing chips with Wi-Fi 7 onboard. ( IPQ9574 / PQ9554 / etc etc)
Would it not be great to see MT get on the front foot and start implementing some of these chips into upcoming products( even if RoS is not ready just yet ). And I would think ideally into MT’s ceiling mounted gear 1st up(CaP AC/ Cap XL etc)..
wifi 7.png

CPU – Quad-core Arm Cortex-A73 @ 2.2 GHz
System Memory – DDR3L, DDR4 16/32-bit
Storage – eMMC, NAND, Serial NOR, SD/eMMC
Networking
Wired – 6 Port Integrated Ethernet Switch 4 x 2.5 GbE + 5 GbE + 10 GbE
Wireless
Wi-Fi Standards
Wi-Fi 7 (802.11be), Wi-Fi 6E, Wi-Fi 6 (802.11ax),
Wi-Fi 5 (802.11ac), Wi-Fi 4 (802.11n), 802.11a/b/g
Wi-Fi Spectral Bands – 6GHz, 5GHz, 2.4GHz
Spatial Streams – 16 streams
Channel Support – 320MHz, 249MHz, 160MHz, 80MHz, 40MHz, 20MHz
Peak Speed – 33 Gbps PHY rate, up to 10+ Gbps PHY rate per channel
Wi-Fi Features
Simultaneous & Alternating Multi Link Puncturing for wideband operation in presence of 5G/DFS & 6G/AFC interferers
4K QAM
UL/DL OFDMA up to 37 users per channel
UL/DL MU-MIMO, up to 8 users per channel
Sustained throughput for up to 2000 users
TxBF, 802.11be QoS
Wi-Fi Security Suite
WPA3 Personal, WPA3 Enterprise, WPA3 Enhanced Open, WPA3 Easy Connect, WPA2, WPS, 802.11i security
AES-CCMP, AES-GCMP, PRNG, TKIP, WAPI2, WAPI1, WEP
Platform Extensions – Bluetooth, 802.15.4 (Zigbee/Thread), 4G/5G FWA
Packet Processing – WAN tunnel offload engine, integrated Ethernet switch, enhanced security processor, high-performance QoS/TM
Peripherals
Audio – I2S
USB – USB 3.0
I2C, SDIO, SPI, UART
4x PCIe 3.0 controllers
2x USXGMII (Universal Serial 10GE Media Independent Interface), 1x USXGMII-M
Process Technology – 14nm

https://www.qualcomm.com/products/application/wireless-networks/wi-fi-networks/networking-pro-series#product-list

“Qualcomm says the new WiFi 7 Networking Pro SoCs can run Openwrt with Linux Kernel 5.4. With up to 2000 clients, the Networking Pro 1620 is designed for highly-congested venues (e.g. stadiums), enterprise, small-to-medium business, and prosumer home environments. The Networking Pro 1220 and 820 both target enterprise, SMB, prosumer, and premium home mesh systems, while the Networking Pro 620 is better suited for enterprise, SMB, gaming, and home mesh systems.”

For me in that tabel the 6GHz is the most interesting improvement, as there is more interfering 5GHz around us already. Wifi 6E , but probably need Wifi 7 to implement 6GHz properly.
The other things are more incremental gains, that will need undisturbed signal for those high MCS encoding, IMHO.

I seriously doubt that Mikrotik will come out with a “WiFi 7” product in less than 2 or 3 years from now.
It also looks like Mikrotik totally missed the “Wi-Fi 6e” market.

Some ( non-Mikrotik ) WiFi 7 , Wi-Fi 7 products are shipping right now.

IMO , Mikrotik has great reliable wireless products for Wi-Fi 5 products but hardly anything for Wi-Fi 6. I’ve been waiting for years for Mikrotik to make a product that operates in the 6 GHz to low 7-GHz bands.

Well to be fair TPLINK just live streamed its wifi7 product release this past week or so, and none of the business class APs are showing up on their websites yet.

Dont forget TP-LINK mention 2024 as the year expected for wi-fi 7

Will we ever get a decent MU-MIMO and OFDMA?
New standards are just broader channels and more chains to show more numbers on the paper. 4-8-16 chains give a lot of throughput but have anyone seen a wireless client with more than 2*2 MIMO?
I hardly believe that one can’t find a free 5GHz channel indoors. Is it so important to have 6-7GHz?
MU-MIMO is not working so what is the different?
Just a broader channel and higher potential MCS (u hardly get)

Of all the new bells and whistles coming with WiF7, I’m most optimistic about MU-MIMO because it mostly relies on AP’s capabilities and it isn’t impossible to build AP with large array of antennae (which will make AP huge or ugly or both). But that doesn’t increase speed for clients when there’s only very few active at a time (i.e. in typical SOHO case). So the benefit of it will only be there for large commercial installations.

The rest of features are, as already written, harder to achieve and it’s quite likely that we’ll see even larger discrepancy between theoretical capacity and reality.

Noting the expected 2024 timeframe for most manufacturers to have product(s) ready.
It would be now where the R&D teams should be getting there sample/test boards ready for a ~1year turn-around to market.

Of all the new bells and whistles coming with WiF7, I’m most optimistic about MU-MIMO because it mostly relies on AP’s capabilities and it isn’t impossible to build AP with large array of antennae (which will make AP huge or ugly or both). But that doesn’t increase speed for clients when there’s only very few active at a time (i.e. in typical SOHO case). So the benefit of it will only be there for large commercial installations.

As I stated earlier, probably a good initial product would be the Cap XL(RBcAPGi-5acD2nD-XL) type of product to get this gear into.
I currently run a campus with about 60 roof mounted MT AP’s at the moment, and could easily see the new WiFi-7 boards being utilized in 2 years time..

yeah… 320mhz @ 4096qam with an 16x16 array…

the baseband/DSP alone will burn a lot of power…

the only “good” thing about “wifi7” is the “MU” part, and maybe some improvements to how OFDMA works.

the laws of physics will keep us closer to 40mhz@2x2 streams for most applications.


i would rather have working wave2+capsman for the existing AX line and CapAC.
That will have some real impact in real scenarios, with real phones that real people have from the last 5+ years
would make a lot more people happy that way (let the half a dozen wifi{next} enthusiasts suffer with tplink for a while)

Why no one is talking about MLO… That for me is the biggest improvement on Wifi7, at least on performance even without a huge Channel width..

320Mb channel width? … say goodbye to 16 channels at once … a few devices will “kill” whole block of flats especially if default configuratin set the max power “just to be safe not to be weekest one”.
BTW … what channel combintaion in 5GHz spectrum will let using 320MHz mode?

The cAP AC XL that was released in 2022 is Wifi ACv1 with no upgrade path.

So 2014 radio standard… 8 years later.

The audience wifi wave2 driver would bring it to ACv2. A 2016 standard. Which they are just getting into caps-man in 2023… 7 years late.

Outside of the 60Ghz PtP line… It’s pretty hard to take Mikrotik WiFi seriously anymore.

So wifi7 is now here. What does it look like?
Is there already hardware with 6 GHz?
I’ve had your hardware with 60 GHz in the house for many years and would love to have everything from a single source.

wifi 6 started in 2019 and MT has new products now. Wifi 7 started now so wait 4-5 yers…

Not that long. AX took longer, since we had to move from our own drivers to chipset manufacturer drivers. From now on with wifi.npk package, it will be much easier.

Ahh so you are at work Normands, still waiting for the detailed response to this post ( you can use email if you prefer ) http://forum.mikrotik.com/t/new-feature-back-to-home-vpn/168434/243

Looking at some of the MT YT videos, and the consumer products line up
I come to conclusion that MT is targeting products that “wider public/consumers”
will buy, rather then creating “latest” products …

Or, as it’s evident from the YT videos, operators or large ISP’s in bulk
at least for “consumer grade products (SOHO)”.


I could be wrong …

I am waiting for some sort of “blanket refund” for the CRS354.

Just put it down as a business loss, sending you postage to send to my location. :slight_smile:

Its not something I would consider funny.

https://forum.mikrotik.com/viewtopic.php?t=160561&start=300