Wireguard at 200 Mbps does seem too high for MT7621. Wireguard on an ER-X (MT7621) at 160 Mbps mean is reported here.
Not sure what you mean by qca9563 being faster than the MT7621 though. The qca9563 is single core 775 MHz MIPS. The MT7621 is dual core 880 MHz MIPS. Wireguard also can make use of multiple cores. I'd be quite surprised if a qca9563 was faster.
FWIW, I found my ER-X to be speedier than an Archer C7 for SQM - don't recall specifics, but the ER-X was definitely an upgrade.
mips 24k vs 74k. The latter is faster per clock.
My experience being my ER-X was faster than my C7, this piqued my curiosity.
The 1004K MT7621 is indeed slower per MHz at 1.6 DMIPS/MHz and 3.05 Coremarks/MHz than the 74K qca9563 at 1.93 DMIPS/MHz and 3.48 Coremarks/MHz (ref: https://www.mips.com/products/classic/).
However, the MT7621 looks like it makes up the Coremarks performance difference per core with its higher clock rate (~14% higher clock rate, ~14% lower Coremarks/MHz), plus it adds an extra core.
Sure but that implies ethernet is not limited to one core (it is). I'm confident the ethernet driver is providing a speedup.
bhphotovideo is showing this particular model as 'discontinued' https://www.bhphotovideo.com/c/product/1713237-REG/netgear_wax202_100nas_wifi_6_ax1800_dual.html RIP can't get anymore
could try ebay. there are usually deals there.
Over here in Europe there's no indication it's being phased out.
Tests done at 28 dBm transmit power. I'm doing ordinary ookla speedtests since I have 1gig FTTH and the server I've been using can easily reach 940/280 (full bandwidth)
Output of iw dev wlp3s0 link command:
Connected to 80:cc:9c:ea:4c:fe (on wlp3s0)
SSID: Soggiorno
freq: 5320
RX: 455816260 bytes (3294617 packets)
TX: 897691645 bytes (386840 packets)
signal: -41 dBm
rx bitrate: 1200.9 MBit/s 80MHz HE-MCS 11 HE-NSS 2 HE-GI 0 HE-DCM 0
tx bitrate: 1200.9 MBit/s 80MHz HE-MCS 11 HE-NSS 2 HE-GI 0 HE-DCM 0
bss flags: short-slot-time
dtim period: 2
beacon int: 100
Speedtest without obstruction, 5 meters apart from the router
Speedtest by Ookla
Server: EOLO - Milan (id = 11427)
ISP: Dimensione Srl
Latency: 9.24 ms (0.31 ms jitter)
Download: 747.41 Mbps (data used: 759.5 MB )
Upload: 278.27 Mbps (data used: 222.3 MB )
Packet Loss: 0.0%
Then I moved to the other room, so there's just one wall (bricks) between me and the router
Speedtest by Ookla
Server: EOLO - Milan (id = 11427)
ISP: Dimensione Srl
Latency: 8.73 ms (0.42 ms jitter)
Download: 589.68 Mbps (data used: 606.2 MB )
Upload: 278.73 Mbps (data used: 219.3 MB )
Packet Loss: 0.0%
So I then walked to the other side of the house (3 walls in between, 9 meters from the router)
Speedtest by Ookla
Server: EOLO - Milan (id = 11427)
ISP: Dimensione Srl
Latency: 9.24 ms (3.87 ms jitter)
Download: 209.33 Mbps (data used: 196.8 MB )
Upload: 108.29 Mbps (data used: 54.4 MB )
Packet Loss: 0.0%
Same test, done at 20 dBm transmit power.
Hardware used is Thinkpad X220 with AX200 wireless card.
No walls:
Speedtest by Ookla
Server: EOLO - Milan (id = 11427)
ISP: Dimensione Srl
Latency: 8.85 ms (0.47 ms jitter)
Download: 649.96 Mbps (data used: 694.9 MB )
Upload: 276.84 Mbps (data used: 324.6 MB )
Packet Loss: 0.0%
1 wall: (got better results this time)
Speedtest by Ookla
Server: EOLO - Milan (id = 11427)
ISP: Dimensione Srl
Latency: 9.23 ms (0.14 ms jitter)
Download: 610.30 Mbps (data used: 579.4 MB )
Upload: 277.79 Mbps (data used: 225.2 MB )
Packet Loss: 0.0%
3 walls: (almost lost signal)
Speedtest by Ookla
Server: EOLO - Milan (id = 11427)
ISP: Dimensione Srl
Latency: 9.44 ms (0.47 ms jitter)
Download: 88.80 Mbps (data used: 107.2 MB )
Upload: 64.49 Mbps (data used: 110.7 MB )
Packet Loss: 0.0%
In the end, I might as well consider to lower the transmit power a tiny bit, since I have other equipment in the house that help me cover it all (another wireless router in AP mode with fast roaming).
Hi, what's the lowest that Tx power reduces to?
I have my router in my office on my desk and like to have it on lowest power
If you select 3dBm for example and press save and apply, does it actually get applied?
With some WiFi adapters I tried, they don't actually let you change Tx power, even if you select.
Curious about WAX202 and if it properly applies the Tx power you select
You can set it higher for sure, I guess you could set it lower as well.
resellers NL are increasingly indicating ‘unavailable’ too. Only three shops still sell the product. Assuming it’s EOL from the manufacturers pov.
I've got one seemingly in backorder for over a week now. Was available at 60 € when I ordered but hasn't shipped yet. Hoping it comes through.
This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.