I have not had a great experience with Wi-Fi performance with comparison to stock on an Archer C20 v1; the range seems to have worsened a bit and it used to cut out fairly often. I can't judge throughput differences because they are already capped low at ~70Mb/s.
Now I've bought EAP615-Wall and I'm wondering whether or not to install OpenWrt.
Have any of you compared the throughput, range, and maybe even power consumption before and after installing OWrt?
Thanks.
On a side note, I've been wondering if EAP615-Wall is even the "optimal" choice in its price range.
Would someone suggest a different model in a similar price-range that is more suited to general home use? Range, low power consumption, and Wi-Fi 6 are the priorities.
EDIT: I live in Europe.
So this needs to be a pin posted. The stock WiFi firmware will ,most of the time, out perform OpenWRT WiFi. There is no way around it. It's been like that for 10-15 years. People will disagree but they are wrong, bias or ignorant.
You really need to forget WiFi on OpenWRT and get a router that runs OpenWRT and external WAP like Unifi or tp-link.
That's what I was afraid of.
I very much agree that this should be mentioned in a pinned post here.
It would be interesting to see an OEM stock- vs OpenWrt firmware comparison for a matured device like the EAP615, i.e. using tools like
ps: I have a spare EAP615 lying around so might see if I can run some tests this weekend (provided that I succeed in reverting from OpenWrt- to OEM firmware)
edit: note to myself: I need to bring back an old server to life as backend for iperf3 server and netperf (netserver)
Good idea. I'll try doing some tests of mine in the next few weeks too and post them here.
Is dumping them in this thread OK or should we start a new one?
You could post results in this thread as far as I am concerned (unless the admins have objections)
fyi my plan is to make tests with 5-10 measure points using
I have an iPhone11 for 802.11ax iperf3 tests and a laptop for 802.11n/802.11ac netperf/wifi survey tests
ps: take note of following EAP615-WALL: installation fails with "Failed to check for new update"
If your stock firmware contains v1.1.7 or higher then you can currently only install an OpenWrt snapshot version. (installing 23.05.2 is possible as well, but then you need to edit the .bin file with a hex editor)
Edit: I will place the EAP615 at 2m height during the tests
On a sidenote: besides whether Wifi performance is better or not, there are also other reasons for installing OpenWrt on an Access Point like
I agree, but I wouldn't want to inconvenience other people living with me by fiddling around with an actively used device just for the sake of it. I can always entertain myself by poking around in other devices.
This morning I ran a number of tests with OpenWrt firmware on EAP615.
Tomorrow I will run same with OEM stock firmware (provided I succeed in installing this)
router - RPi4
switch - 8x 1Gb port
accesspt - EAP615 placed outside house in shed at 2m height
server - Ubuntu 24 running iperf3 -s -D
laptop1 - Ubuntu 24 with Intel AC8265 802.11n/ac
laptop2 - Windows11 with Intel AX211 802.11ax
phone - iPhone11 802.11ax
ch1 2.4Ghz, 20Mhz, 20 dBm (100mW)
ch36 5Ghz, 80Mhz, 21 dBm (125mW)
ch100 5Ghz, 80Mhz, 21 dBm (125mW)
Used apps
$ xhost +local:docker
$ sudo docker run --net="host" --privileged --name survey -it --rm -v $(pwd):/pwd -w /pwd -e DISPLAY=$DISPLAY -v "$HOME/.Xauthority:/root/.Xauthority:ro" \
jantman/python-wifi-survey-heatmap wifi-survey -i wlp2s0 -s <host> -p floorplan.png -t ch1
$ sudo docker run -it --rm -v $(pwd):/pwd -w /pwd jantman/python-wifi-survey-heatmap wifi-heatmap ch1
Results - Stock 1.1.9 vs OpenWrt SNAPSHOT r24723
1.) Wireless throughput in Mbits/sec sta > server 1m no obstacles
phone laptop1 laptop2 cpu usage
Stock OpenWrt Stock OpenWrt Stock OpenWrt Stock OpenWrt
ch1 144 141 101 100 161 161 20% todo
ch36 748 829 448 431 n/a n/a 30% todo
ch100 - 750 - 472
2.) Wireless throughput ch1 in Mbits/sec sta > server on 1st- and 2nd floor
phone laptop1 laptop2
Stock OpenWrt Stock OpenWrt Stock OpenWrt
1st floor 15 20 35 40 30 30
2nd floor 0 4 20 15 5 10
3.) wireless survey: Stock (OEM) vs OpenWrt UDP Download jitter & TCP Download
Notes:
@Ruscal Result comparison:
Hmmm... I attempted to revert to stock with these instructions https://openwrt.org/docs/guide-user/troubleshooting/vendor_specific_rescue#rescue_firmware_partition but I somehow can't access the 'mini-web server'
edit: turns out that the web server is located at 192.168.0.254 instead of 192.168.0.1
Unfortunately uploading a stock firmware failed at 40% (I will create a new topic for this - see EAP615-Wall reverting from OpenWrt to OEM fails)
edit Jan 13th: I finally succeeded in reverting to stock and will resume tests tomorrow.
Did you ever get a chance to test the performance with stock firmware? I’m really curious to know how Openert fares with mt76 since it’s supposed to be one of the most actively developed platform
Because it's actively developed, so it shows even more differences, I am not talking about signalling here (since I don't have proper measurement tools), but the features, and issue fixes. Just like previously there is VHT160 problem on some devices, causing a huge performance degrade when using with Apple devices (which is mainly Broadcom based device), for stock firmware, when there is no update, you get no fix, so people will say those routers do not work with Apple and buy a new one (at least I've seen multiple forum discussions about it), some routers are capable to flash OpenWrt which I told them to try, and they found it more usable afterwards.
Yes, I tested stock firmware as well - results are in the edited post from 18d ago.
(thus far I did not see a noticeable difference between Stock and OpenWrt)
