Posted: Wed Nov 12, 2025 12:34 Post subject: R6300v1: 802.11ac performance
Hello,
today I have noticed that the 802.11ac performance of my R6300v1 is slightly suboptimal. I get ~360 Mbit/s sending and ~230 Mbit/receiving via iperf3 when right next to the AP. This translates to approximately 200 Mbit/s max in speedtest, which is a shame because I have a 300 up and 150 down Mbit/s connection.
When connected directly to my FritzBox 7530 acting as a router, I get ~460 Mbit/s sending and ~440 Mbit/s receiving, which means I can max out the 300 Mbit/s line. This is also more in the range of what I would expect the maximum real world throughput of an 866 Mbit/s 802.11ac link to be.
I have followed most of the guide here [1], except the numeric settings.
Can the fact that the iperf sender, receiver and the AP are connected via the same RJ45 jack to the router? But then it should not influence the speedtest.
Channel is also clear according to WiFiAnalyzer.
Is this performance level expected?
Joined: 18 Mar 2014 Posts: 13880 Location: Netherlands
Posted: Wed Nov 12, 2025 16:54 Post subject: Re: R6300v1: 802.11ac performance
belegdol wrote:
Hello,
today I have noticed that the 802.11ac performance of my R6300v1 is slightly suboptimal. I get ~360 Mbit/s sending and ~230 Mbit/receiving via iperf3 when right next to the AP. This translates to approximately 200 Mbit/s max in speedtest, which is a shame because I have a 300 up and 150 down Mbit/s connection.
When connected directly to my FritzBox 7530 acting as a router, I get ~460 Mbit/s sending and ~440 Mbit/s receiving, which means I can max out the 300 Mbit/s line. This is also more in the range of what I would expect the maximum real world throughput of an 866 Mbit/s 802.11ac link to be.
I have followed most of the guide here [1], except the numeric settings.
Can the fact that the iperf sender, receiver and the AP are connected via the same RJ45 jack to the router? But then it should not influence the speedtest.
Channel is also clear according to WiFiAnalyzer.
Is this performance level expected?
I do not have your router so not 100% sure but it is a single core 600 MHz, your Fritzbox is quad core 800 MHz so that could explain the difference.
What could help a bit is if your router has CTF & FA as flow off load (see basic setup page) but I doubt that is the case, you can try to use CTF instead of SFE and see if that makes a difference but I doubt that is the case.
So this might just be what is possible on these old single core low clocked routers but as said I do not have it so not sure.
My R6400v2 dual core 1000 MHz tops out at around 450 Mb/s, NetGear R7800 550 Mb/s and my AX Dynalink DL-WRX35 does around 750 Mb/s under ideal circumstances all measured at 5 GHz 80 MHz channel width and iperf server running on a PC attached via ethernet on the same subnet _________________ Routers:Netgear R7000, R6400v1, R6400v2, EA6900 (XvortexCFE), E2000, E1200v1, WRT54GS v1.
Install guide R6400v2, R6700v3,XR300:https://forum.dd-wrt.com/phpBB2/viewtopic.php?t=316399 Install guide R7800/XR500:https://forum.dd-wrt.com/phpBB2/viewtopic.php?t=320614 Forum Guide Lines (important read):https://forum.dd-wrt.com/phpBB2/viewtopic.php?t=324087
CTF/SFE are an unlikely culprit. Apologies if this was not clear, but I am using my R6300v1 as a Wireless AP/Switch, it is not doing any routing. It is connected to the FritzBox via ethernet.
Out of curiosity, if the desktop PC is attached to Netgear's LAN and the laptop to it's WiFi, do the packets need to travel to the router and back via Ethernet? Or can the R6300 route (or rather switch) the packets between the devices directly?
I guess the time has come to finally retire this device after almost fifteen years. The next question becomes whether I should grab a used FritzBox 7520 (802.11ac) for a couple dozen euros, or finally modernise and get something supporting 802.11ax or even 802.11be. Are real world improvements worth it? While I do have some ax clients, the improvements seem fairly modest [1].
Joined: 18 Mar 2014 Posts: 13880 Location: Netherlands
Posted: Wed Nov 12, 2025 17:34 Post subject:
belegdol wrote:
CTF/SFE are an unlikely culprit. Apologies if this was not clear, but I am using my R6300v1 as a Wireless AP/Switch, it is not doing any routing. It is connected to the FritzBox via ethernet.
Out of curiosity, if the desktop PC is attached to Netgear's LAN and the laptop to it's WiFi, do the packets need to travel to the router and back via Ethernet? Or can the R6300 route (or rather switch) the packets between the devices directly?
I guess the time has come to finally retire this device after almost fifteen years. The next question becomes whether I should grab a used FritzBox 7520 (802.11ac) for a couple dozen euros, or finally modernise and get something supporting 802.11ax or even 802.11be. Are real world improvements worth it? While I do have some ax clients, the improvements seem fairly modest [1].
CTF&FA also works across the bridge so it can help a bit even if you are using this as a Wireless Access Point (it does not directly offload wireless but keeps the CPU free so it does not do wonders in this case as it does for LAN<>WAN throughput), but yes it probabaly is time to get something better.
I would not go for "be" but "ax" is worth considering, if you can find something below < E 80.
SFE was already enabled. @Per Yngve Berg is right, the CPU usage spikes when running iperf. I do not quite understand why the load would be lower when receiving data than when sending it, but whatever.
I tried overclocking and upped the frequency to 632 MHz. Router rebooted upon clicking Save and Apply and took really long time to come back up, is this normal? I initially thought I bricked it.
There appears to be no significant change in thoughput anyway.
I then tried CTF and it indeed worked wonders, iperf got up to ~380 Mbit/s downloading and I was able to max out my ISP connection. Thanks!
ETA: lack of IPv6 support is mentioned as one of CTF's drawbacks. Does it matter for devices running in AP/Switch mode? My ISP provides internet access via DSLite so IPv6 support is pretty essential, but I was not able to spot any issues yet and https://test-ipv6.com/ shows 10/10.
"Until it doesn't" is a fair description. I was on 62418 when I tested it and CTF helped. I now upgraded to 62606 and I am again stuck at 200 Mbit/s when doing a speedtest. I will see if this is just a fluke, but I might need to downgrade otherwise.
ETA: I downgraded and it is still slow. I will test again later, hopefully the improved speed is not gone forever.
The slowness would not go away in its own unfortunately. I then set SFE to Disabled and then back to CTF, and the speed came back. So it might have had anything to do with the update, but I am a bit afraid to try again.
OK, I think I got this narrowed down. The performance suddenly disappearing seems to be caused not by upgrading dd-wrt, but rather by rebooting the router. Back on 62418 I am noticing the following behaviour:
- boot up with SFE set to CTF: slow performance, iperf3 capping at around 250 Mbit/s
- disabling SFE entirely: 260 Mbit/s
- re-enabling CTF: 500 Mbit/s
So it appears that whatever enabling CTF does to WLAN <-> LAN switching performance, it only has an affect if CTF has been enabled via the UI after booting the device.
It is definitely odd. I have now confirmed the same behaviour with 62778.I am attaching the syslog in case there is anything interesting in it. I re-enabled CTF at 21:07.
