Out of curiosity, do you actually have long-term traffic volumes sustained at gigabit speeds?
It is nice to have that "full gigabit" at speed test, but the difference between 930/900 and 870/800 sounds pretty uninteresting to me from real-life perspective.
I am somewhat amused with the general eagerness to have "full gigabit" at speedtest as I believe that most servers in the wider internet will never sustain gigabit traffic with you except for short bursts. There are bottlenecks either at servers' total bandwidth or in the intermediate hops.
This is a much newer SoC, MT7622, Aarch64-based rather than MIPS and many changes made also to the Ethernet part (compared to MT7621), such as support for 2500Base-X as PHY interface mode. So this is most likely not a device-specific problem but rather a driver/SoC-specific one.
For a general use of internet and even if your are using it to see 4K videos or watch TV and have several people at home, you don't need 1 Gb ethernet at home. 500 or 300 are generally speaking more than enough.
If you have a hotel or something like that, may be you need that speed in order to get good simultaneous usage of the internet.
But most places won't give you even 300 Mbps for serving streams of data in a sustainable rate.
Some special uses like tunnelling two distant offices with huge traffic between them may justify to be worry about that peak performance.
But most of the time is more a matter of having paid for it and technical personal challenge than a real need: if we are here is because we have a geek inside us (to a greater or lesser extent).
It is not but for the community, it is good: may be we don't need it now, but when the neccesity arises there has been people trying it that has paved the way.
Did you have it set to "driver default"? For mine with 2.4 GHz disabled, it claims to be using 27 dBm. I wonder if you switch off 2.4 GHz if it will jump up?
EDIT: I think it's more about which channel you use... lower channels for me (36 for example) only allows 23 dBm whereas higher ones like 157 allow 27 dBm.
I regularly download games from steam that manage to saturate my connection. I have a relatively small SSD, and play many different games, so I regularly uninstall and redownload games.
It's not just about the speed though. The fact that the router is the bottleneck, means that there is also less power left on the router to do other interesting stuff like Adblocking, dnssec validation, etc. There's nothing wrong with wanting to optimize what you have. Even if it's just for the sake of tinkering, since tinkering itself can be fun
Is there a way to find out whether:
This is an issue with my specific SOC, eg I need to return the device
This is a general issue with the SOC, eg everyone will run into this issue, and it probably cannot be solved with a software update.
This is a general issue with the software, eg everyone will run into this issue, but it probably can be solved with a software update.
For some reason, PPPoE has a massive CPU overhead in OpenWRT (even though it's a relatively simple protocol). With no software flow offload (which is a requirement for SQM to even do anything), I cap out at 600-700 MBit. That is WITHOUT sqm.
PPPoE requires to diassamble all received packets and reassembly with changed header fields and the PPPoE header removed from the ethernet payload. That requires deeper packet processing than plain forwarding.
My son upgrades his games (menay GB) from steam and I have only 300 MBps.
When he is downloading we not realized of it because downloading speed from steam here is quite low, too low in fact it takes quite long to download a game.
It surprised me that you could download games at speedn near 1Gbps.
I thinks steam uses downloading from peers, I am not sure, and that may be the reason downloading speeds can vary a lot from region to region.
Or may be they do not distribute their servers well across the globe.
I have substituted the ISP functions by my RT3200.
The ISP router now acts simply as a bridge (well it does telephony too) and fowards all data traffic to my RT3200 that is configured with PPoE in its wan interface.
There is the possibility to use the ISP router to connect PPoE in its external interface and configure the RT3200 with a fix IP in its WAN interface and use DMZ to pass all data traffic to it.
Would it be a better aproach and let RT3200 to be less busy?
My RT3200 does not seem to be busy right now, with 300 MBps it seems to be quite free, but if it is a better solution, meanwhile I continue using the ISP router and don't eliminate it...
Enabling packet_steering and installing & enabling irqbalance did the job Same speeds as previously on hardware flow offload, but now purely on software flow offload This device really is a beast!
I can confirm too that hw offloading is crashing the router after a while. But the good news is AFAIK there is active development here including LAN-WLAN offloading which currently takes a big hit on the CPU.
Are there any known failures of "packet_steering"? Just asking as hw-offloading tooks some days to find the source of the problem. (And if there are none: why is only 1 cpu used by default?)
