Most reliable, best supported router for OpenWrt?

I have an x86 based router with a Compex Ath10k wifi card, and it works just perfectly (performance + coverage). Maybe you had a bad experience.
Nevertheless it's a single band wifi, so I have to choose between 2.4 and 5GHz, while a classical router/AP have both bands. That may be an argument for using a dedicated AP anyway.
And last not least ... I finally unclip the wifi card and only use the AP. It's only a choice of energy saving. The wifi card was heating 24/7 while rarely used. So I only power on the AP when I need it.

I also had some ugly crashes with an Intel card back in the 4.4 kernel days. I really hope it's fixed now 25+ cycles later as I haven't used one since.

You should get a Pi-router supported by OpenWRT and then just add a good Wireless AP to it.

Problem is, it's a lot more expensive, assuming you can actually get your hands on one, it think they're still in shortage.

NanoPi routers are still easy to get and not very expensive.

Interestingly, as more devices have come up on my network, both wired and wireless, and I installed and configured more features including relative CPU intensive ones (SQM, QoD, mwan3, adblock, snmp for monitoring), I haven't had a crash in nearly a week so far.

I am starting wonder if the reason for the various random crashes I have been experiencing is a bug in power saving on this SoC. My current theory is that when the SoC goes down to a certain level of deep sleep, sometimes it doesn't wake up reliably and instead hangs and crashes.

Does anyone know if there is a kernel boot or /sys or /proc parameter that can be tweaked to disable CPU power saving? I already tries disabling WiFi power saving and that didn't make any difference before, but since there has been a little bit more load on the CPU, it has been suspiciously solid, presumably because it no longer goes into as deep a sleep as frequently.

There is, but it might be platform dependant.

So going for a HP T620 Thin Client (x86-64) vs a "Pi Router": NanoPC-T6 (ARM) vs Asus TUF AX6000 (MT7986). Is there really a big difference? I also want a very stable network, handle more than 4000+ open connections and high traffic.

Ideally 2.5GB or higher on the WAN + LAN. My requirements aside, let's focus on the core question again;
would a x86-64 vs ARM vs MT7986 (and the mentioned models above) indeed be that much different in terms of stability on high throughput/traffic, 24/7 up-time, large nr. of connections, etc.?

If you say yes.. I really should go for a NanoPC-T6 instead of the Asus TUF AX6000 and buy a separate WiFi 6 AP, please let me know. Or even go for this HP system...?

All non x86 routers requiring more specific linux support to make them work nicely, once they are in kernel it should be working well (the former Nano Pi R4S with RK3399 is good example, rock solid as dual 1G ports router).

Currently RK3588 (the NanoPC-T6, I own NanoPi R6S which is almost the same) isn't upstream to LTS kernel so no official OpenWrt support (however some custom build right now showing quite good performance), the MT7986A just got it's support however the 4 x A53 is definitely a lot behind the RK3588 (4 x A76 + 4 x A55), while the upcoming MT7988 has 4 x A73 which performs a lot better (but no upstream kernel support yet). And BTW the TUF AX6000 is not getting official support yet, why would you consider if you really need stability?

incorrect - supported in snapshots

For me anything not in stable release version won't be counted as "official support".
Anything can happen before they finally coming into release, if I remember correctly the IPQ807x target was introduced quite long time ago however it was dropped, later probably issue fixed then it came back and now getting support again.

Do you have git reffrence of this happening?
As for officiall support as long as it's in git it's officially supported

But device ASUS TUF-AX4200 has been in the stable branch for a long time:

Both devices TUF-AX4200 and TUF-AX6000 are as maximum identical.

And both devices have errors in the repository that no one pays attention to:

  1. https://github.com/openwrt/openwrt/pull/13422#issuecomment-1780773692
  2. https://github.com/openwrt/openwrt/pull/13125#issuecomment-1829308456
1 Like

Looks like this bug limits the maximum image size to 16 MiB. Kinda bad but not something that breaks the default image. Did you post the patch or do you want me to do it?

First of all thank you all for your fast replies.

As remittor mentioned already it's indeed part of snapshot. Despite that, the TUF AX4200 is already in the stable branch, uses a completely similar platform and therefore the firmware for router TUF AX6000 can be considered stable as well. I think it will be part of the official OpenWrt 24.x release.
Read this thread.

MT7986A is WLAN, this thread mainly focuses on performance on LAN/WAN ports. Apart from the errors that remittor mentioned above, I think we can say this device runs stable. Not sure about the performance difference between NanoPC-T6 vs TUF AX6000. In my case my router is before a busy server, so it's not only about speed but also about being able to handle the large amount of parallel TCP/UDP connections.

EDIT: for people who are interested in MT7988A chip (which is the ARM core chip not the WLAN chip) see: Banana Pi BPI-R4. Granted Banana Pi BPI-R4 looks also very promising (too bad I just noticed that it only has 1GbE on WAN port).

I would like to see benchmarks tests between the Asus TUF AX6000 vs NanoPC-T6 vs Banana Pi BPI-R4 with various loads all using OpenWrt.

EDIT: EDIT: https://openwrt.org/inbox/toh/sinovoip/bananapi_bpi-r4 Says "Not supported"!? That is not good..

That might soon (fingers crossed) change

While that is technically correct, it doesn't really apply either.

Yes, blogic originally did introduce a development target for ipq807x - but this target was never built, nor did it support any actual device at all (maybe just enough to boot the Qualcomm evaluation board HK01, maybe…). This was indeed dropped again, as it didn't support any device and wasn't moving anywhere - while it was causing quite some pointless rebasing work for the actual development then taken up elsewhere.

Ansuel and robimarko then did the heavy lifting of getting ipq807x actually functional, including half a dozen of supported devices (and meanwhile this list has grown), and merged (both mainline and for OpenWrt).

A no point was device support dropped, not even temporarily, for the simple reason because the initial ipq807x target didn't support any.

1 Like

If your primary interest is ethernet performance but not WiFi, then the best should be x86 based one. Also NanoPC-T6 RK3588 is a lot faster than the TUF AX6000.

BPi R4 has the following ports, I do not see anywhere saying it has only 1G WAN:

  • 2x 10Gbe SFP slot (option 1x 10Gbe SFP and 1x SOC embedded 2.5Gbe PHY. NOTE: Need to modify hardware)
  • 4x Gbe network port

This platform is too new.....you'll have to wait, but Mediatek seems to work with the community actively so I think it should be coming.

Also NanoPC-T6 RK3588 is a lot faster than the TUF AX6000.

Are there any benchmarks I can watch and see?

I do not see anywhere saying it has only 1G WAN
4x Gbe network port

So that is one of those GbE ports can be used for WAN. Hence 1G wan.

Don't know why you insist to see the numbers when there is a big difference in CPU family.

But you can check out this post to find the MT7986 RSA speed, and go into the table to find the RK3399, you'll see RK3399 is already a lot faster than MT7986 (single core), and RK3588 is a more advanced one with even more core when compared with RK3399.

You can define which port to be used with WAN, there are 2 x 10G SFP+ ports, so you can have 10G WAN no problem.

1 Like