Anyone have experience running Mullvad on an AX1800 Flint router?
Anyone have experience running Mullvad on an AX1800 Flint router?
We’ve recently got a decent fibre gigabit connection, so after a lot of reading I bought a gl.i ax1800 Flint router to run Mullvad on, rather than having it on every device individually. The setup was simple, and while I know that it will impact speed to an extent, it seems to be really limiting the connection. Like to about 13mbps, which is slower than my almost non-existent 4g connection. From what I’ve read, I should expect the flint to be capable of giving me about 4/5ths of my connection speed, so on ours I’d been expecting something in the mid-hundreds, as I’m seeing 800mbps or so just on the cat 7 from the modem. It’s almost certainly my fault, or something I’ve overlooked in the setup, so any help or pointers would be much appreciated.
VPN can be computationally heavy, I'm not really surprised those are the speeds you're getting from it.
With that said, adjusting your MTU value, usually lower, to help reduce fragmenttation, can usually claw back some throughput. You'll have to play with it a little to find optimum values. Use PING with the df (do not fragment) bit set to sus out what works best for your setup.
Hope that helps!
Thanks for the reply. It’s a router designed specifically to handle the loads of vpn usage with a fairly quick dual-core processor, and looking a various people’s speed tests, they seed to be miffed if they lose 25-40% of their bandwidth. I’m loosing about 90%, so I doubt it’s the router’s hardware’s fault. None of the tests I could find were seeing speeds of less than 100mbps, from a 275mbps connection for instance. Most were showing up to 800mbps with a vpn running on Wireguard from a gigabit connection.
I’ll have a look at the MTU value when I’m at home, and see if that gets me anywhere.
I am actually somewhat familiar with the hardware, I have an older (2 years?) version that I use for wireguard when in hotels, etc. I get about 40% of their rated 'wireguard' throughput with it, which is where my assumptions came in about performance. I actually switched to an old android phone with a USBC to Ethernet and a local hotspot for this purpose instead. Glad to hear the newer gear is more inline with their estimates, though. Might be worth swapping mine out.