I am currently using IPFire on a Raspbery pi 3 and using QOS
Today I played a game and I felt latency in my game but I had a good ping so I tried to do packet loose test and I figured out that behind ipfire I loose performance and I don’t know why. My connection can go up to 11MBPS or 12 MBPS max.
QOS config :
|Downlink speed (kbit/sec):|10670||
|Uplink speed (kbit/sec):|533|
My configuration for default qos (I changed settings for gaming because it is random port):
Paquet loose constated behind my router :
Packet loose behind the IPFire Raspbery pi 3 :
I tried too without QOS but that the same. So I don’t really know from where the problem come. On my side I keep looking for find from where that could come. I have to mention to that everything is passing in vlan in a TP-link Switch who isn’t configured with QOS. I have vlan on Red Zone and Green Zone.
If somebody already face to this problem and know more or less from where it is coming I am answering for your help and I will take any help available. Thank you of taking time to look my post guys. Have a nice day !
Speed issues can be due to bandwidth allocation but also to CPU bottle neck. You should check if the CPU cores are maxed out while you play the game. Especially if the CPU needs to take care of the TCP-IP traffic as well, like in a RPI.
Thank for you Answer but I don’t think the problem come from my computer before I installed IPFire I’ve never meet this problem on my computer and I always have a 170 FPS stable on my game I got a Ryzen 7 2700X an RX 5700 XT and 16gb RAM so few possibilitty that is a bottleneck and the bandwith I tried the packet loose test without QOS for hope of a better result and that the same but in the doubt I will make other test in my game with the stats of CPU and RAM. Like the screen I showed behind the router there is no packet loose and behind ipfire I loose a lot. I saw the acceptable packet loose is 2 % and right now I am at 9% of loose.
Just to be sure, we are talking about the RPI running IPFIre and directing the traffic in and out your network, not the machine you are using behind IPFire. My hypothesis is that the RPI cpu can’t cope with that traffic. The best way to check this hypothesis is to open a console access to IPFire and run the program
top while playing the game. If one of the 4 cores is maxed out, this would be consistent with my hypothesis. You can also check the web user interface and go to status/system but then you need to find in the graph the time frame when you were playing the game.