3 ms·
Speed is not the only factor. In fact, most people thought their Wi-Fi was fine because for non-real-time usages, TCP will conceal occasional dropouts. Your in-
by Nextgrid 5y ago
Speed is not the only factor. In fact, most people thought their Wi-Fi was fine because for non-real-time usages, TCP will conceal occasional dropouts. Your in-browser test might be showing 500Mbps while what's really happening is +500Mbps in between bursts of nothing. Applications such as iperf3 would be able to tell you more (as they can report in much quicker intervals, allowing you to see spikes that would be averaged out) but a quick test is to just ping your gateway and move around in any area you will potentially use your device (even your orientation matters; if your body is blocking the line of the signal it could have a disastrous effect on an already-weak connection) and see if you notice any spikes. If not: congrats, if yes: you have shit Wi-Fi that TCP has been successfully concealing but the illusion will break down when you use a real-time application where retransmissions can't work (because they would be too late).
- Aeolun 5y agoAh, the good part is probably that the connected machines are not going to move very much, nor is the AP (e.g. the connections between the AP and machines are all cables). I simply don’t have a cable pulled from the top floor (where my fiber enters the house) to the bottom floor where my office is. I’ve played realtime games over it, and haven’t had too much issues, so I guess it’s working well enough for my use case. It’s possible the actual speed is higher though, and the occasional bursts of nothing are still there (the link to my home is 2Gbit).