I have what I would consider an anomaly in my setup and am curious if there is a way to take advantage of it (DL-WRX36)

I just set up a new router (DL-WRX36) with OpenWrt and SQM for my internet. I have 500/20 speeds from Spectrum (cable/DOCSIS) and my ping to 8.8.8.8 is around 30ms. I play Valorant and during peak hours my ping is usually 40-50 to my closest server (Georgia) and my absolute best ping in game can sometimes drop to below 30, barely, about 28-29ms late at night.

Before SQM my bufferbloat tests varied drastically from A to F depending on time of day, and sometimes would go from A to F in a matter of minutes. After SQM I consistently get A+. I've been running the tests while in game on Valorant and notice that during download tests my jitter increases slightly with small spikes which don't appear on the bufferbloat test, however, during upload tests my jitter almost goes flat. It reduces my jitter to almost consistently under 1ms and reduces my in game ping to around 35ms from 40-50. After the test ends, my ping quickly creeps back to 40ms+ again.

Like I said, I would consider this an anomaly but maybe this isn't so uncommon. My real question is, how could I trick my router into outputting this type of lower jitter and latency consistently? Is there a certain way that the Waveform bufferbloat test is sending packets that causes this, or the type of packet? I'm obviously a noob but I'm intrigued.

That seems... suboptimal. At full chat the upstream would likely be saturated with ACK packets, causing a bottleneck. Any chance your ISP can increase your upstream to a sensible value? At an absolute rock-bottom bare minimum for serviceability you should be looking at 50 upstream, preferably more.

Unfortunately not, unless I bump it up to a gig, then I think I’d get closer to 50. I just not need it unfortunately and it’s not cheap, I think it’s $80 for what I have right now.

Duplicate:

1 Like

3.1 or 3.0 ? The 3.1 technology has several changes that address latency.

1 Like

I’m 90% certain it’s 3.1

1 Like

This thread appears to be really similar to another one by the same OP found here.

Since they are so similar and can result in bifurcating the discussion, I will be closing this thread. If this should be re-opened, please feel free to send me a PM with a quick explanation about how these are two are different.

1 Like