BE98 Latency

[jixiangyuan]

Hi all, I’ve noticed that the latency of my BE98 is consistently higher than my previous AXE11000. With my AXE11000, the ping is usually at 7ms to the nearest server at both 5GHz and 6GHz. But with the BE98, I get 7-8ms on 6GHz, 8ms on 5G-1, and 9ms on 5G-2 most of the time with the same server. So, I was wondering if it’s just me, or if it’s because of the firmware, or if it has something to do with the hardware design (I heard that dual 5GHz will always be worse than single 5GHz due to less FEM per band as well as antenna interference).

 

[TheLostSwede]

Sorry, but you worry about 1 ms ping difference over WiFi?
Here, try this test, it's much better than some kind of ping test and matters a lot more.

Bufferbloat and Internet Speed Test - Waveform

Is bufferbloat causing issues with your internet connection? Run this test to find out.

www.waveform.com

 

[jixiangyuan]

Sorry, but you worry about 1 ms ping difference over WiFi?
Here, try this test, it's much better than some kind of ping test and matters a lot more.

Bufferbloat and Internet Speed Test - Waveform

Is bufferbloat causing issues with your internet connection? Run this test to find out.

www.waveform.com

Well, from 7ms to 9ms, that's a 28.6% increase in latency. High throughput and low latency is what we wanted, isn't it? Otherwise why won't we just stick with the free router from the carrier. I'm getting similar results from the website you provided. Not sure if that's more accurate than the latency I got using Speedtest by Ookla though, as this website doesn't disclose the server it use so you might be connected to a different server when switched to another band.

 

[TheLostSwede]

Loaded latency is what matters, so very different test in that sense.
Also, WiFi can behave differently just based on the weather that day or if you placed the new router differently compared to the old one.
I get 5-6 ms on my phone with ookla, but it means nothing in real world terms.

 

[Tech9]

but it means nothing in real world terms.

There was a person here not long ago chasing low latency on a phone because it makes a difference in financial transactions...

So you never know... some superhumans may have developed 20ms reaction time and 120Hz vision refresh rate competing with cats.

 

[L&LD]

The latency on tech hardware doesn't directly relate/compare with the differences that humans can sense. Regardless of the 'theory'.

I have customers who can sense smaller differences than me and others who don't see gross differences in network (and computer) performance. People vary, and averages don't reflect those outliers.

I've seen some very slow cats too.

 

[sfx2000]

If one is concerned about latency/consistency - don't do WiFi - stay on the wire...

 

[jixiangyuan]

Loaded latency is what matters, so very different test in that sense.
Also, WiFi can behave differently just based on the weather that day or if you placed the new router differently compared to the old one.
I get 5-6 ms on my phone with ookla, but it means nothing in real world terms.

Wait, weather could affect in door APs? That doesn't sounds right. Is there any evidence to support this?

 

[jixiangyuan]

If one is concerned about latency/consistency - don't do WiFi - stay on the wire...

Well, you know, not everything can stay on the wire, phones for example (or it doesn't really make sense - i.e. Ethernet to USB converter) .

 

[TheLostSwede]

Wait, weather could affect in door APs? That doesn't sounds right. Is there any evidence to support this?

Try testing during a thunderstorm and you'll understand.

 

[jixiangyuan]

Try testing during a thunderstorm and you'll understand.

Very scientific, I love it

 

[TheLostSwede]

Very scientific, I love it

Obviously most weather doesn't affect your WiFi, but sometimes it's the little things, like an old microwave oven or even your neighbours WiFi that could cause interference. As such, you can't expect the same exact performance all the time on WiFi.

 

[jixiangyuan]

Well, it just so happens that I’m living in a semi-remote place, where there is no interference from neighbours or microwaves whatsoever, making it the perfect playground for testing latency.

 

[Matthew L]

@jixiangyuan You might have some services running that may not have been enabled on your Axe11000.

You may want to check to see if Traffic Analyzer and AiProtection are running as both would add some latency.

 

[jixiangyuan]

@jixiangyuan You might have some services running that may not have been enabled on your Axe11000.

You may want to check to see if Traffic Analyzer and AiProtection are running as both would add some latency.

View attachment 58042View attachment 58041

Thank you for your reply. I’ve just checked my router settings, and it appears that all those settings are off by default. I still think this is hardware-related, as the latency of 5GHz-2 is significantly higher than that of 5GHz-1; they should have the same increase in latency if it’s software-related. Maybe it’s just due to fewer FEMs per band or internal antenna interference? The best way to find out is for me to buy the BE96U, as it’s a tri-band router. Or it might just be firmware-related. Let's hope ASUS can push another firmware update soon.

 

[jixiangyuan]

To make this even funnier, this is the lantency when using the free router comes with the fiber plan.

 

[jixiangyuan]

And this is what I get with my £699 BE98: