Bufferbloat question

[Welshy]

I've been setting the bufferbloat up on PC and checking via dslreports and if I'm lucky I can get a A rating and got to be honest the results can be hit and miss.  

 

Last night I thought I check the same PC bufferbloat settings which gave me a A rating via the explorer on the console and the rating was C so adjusted on PC and checked on console until l got as close as I could to A.  Now the difference was night and day and this time other people were rage quitting lol.

 

So the question is, should we be using the explorer on the consoles to check the bufferbloat as that is what it impacts?

 

Also, I can't use the same settings for PS4 and xbox (both checked on the console explorer) with the xbox being the hardest one to get close to the A.  Any ideas why?

[Netduma Fraser]

Well when you do a bufferbloat test it is doing it for that specific device not for your entire network which is why its different when you check it on the consoles. Bufferbloat happens when your connection becomes saturated with 70% anti-flood it does not get a chance to become saturated.

[Welshy]

I wish 70% worked. On the PS4 I've had to wind it down to about 50% to get A and xbox seems to be very inconsistent.  Actually on the xbox I've had to use roughly about 480kbps per device, share excess unticked and on 100% to get in the B region.  Any suggestions? 

[Netduma Fraser]

As I said, bufferbloat isn't an issue until your connection becomes saturated but 70/70 never lets it get saturated so bufferbloat isn't an issue. I don't really have any suggestions for you in regards to bufferbloat unfortunately, there are lots of people who do the tests though so I'm sure they'll comment with advice

[e38BimmerFN]

Also modems and ISP cabling up to modems can play a roll in signals and line quality. Any bad splitters on the line isn't recommended.

If your the only one gaming at night, like me and nobody else using bandwidth, I keep my congestion controls at 100/99. I get A grades across the board with my 100/3 ISP service using a Arris SB6180/6183 cable modems.

 

Also some test servers that dslreports uses maybe busier then others so use different test server sites near you.

[ColonicBoom]

Ion the xbox I've had to use roughly about 480kbps per device, share excess unticked and on 100% to get in the B region.  Any suggestions? 

 

 

It could be down to the crappy browser on the Xbox.

 

I don't bother with dslreports anymore, I find it can give erratic results that will change daily, hourly, even after only a few minutes at times. I also found that it would nearly always give me A+ across the board with my previous ISP even though games were completely unplayable on that line due to horrible lag from the ISP.

 

When I want to test for bufferbloat...

Set your sliders at 100/100 and run a speedtest from a computer over ethernet. You should see some big spikes on the graph.

Don't stop PP running (for comparison), move your sliders down and run the speedtest again.

If you don't see the difference on the graph then adjust your sliders down further.

Once you see the spikes are being suppressed, leave them there whilst you play.

That's all you need to do as far as buffer-bloat goes really.

 

Here's a link to a thread about it...

http://forum.netduma.com/topic/19746-does-congestion-control-work/

[XSXS]

My pc usually scores better than using MS edge via the xbox dashboard. I even used the same network cabling and everything. My pc A+, A, A+. Xbox edge A, buffer bloat A or B and A. I rarely see any A+ ratings on the console where my PC usually scores 2 out of 3 A+. I think the browser on the xbox is junk.

[fuzzy clam]

It could be down to the crappy browser on the Xbox.

 

I don't bother with dslreports anymore, I find it can give erratic results that will change daily, hourly, even after only a few minutes at times. I also found that it would nearly always give me A+ across the board with my previous ISP even though games were completely unplayable on that line due to horrible lag from the ISP.

 

When I want to test for bufferbloat...

Set your sliders at 100/100 and run a speedtest from a computer over ethernet. You should see some big spikes on the graph.

Don't stop PP running (for comparison), move your sliders down and run the speedtest again.

If you don't see the difference on the graph then adjust your sliders down further.

Once you see the spikes are being suppressed, leave them there whilst you play.

That's all you need to do as far as buffer-bloat goes really.

 

Here's a link to a thread about it...

http://forum.netduma.com/topic/19746-does-congestion-control-work/

I think that's the mistake people make when they test for BB,they don't put any load on their connection (aside from the DSL reports test),I found a huge difference running PP,and Netflix or YT while checking the BB,the settings weren't even close to being the same.

[ColonicBoom]

I think that's the mistake people make when they test for BB,they don't put any load on their connection (aside from the DSL reports test),I found a huge difference running PP,and Netflix or YT while checking the BB,the settings weren't even close to being the same.

 

Same, it was miles off! And led to a lot of unnecessary tweaking and stressing.

And as it's people that are streaming in the house that are going to cause the congestion, it seems like a no-brainer to test that way.

 

I can't take credit for this though (although I will).

The force is strong with this method 

[fuzzy clam]

Yeah I love Yoda always a few steps ahead of the game..

 

And your exactly right it's the load on your connection that's gonna cause the issues, so why not stress it as that's what will actually happen on most home networks.