Jump to content

congestion control


Vroomfondel1965

Recommended Posts

hi guys,

I've had my Netduma for a couple of years now so I'm not a total noob but........I've added a linksys router for 5Ghz wi fi but when I adjust the sliders on cc the speed to the linksys also drops, Is there a setting I'm missing as I thought it would have greater access to speed when the sliders are lowered for my Xbox. I have share excess ticked also.........thanks in advance for any replies.

Link to comment
Share on other sites

That is correct CC will give headroom to all attached devices including the Linksys so when you game network usage by other people will not create jitter spikes.

 

If you want more bandwidth when you are not game then set the sliders to 100%

Link to comment
Share on other sites

That is correct you lower the bandwidth so you are leaving headroom so the others using the bandwidth do not create jitter.

 

It will lower the bandwidth on the connected Linksys AP.

 

This is what the congestion control is for, control congestion so other users in your network do not make you lag.

 

For me this is indispensable as a gamer on a utilized network

Link to comment
Share on other sites

Ok, thanks for your patience. Looking at options on how to run my "network", modem mode to duma then linksys will lower bandwidth to the rest of the house because of above. Am running duma and linksys side by side which means any downloading affects my gaming. I've tried modem mode to linksys to duma which I think affected gaming and meant duma only operated host filtering and cc......any suggestions welcome.

Link to comment
Share on other sites

Modem > Netduma > Linksys AP is the best way to do this.

 

Yes this will lower the bandwidth but this is a good thing while you game 70% each way using reactive is the sweet spot for myself other users use other percentages.

 

I leave my sliders at 70% all the time so my web pages open fast even if my line would be saturating as my son does a download.

 

Others like to switch back to 100% while they are not gaming.

Link to comment
Share on other sites

I know and I have tried this set up before but to try and get a good bufferbloat test I'm reducing download cap to 10% / 20Mbps.... better if I go lower. I'm just concerned that the rest of my network only has that to share. I thought if my Xbox is down to 10% then the rest would share remaining 90%.

Link to comment
Share on other sites

I would reset the bandwidth flower, tick share excess and try 70% each way reactive and then run a test.

 

I use ping plotter to test for jitter as I find DSLreport's not to be that accurate.

Link to comment
Share on other sites

Thank you both for your time. Have spent/wasted a heck of a lot of time on bufferbloat then....d'oh!!..... will have to research ping plotter now, I have it but not sure what I should be seeing..... thanks again.

Link to comment
Share on other sites

Ping twitter.com using ping plotter run for around 10 minutes and post your results and we can let you know.

 

On ping plotter File, save image then upload the picture

Link to comment
Share on other sites

Bufferbloat only occurs when your connection is saturated. Use 70/70 anti-flood as Zennon suggested and that won't happen. Do not worry about trying to get a perfect score on a test because it doesn't matter.

My CC is 30% download and 99% upload and I get some really good bufferbloat scores.  If you are saying that it does not really matter the scores I'd rather leave it at 70/70 if I will get the best and most consistent results there.  I too like our friend xAngryNeilx have spent countless hours and days on bufferbloat.

Link to comment
Share on other sites

I would reset the bandwidth flower, tick share excess and try 70% each way reactive and then run a test.

 

I use ping plotter to test for jitter as I find DSLreport's not to be that accurate.

Any time of day is a good time to do Ping Plotter test?

Link to comment
Share on other sites

  • Administrators

My CC is 30% download and 99% upload and I get some really good bufferbloat scores.  If you are saying that it does not really matter the scores I'd rather leave it at 70/70 if I will get the best and most consistent results there.  I too like our friend xAngryNeilx have spent countless hours and days on bufferbloat.

 

The scores don't matter, it's whether you don't experience any local congestion and that devices have enough bandwidth as 30% is quite low and likewise 99% upload is very high. It doesn't have to be 70/70. If you REALLY wanted to be methodical about it I would have a PingPlotter running and really hammer your connection - downloads, streams etc. Then try at different percentages to see which brings you the closest back to your pre downloading/streaming ping and what is most consistent. Obviously that'd take awhile though.

 

Any time of day is a good time to do Ping Plotter test?

 

Generally wouldn't do it at peak times because your ping/stability is more than likely going to be worse & you could perceive that as a problem with your setup.

 

1509471100-U10165.png

OK, sliders at 70/70.....pre emptive......network not being used by anything else

 

That was the conditions for the above test? Seems a little inconsistent if that is the case. Not terrible but would expect better if congestion control was working ideally.

Link to comment
Share on other sites

The scores don't matter, it's whether you don't experience any local congestion and that devices have enough bandwidth as 30% is quite low and likewise 99% upload is very high. It doesn't have to be 70/70. If you REALLY wanted to be methodical about it I would have a PingPlotter running and really hammer your connection - downloads, streams etc. Then try at different percentages to see which brings you the closest back to your pre downloading/streaming ping and what is most consistent. Obviously that'd take awhile though.

 

 

Generally wouldn't do it at peak times because your ping/stability is more than likely going to be worse & you could perceive that as a problem with your setup.

 

 

That was the conditions for the above test? Seems a little inconsistent if that is the case. Not terrible but would expect better if congestion control was working ideally.

Can someone please translate the 2 images below for me.  I pinged google.com and twitter.com while playing BF1 on PS4 and CC is at 70/70 reactive.  

 

1509673865-U6308.jpg1509673916-U6308.jpg

Any help is greatly appreciated.

Link to comment
Share on other sites

  • Administrators

Well the Google server itself can misbehave so I would account the mini spikes to that although it's pretty good overall.

 

Twitter looks fairly consistent, I pinged the exact IP myself and that server is located in the UK (I get 4-7ms) so that is why it is higher than you might expect.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...