Jump to content

Bufferbloat


Lyncho84
 Share

Recommended Posts

Can anyone advise me on how to achieve better results on dsl reports when doing tests?iv tried re-active and pre-emptive on different cc slider settings but I alway spike into the red on bufferbloat meter which acco ding to videos I've watched is a no no.my isp is bt infinity 52down 10up package

Link to comment
Share on other sites

Seems odd. I had problems with congestion on Virgin Media - I could never get an A grade regardless of much I reduced my bandwidth - yet with BT Infinity I can use 99% reactive sliders and average under 3ms of bloat.

 

Might there be something wrong with your cables? What do network diagnostic or PingPlotter tests look like for you with nobody else using the connection?

Link to comment
Share on other sites

the thing is with dslreports you are relying on their servers and you are running it through a web broswer. you would be wise to install pinglotter and see how your line is performing and wether there is jitter on your line. you could have bad cat cable, faulty isp router or may need a new socket faceplate.

Link to comment
Share on other sites

the thing is with dslreports you are relying on their servers and you are running it through a web broswer.

Forgot to mention this. The jitter test can show 0.2ms PDV one day and 20ms the next, while pinging a server further out (3400 miles away in New York for example) can show me 70ms latency with 0.2ms PDV because it's not taking a dump.

 

PingPlotter is a little better as you have the choice of pinging whatever you want; if something is showing a ton of packet loss or is down, you can try something else near you. I tend to avoid the usual 8.8.8.8 and ping my local (UK) IW server and it's consistently good. Shame I can't say the same for gameplay lol

Link to comment
Share on other sites

Ping Plotter IMO gives you the best and most accurate results. I test on DSLReports and it gives me different ping results, in my case 20ms to 60ms difference in my ping time, whilst PP is only from 10ms to 15ms. Plus I ping Twitter for the most accurate and consistent test.

 

I only use DSLReports to check my speeds.

Link to comment
Share on other sites

Yeah the DSLreports.com is more of a general consensus of how your connection is. definitely do as suggested with ping plotter. Although, if you do want to keep using that website, you should try to have both the lines as flat as possible. You should also try to make sure that there are no red bars and everything is green. As well as, try to have the excess go no higher than 10. If you check all of these boxes, you should have a+ each time almost.

Link to comment
Share on other sites

Another thing I should add as I've just noticed this today: make sure you take into consideration all elements of the connection equation, so to speak.

 

I used to think preemptive was best 100% of the time until I realised that reactive gave me less jitter if there was background activity on my line, like a download in progress.

 

However I just ran a ton of tests, using many different slider combinations with preemptive vs reactive, and I discovered that reactive always gives me packet loss. It varies from 0.2-8%, but I never got 0% after around 50 tests. When using preemptive I had 0% packet loss after the same number of tests, regardless of whether I had 100% sliders or as low as 15% (seemed silly going lower). I'm on 50/10 BT Infinity though; your mileage may vary.

 

If any of the wizards around here could possibly explain why reactive always causes packet loss and preemptive never drops packets, I'd be very grateful

Link to comment
Share on other sites

If any of the wizards around here could possibly explain why reactive always causes packet loss and preemptive never drops packets, I'd be very grateful I never see anything but exceptional for packet loss on the Netduma diagnostic test though. Does exceptional cover a range such as <1% like DSLReports?

 

What's showing packet loss? Your Xbox/PS4 or dsl reports?

 

 

My understanding of it (not sure if I'm right)...

 

Hyperlane is there to safeguard your gaming data above ALL other data going through your line. In the event of a flood of data the Netduma can drop non-gaming packets to ensure that the gaming data goes in/out when it's supposed to.

 

When your line gets flooded by the bufferbloat test, the Netduma notices the rapid increase in throughput and 'reactive' decides that the data is not priority and can drop some packets in an effort to prevent the flood.

Because 'pre-emptive' is more aggressive, and is limiting your bandwidth to a greater extent, it's already prepared for the influx and so doesn't necessarily need to drop those packets during the bufferbloat test (but you'll have less bandwidth).

 

Either way, if your console is in hyperlane then it shouldn't matter because the packets getting dropped are not your bullets.

 

Might be wrong though.

Link to comment
Share on other sites

What's showing packet loss? Your Xbox/PS4 or dsl reports?

 

 

My understanding of it (not sure if I'm right)...

 

Hyperlane is there to safeguard your gaming data above ALL other data going through your line. In the event of a flood of data the Netduma can drop non-gaming packets to ensure that the gaming data goes in/out when it's supposed to.

 

When your line gets flooded by the bufferbloat test, the Netduma notices the rapid increase in throughput and 'reactive' decides that the data is not priority and can drop some packets in an effort to prevent the flood.

Because 'pre-emptive' is more aggressive, and is limiting your bandwidth to a greater extent, it's already prepared for the influx and so doesn't necessarily need to drop those packets during the bufferbloat test (but you'll have less bandwidth).

 

Either way, if your console is in hyperlane then it shouldn't matter because the packets getting dropped are not your bullets.

 

Might be wrong though.

DSLReports shows packet loss if I use reactive mate. It varies but it's never 0% on reactive, while preemptive isn't capable of causing packet loss at all apparently lol

 

I think I had PSN in the hyperlane while testing and it never made a difference on reactive so idk...

Link to comment
Share on other sites

I think those blufferbloat test are bs I get a A+ across the broad and games still run like crap

LOL yeah, maybe you need a D grade while streaming to be one of those unkillables like the majority of connection-ignorant streamers who can just plug and play on a crap line and melt like they're terminators

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...