Jump to content

Reactive vs pre


Recommended Posts

Posted

I find reactive works best for me as do many others. Other people find preemptive works better for them, it’s really down to experimenting with the algorithms and finding what works best for you.

Posted

So ww2 is having issues but with my battlefield reactive has a better experience. You know why vid games don’t play good with reactive?

As far as I can tell, reactive on newer firmware causes a small amount of packet loss

Posted

Am on fibre A+ bufferfloat ping 7-8 ms. Very stable line. I'm get slaughtered too. Think their code really punishes the low ping high speed users.

 

Also same happens on BF1 where I am usually the lowest ping (typically 30-40 ms to next lowest ping player).

 

Don't assume games will be better magically with fibre.

Duma can't help in these situations either.

Posted

Me too. Am on Fibre optic and after the dedicated servers gone back up i have the toughest time in COD history. Bullets don't register. I don't know what to do any more

Posted

I'm in the UK and Duma wasn't putting me on dedi's. I blocked the only one it found in Belgium which was unplayable and thereafter only found P2P games.

  • Netduma Staff
Posted

I'm in the UK and Duma wasn't putting me on dedi's. I blocked the only one it found in Belgium which was unplayable and thereafter only found P2P games.

 

So you temp-banned a dedicated server in Belgium which led to you only connecting to P2P games? Did this ever change, i.e when the temp ban ran out did you connect again to the Belgium server?

Posted

So you temp-banned a dedicated server in Belgium which led to you only connecting to P2P games? Did this ever change, i.e when the temp ban ran out did you connect again to the Belgium server?

Posted

Stayed on P2P as I never saw Temp Ban button again only the Update button which I believe means P2P

Posted

I was looking at the statistical data at the bottom of dslreports where it says something like lost frames and lost ms or words to that effect

Posted

I was looking at the statistical data at the bottom of dslreports where it says something like lost frames and lost ms or words to that effect

Honestly, I would take anything that dslreports says with a pinch of salt. The best way to test your line quality is by downloading PingPlotter and pinging twitter.com. It’s a far better indication of your line quality than dslreports.

Posted

Honestly, I would take anything that dslreports says with a pinch of salt. The best way to test your line quality is by downloading PingPlotter and pinging twitter.com. It’s a far better indication of your line quality than dslreports.

It happens to me too. 100% reactive is fine but anything else causes me to have anything but a 0% packet re-xmit stat. Unticking share excess does the same too. I get 0% with any combination of sliders on preemptive.

 

However this doesn't happen on firmware 1.03.4, which I was using for quite a while thanks to the 1.03.6 GUI slowdown when using port forwarding before UPnP started opening port UDP 3074 by itself. DSLReports test should be taken with a pinch of salt, of course, but I've tested this probably a hundred times now and it occurs regardless of the server in use lol

Posted

But dont u want 0 packet loss? So how does reactive perform worse or give u packet loss. Reactive for me gives me more stable pings while my line is being hammered by the people at mt house. Preemtive heps but the ping spikes are a little higher.

Posted

But dont u want 0 packet loss? So how does reactive perform worse or give u packet loss. Reactive for me gives me more stable pings while my line is being hammered by the people at mt house. Preemtive heps but the ping spikes are a little higher.

You'd think absolutely zero packet loss with that latency and jitter would be amazing. Apparently not lol

 

I have noticed that reactive seems better when the line is being hammered as you say though. If I download something I can drop it to just 97% on reactive and there'll still be next to no jitter, whereas preemptive seems a little more unstable and you get the bandwidth drop too. 100% preemptive gives me like 75% of my download lol

Posted

Could argue all testing tools inflate results. Only using dslreports as a comparison tool I.e. Test, make a changes, test. Repeat to check result consistent.

 

Ping plotter may be another option however I can monitor ping in game where jitter is less than 1ms so ping doesn't appear to be the culprit. Suspect lag comp is the culprit and the developers lag algorithm has the decimal point in the wrong place.

 

Perhaps if they insist on following this approach (and forcing a fictitious game outcome) then they should put up the proposed game scoreboard at the start of the match and let us decide whether we want to play that lobby.

Posted

Could argue all testing tools inflate results. Only using dslreports as a comparison tool I.e. Test, make a changes, test. Repeat to check result consistent.

 

Ping plotter may be another option however I can monitor ping in game where jitter is less than 1ms so ping doesn't appear to be the culprit. Suspect lag comp is the culprit and the developers lag algorithm has the decimal point in the wrong place.

 

Perhaps if they insist on following this approach (and forcing a fictitious game outcome) then they should put up the proposed game scoreboard at the start of the match and let us decide whether we want to play that lobby.

It's so random you might as well just pick a captain for each team, flip a coin, call it and go back to the prelobby lmao

Posted

DslReports never shows me Bufferbloat anymore, not sure why :/

I've had this problem for the last few months. The graph /dial pops up for a second then disappears.

Posted

I've had this problem for the last few months. The graph /dial pops up for a second then disappears.

I tend to get this randomly. Sometimes a bufferbloat grade shows up as just a dash in a grey circle too lol

 

I normally just need to reload the test and nine times out of ten the dial will reappear.

Archived

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

×
×
  • Create New...