RL317 Posted May 23, 2017 Share Posted May 23, 2017 I turned on my PS4 today and apparently my hard drive committed suicide, because it asked me to insert a USB drive with the latest 4.55 software for reinstallation. I reinstalled it and went about sorting all my settings and whatnot out again. Everything got wiped I was bored while waiting for my IW update to download, so I thought I'd get on the browser and find all of the sites I had bookmarked. I got to the DSLReports site and decided to run some tests just to see how my line performed while it was busy. Bear in mind that as well as downloading the game patch, we had the Amazon Fire TV on downstairs streaming too, and that alone was using around 25Mbit according to the Duma network monitor. I pay for 50/10 from BT. At first I started out on the usual 70/70 preemptive, but I noticed that the test was starting out with more than my usual 8ms to the London server (more like 20ms), and there was a ton of jitter too. I pulled up my Duma GUI on my phone and switched from preemptive to reactive, and to my amazement the minimum latency immediately fell to 8ms and the jitter started tumbling down too. I ran another test, this time starting out on reactive, and it would give me either 8 or 9ms minimum with <0.5ms jitter. If I swapped back to preemptive it would immediately start climbing to as much as 25ms of jitter. I repeated the test several times to take screenshots, but I forgot which sliders I'd used for each test so I got rid of them to prevent confusion lol. I ran one more, starting at 90/90 reactive as you'll see here, and as you'll see it shows 9ms min latency and 0.4ms jitter. Halfway through I swapped to 90/90 preemptive and it started climbing again, as shown in the second screenshot. The moral of the story is that while using 70/70 preemptive is a good general guideline when first setting up your Netduma, don't be afraid to test in depth for yourself to find something that keeps your latency down while your connection is busy. I always assumed 70/70 preemptive was best for my line type on 50/10 FTTC, and it makes sense to assume it would be best from a gaming perspective to keep spikes down at the cost of bandwidth, but even 99/99 reactive works better for me than 70/70 preemptive! Now if only low latency and jitter meant good IW gameplay... fuzzy clam, Netduma Jack and TryHardUndies 3 Link to comment Share on other sites More sharing options...
Zennon Posted May 23, 2017 Share Posted May 23, 2017 That's why you should always test for yourself. Even when I was on 6/1 reactive worked out best for me fuzzy clam and RL317 2 Link to comment Share on other sites More sharing options...
Lyncho84 Posted May 23, 2017 Share Posted May 23, 2017 Interesting must try this out cuz it is awful at the min for me and have pretty much the same but package as urself RL317 1 Link to comment Share on other sites More sharing options...
N3CR0 Posted May 23, 2017 Share Posted May 23, 2017 Reactive has given me less BB when I was on VM and on BT. RL317 1 Link to comment Share on other sites More sharing options...
PharmDawgg Posted May 23, 2017 Share Posted May 23, 2017 And alot of it depends on your ISP and the type of service you have be it cable or dsl. I on the other hand have Comcast cable and I have always used reactive but I have gone the opposite and I use preemptive. It keeps my network congestion to a minimum because my little mini me's like to stream and such while I am trying to game. I have come to find out that they can affect my gaming experience alot more that my ISP can. But like it said I use preemptive with 80/80 with A+'s across the boards. RL317, fuzzy clam and ColonicBoom 3 Link to comment Share on other sites More sharing options...
RL317 Posted May 23, 2017 Author Share Posted May 23, 2017 Yeah it's pretty weird. I can use 97/97 and get A+ across the board, but I never thought to do a test when the line was busy. I've only ever really used PingPlotter when nothing else was online for example. I rarely game with other devices online anyway, but I thought it was interesting and weird enough to share Link to comment Share on other sites More sharing options...
RL317 Posted May 23, 2017 Author Share Posted May 23, 2017 Reactive has given me less BB when I was on VM and on BT. On VM 100/6 I had to drop my sliders to around 12/35 for down and up respectively just to get an occasional A+ bufferbloat, sometimes an A. On BT I can use 97/97 and it's never anything but A+ for both bufferbloat and quality. Weird how my gaming experience from Ghosts onwards seems to have gone downhill as my setup has improved lol N3CR0 1 Link to comment Share on other sites More sharing options...
iAmMoDBoX Posted May 23, 2017 Share Posted May 23, 2017 Are you sure your PS4 wasn't in hyper lane while doing this testing? That would cause those results. Link to comment Share on other sites More sharing options...
N3CR0 Posted May 24, 2017 Share Posted May 24, 2017 On VM 100/6 I had to drop my sliders to around 12/35 for down and up respectively just to get an occasional A+ bufferbloat, sometimes an A. On BT I can use 97/97 and it's never anything but A+ for both bufferbloat and quality. Weird how my gaming experience from Ghosts onwards seems to have gone downhill as my setup has improved lol On VM, I had best BB 15/15 but still rarely got A+. On BT, 70/70 is an A+ everytime. Link to comment Share on other sites More sharing options...
ColonicBoom Posted May 24, 2017 Share Posted May 24, 2017 On Virgin Media, 100/100 would give me an A+. Which is why I don't bother testing with dsl reports. N3CR0 and RL317 2 Link to comment Share on other sites More sharing options...
RL317 Posted May 24, 2017 Author Share Posted May 24, 2017 Are you sure your PS4 wasn't in hyper lane while doing this testing? That would cause those results. I'm on firmware 1.03.4 (which doesn't have hyper lane) so I can use the geofilter without the GUI lagging lol Link to comment Share on other sites More sharing options...
RL317 Posted May 24, 2017 Author Share Posted May 24, 2017 On Virgin Media, 100/100 would give me an A+. Which is why I don't bother testing with dsl reports. I found VM would give me different results depending on how I tested. Fully wired it was terrible, yet I could connect my phone to WiFi from the other side of the house, run the desktop version of the site on my phone's browser and get A+ with 97% reactive lol Link to comment Share on other sites More sharing options...
fuzzy clam Posted May 30, 2017 Share Posted May 30, 2017 Everyone's network...ISP,set up etc is different. The 70/70 settings are a guideline and a great starting point, the more you get comfortable with the duma and it's controls you just tweak from there some will end up less than 70,some more, some reactive, some preemptive. That's why as Z said it's always a good idea to test for yourself and see for yourself what results are best for your network. ColonicBoom and RL317 2 Link to comment Share on other sites More sharing options...
RL317 Posted May 30, 2017 Author Share Posted May 30, 2017 Yeah I think I recall Zennon saying in a post once that he got far better results using reactive on something like a 5 down 1 up line? I guessed there were other factors in play affecting how each algorithm works with different lines but I just assumed preemptive would be better as the majority of people report they find it's better for gaming, while reactive is better for raw bandwidth. There doesn't seem to be much difference, if any at all, between preemptive and reactive when gaming if nothing else is connected though. I'm testing a whole different bunch of variables in IW right now as I've think I've discovered a way to make it actually playable on 7ms pings, so I'll keep comparing those two as well fuzzy clam and N3CR0 2 Link to comment Share on other sites More sharing options...
TrayDay Posted May 31, 2017 Share Posted May 31, 2017 Yeah I think I recall Zennon saying in a post once that he got far better results using reactive on something like a 5 down 1 up line? I guessed there were other factors in play affecting how each algorithm works with different lines but I just assumed preemptive would be better as the majority of people report they find it's better for gaming, while reactive is better for raw bandwidth. There doesn't seem to be much difference, if any at all, between preemptive and reactive when gaming if nothing else is connected though. I'm testing a whole different bunch of variables in IW right now as I've think I've discovered a way to make it actually playable on 7ms pings, so I'll keep comparing those two as well Now that would be definitely worth a post lol. RL317 1 Link to comment Share on other sites More sharing options...
fuzzy clam Posted May 31, 2017 Share Posted May 31, 2017 Yeah Z had a 6/1 line and ran reactive and I believe his line had literally zero jitter as rock solid as you could get. RL317 1 Link to comment Share on other sites More sharing options...
Recommended Posts
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 accountSign in
Already have an account? Sign in here.
Sign In Now