lordbaronstein Posted December 21, 2018 Share Posted December 21, 2018 I live in Istanbul and I have two ISP’s available atm, one is FTTC fiber, and the other one is cable, both are 50/5. I’ve tinkered with Speedtest to compare my ping to various locations across EU. Then I realized something very interesting. I had cable on my netduma, and used WiFi to compare fiber. Initially ping results for same speedtest servers was very close, cable being better a few ms if not the same. However, jitter was much much higher on the netduma (cable). For the same speedtest servers (Interoute VDC Frankfurt, London and Paris) I get 1-3 ms jitter on fiber compared to 20-30ms jitter on cable which is connected through the netduma. So I decided to tinker more and disabled the netduma and connected the cable directly to my computer. Jitter reduced drastically to 1-3 ms without the Netduma, while ping numbers stayed roughly the same. I did the exact test connecting fiber network to netduma too and jitter also increased up to 30-35 ms. After browsing the forums I read something about Ethernet cables being the suspect here so I hooked up brand new cables to test out but jitter stayed high again. 30ms jitter May not be the worst but I’ve been having serious inconsistencies on BO4 on and off again and I still feel like I’m behind almost half of a second against the others. Original duma software, or dumaos didn’t made much difference for bo4, 9 out of 10 games feel like utter crap. Taking netduma off seems to help a bit for bo4, still sluggish but 7 out of 10 games feel like crap this time. On the other hand, Ww2 is just perfect with Netduma and horrible without it. I’m very confused. What should I do to improve the experience? Should I need to worry about apparent increase in jitter? Link to comment Share on other sites More sharing options...
Netduma Staff Netduma Jack Posted December 21, 2018 Netduma Staff Share Posted December 21, 2018 4 hours ago, lordbaronstein said: I live in Istanbul and I have two ISP’s available atm, one is FTTC fiber, and the other one is cable, both are 50/5. I’ve tinkered with Speedtest to compare my ping to various locations across EU. Then I realized something very interesting. I had cable on my netduma, and used WiFi to compare fiber. Initially ping results for same speedtest servers was very close, cable being better a few ms if not the same. However, jitter was much much higher on the netduma (cable). For the same speedtest servers (Interoute VDC Frankfurt, London and Paris) I get 1-3 ms jitter on fiber compared to 20-30ms jitter on cable which is connected through the netduma. So I decided to tinker more and disabled the netduma and connected the cable directly to my computer. Jitter reduced drastically to 1-3 ms without the Netduma, while ping numbers stayed roughly the same. I did the exact test connecting fiber network to netduma too and jitter also increased up to 30-35 ms. After browsing the forums I read something about Ethernet cables being the suspect here so I hooked up brand new cables to test out but jitter stayed high again. 30ms jitter May not be the worst but I’ve been having serious inconsistencies on BO4 on and off again and I still feel like I’m behind almost half of a second against the others. Original duma software, or dumaos didn’t made much difference for bo4, 9 out of 10 games feel like utter crap. Taking netduma off seems to help a bit for bo4, still sluggish but 7 out of 10 games feel like crap this time. On the other hand, Ww2 is just perfect with Netduma and horrible without it. I’m very confused. What should I do to improve the experience? Should I need to worry about apparent increase in jitter? Hi, welcome to the forum! I reckon your testing setup is to blame here; I've not heard great things about Speedtest granting accurate Jitter results. It's certainly not the best way, and the results are extremely vague anyway. If you really want to get to the bottom of it, connect the Netduma wired to a PC, then run a Pingplotter test. Choose a target URL like Twitter.com or a local news site (to reduce noise on the test, you'll want a really reliable server so these are good options). Then, run the exact same test with nothing downloading on your Modem instead. This is the kind of testing that will yield accurate results. I hope this helps! Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.