Jump to content

For Users On Time Warner - Lag Issues


375 replies to this topic

#281 Clit Beastwood

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,262 posts
  • LocationSouthern California

Posted 05 February 2015 - 12:14 PM

Yeah they do the same thing but the user set bridging has issues with ipv6, supposedly those issues are not there when the ISP sets bridge mode. The latency also has to do (partially) with a 50hz rate used inside the modem. It's just under-powered for a docsis device thats supposed to be able to do 600mbits. I've always loved the motorola sb6141 (not enough channels for my current speed) and the 6183 was demonstrated to have substantially lower pings. Single function devices > multi function devices is what I always say :) - you can have a device that does several things poorly, or one thing well.

#282 LordSkippy

    Member

  • PipPipPipPipPipPip
  • Overlord
  • Overlord
  • 451 posts
  • LocationAustin, TX

Posted 05 February 2015 - 12:43 PM

If I set my Arris to bridged mode, incoming ports I don't want blocked get blocked. So I run it in NAT. But, only my linux box is connected to it, and wifi is disabled. If it didn't block those ports, I would have it in bridged mode.

#283 Kin3ticX

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 2,926 posts
  • LocationSalt Mines of Puglandia

Posted 05 February 2015 - 12:45 PM

View PostFierostetz, on 05 February 2015 - 12:14 PM, said:

Yeah they do the same thing but the user set bridging has issues with ipv6, supposedly those issues are not there when the ISP sets bridge mode. The latency also has to do (partially) with a 50hz rate used inside the modem. It's just under-powered for a docsis device thats supposed to be able to do 600mbits. I've always loved the motorola sb6141 (not enough channels for my current speed) and the 6183 was demonstrated to have substantially lower pings. Single function devices > multi function devices is what I always say :) - you can have a device that does several things poorly, or one thing well.


Interesting, but that doesn't explain why people still get ping spikes when they are using the supplied combo unit as their router/switch too.

#284 NocturnalBeast

    Member

  • PipPipPipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 3,685 posts
  • LocationDusting off my Mechs.

Posted 05 February 2015 - 12:53 PM

View PostKin3ticX, on 05 February 2015 - 12:45 PM, said:


Interesting, but that doesn't explain why people still get ping spikes when they are using the supplied combo unit as their router/switch too.


This theory sounds good, but I play other games on West Coast servers and never get disconnects, or unreasonable ping spikes.

#285 Kin3ticX

    Member

  • PipPipPipPipPipPipPipPipPip
  • The People's Hero
  • The People
  • 2,926 posts
  • LocationSalt Mines of Puglandia

Posted 05 February 2015 - 01:01 PM

View PostEd Steele, on 05 February 2015 - 12:53 PM, said:

This theory sounds good, but I play other games on West Coast servers and never get disconnects, or unreasonable ping spikes.



I am talking about another phenominom with the combo units. Getting erratic but manageable pings on the lan. Something like

ping 192.168.0.1
4ms
5ms
44ms
5ms
4ms
65ms

when the lan is under no load. It seems to have minimal impact because it happens about 2% of the time.

Posted Image


Let us call it, "The cheap combo unit effect"

Edited by Kin3ticX, 05 February 2015 - 01:07 PM.


#286 Clit Beastwood

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,262 posts
  • LocationSouthern California

Posted 05 February 2015 - 01:26 PM

View PostLordSkippy, on 05 February 2015 - 12:43 PM, said:

If I set my Arris to bridged mode, incoming ports I don't want blocked get blocked. So I run it in NAT. But, only my linux box is connected to it, and wifi is disabled. If it didn't block those ports, I would have it in bridged mode.


Yeah - if you call TWC and ask them to put it in bridge mode, it'll bridge without the port blocking shenanigans.

View PostKin3ticX, on 05 February 2015 - 12:45 PM, said:


Interesting, but that doesn't explain why people still get ping spikes when they are using the supplied combo unit as their router/switch too.


Oh no of course not - it's just a point of interest. The last leased combo router they gave me was consistently adding 10ms to my pings - switched to a motorola sb6141 and my pings dropped (before MAXX upgrade). Just interesting is all :)

#287 LordSkippy

    Member

  • PipPipPipPipPipPip
  • Overlord
  • Overlord
  • 451 posts
  • LocationAustin, TX

Posted 05 February 2015 - 07:40 PM

View PostFierostetz, on 05 February 2015 - 01:26 PM, said:


Yeah - if you call TWC and ask them to put it in bridge mode, it'll bridge without the port blocking shenanigans.



Might have to try that this weekend, when I have time to fix it if it goes haywire.

80ish pings from 6-9:40pm Central tonight. Occasional spikes into the 90s, but usually around 85. Not ready to call it resolved yet, maybe after a week.

Think I saw Kin3ticX in my last match, was that you in the reddish vindicator?

#288 NocturnalBeast

    Member

  • PipPipPipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 3,685 posts
  • LocationDusting off my Mechs.

Posted 07 February 2015 - 03:29 AM

Ok, still needed to use Pingzapper tonight to get sub 200ms pings. I did get a couple disconnects though using the US Central North servers. US West VPN servers are guaranteed disconnect at this time. Jon have you contacted Pingzapper yet to work on that profile?

#289 LordSkippy

    Member

  • PipPipPipPipPipPip
  • Overlord
  • Overlord
  • 451 posts
  • LocationAustin, TX

Posted 07 February 2015 - 08:08 PM

Didn't play Friday night, but pings were in the old usual range of ~80ms.

Not sure why Ed Steele hasn't seen improvement.

#290 Cabbage Merchant

    Member

  • PipPipPip
  • Bad Company
  • Bad Company
  • 69 posts
  • LocationUnited States of America

Posted 07 February 2015 - 08:37 PM

I am no longer experiencing rubberbanding/teleporting as of Friday, though CTDs have increased in frequency. The CTDs only happen occasionally when connecting to a match, which is a huge improvement over rubberbanding during the match.

Overall better gaming experience. Thanks to whoever (or whatever?) fixed this issue!

If it fixed itself then direct my thanks to random Dev so it's not wasted.

#291 Clit Beastwood

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,262 posts
  • LocationSouthern California

Posted 07 February 2015 - 10:14 PM

View PostEd Steele, on 07 February 2015 - 03:29 AM, said:

Ok, still needed to use Pingzapper tonight to get sub 200ms pings. I did get a couple disconnects though using the US Central North servers. US West VPN servers are guaranteed disconnect at this time. Jon have you contacted Pingzapper yet to work on that profile?


Ed, a profile isn't going to change anything - the profiles are just a pre-configured sample of what we already do (tcp/udp, executable, loader) - the dc's don't have anything to do with the profiles. They have to do with either the traffic at the server itself from all the inbound connections, or traffic between you and the server.

#292 SoggyGorilla

    Member

  • PipPipPipPipPip
  • The Money Maker
  • The Money Maker
  • 194 posts

Posted 07 February 2015 - 11:17 PM

as far as i can tell the problem has been resolved since wednesday. my traffic still comes from telia, but now takes a different route through ovh

#293 Tuku

    Member

  • PipPipPipPipPipPipPip
  • The Grizzly
  • The Grizzly
  • 529 posts

Posted 08 February 2015 - 12:39 AM

yup im running steady at 75 and only up to 110 during peak hours (Lots of internet users in the house and my immedeate area.)

#294 Zuri Prime

    Member

  • PipPipPipPipPip
  • The Raider
  • The Raider
  • 120 posts
  • LocationThe Periphery

Posted 08 February 2015 - 01:35 AM

Confirming the lack of high latency spikes without Pingzapper running, although my ping is ~15+ms better with the VPN tunneling in the background through Chicago 1.

For now, I'll keep using Pingzapper for the sake of consistency pingwise.

#295 NocturnalBeast

    Member

  • PipPipPipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 3,685 posts
  • LocationDusting off my Mechs.

Posted 08 February 2015 - 01:59 AM

View PostFierostetz, on 07 February 2015 - 10:14 PM, said:


Ed, a profile isn't going to change anything - the profiles are just a pre-configured sample of what we already do (tcp/udp, executable, loader) - the dc's don't have anything to do with the profiles. They have to do with either the traffic at the server itself from all the inbound connections, or traffic between you and the server.


I am aware, but Jon may be able to give Pingzapper some insight on how to improve their service (as far as tunneling to MWO).

View PostAirmanator, on 08 February 2015 - 01:35 AM, said:

Confirming the lack of high latency spikes without Pingzapper running, although my ping is ~15+ms better with the VPN tunneling in the background through Chicago 1.

For now, I'll keep using Pingzapper for the sake of consistency pingwise.


I am not getting the ridiculous spikes like I was, but my ping without VPN is 150ms - 160ms and not very stable. With Pingzapper, I stay between 105ms to 108ms, but that is still higher than before the server move and TWCs 300Mbps upgrade.

Edited by Ed Steele, 08 February 2015 - 01:59 AM.


#296 Clit Beastwood

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,262 posts
  • LocationSouthern California

Posted 08 February 2015 - 10:23 AM

View PostSoggyGorilla, on 07 February 2015 - 11:17 PM, said:

as far as i can tell the problem has been resolved since wednesday. my traffic still comes from telia, but now takes a different route through ovh


still can't do consistent damage without vpn.

#297 Tuku

    Member

  • PipPipPipPipPipPipPip
  • The Grizzly
  • The Grizzly
  • 529 posts

Posted 09 February 2015 - 07:48 AM

hmmm now the problem came back :( intresting how erattic it is.

#298 BellatorMonk

    Member

  • PipPipPipPipPipPipPip
  • Shredder
  • Shredder
  • 617 posts
  • LocationWallet Closed PGI Knows Why

Posted 09 February 2015 - 07:51 AM

I had started a forum thread at OVH and one of their mods/support responded asking for someone to open a ticket with some detailed trace routes. If any of you network savvy folks want to engage them directly here is the post:

https://forum.ovh.us...te-Peer-Problem

#299 Tuku

    Member

  • PipPipPipPipPipPipPip
  • The Grizzly
  • The Grizzly
  • 529 posts

Posted 09 February 2015 - 11:45 AM

So I emailed Time Warner about it and they want a traceroute and a ping of the game server back ...What is the IP of the game server so i can do that ? anyone know what IP I should be using ?

So I emailed Time Warner about it and they want a traceroute and a ping of the game server back ...What is the IP of the game server so i can do that ? anyone know what IP I should be using ?

#300 LordSkippy

    Member

  • PipPipPipPipPipPip
  • Overlord
  • Overlord
  • 451 posts
  • LocationAustin, TX

Posted 09 February 2015 - 12:24 PM

Try 192.99.40.103 and 192.99.40.105. 103 I was asked to do a tracert to by Jon, and 105 is supposedly a game server.





27 user(s) are reading this topic

0 members, 27 guests, 0 anonymous users