SEARCH
ASUS Member ID
Password
Forgot password?
Join member

ASUS Support Home Location : Forum > Wireless > RT-N66U (VER.B1)  
  Forum  
  Topic : [Problem]Power Saving Mode(PSM) causing latency New Topic
Optimus2357
Member
Posts : 61
Level :
Tech Points : 59
From: USA
Posted:5/30/2017 1:55:00 AM # 1
I am trying to fix a latency issue between my RT-N66U(3.0.0.4.380_7378) and specific wireless clients. The devices effected have latency between 70-400ms just to the router. It seems the only devices effected show as connecting with Power Safe Mode(PSM) under the wireless log. It effects devices connected to the 2,4Ghz and 5ghz network, and have tried different wireless radio settings, including turning on/off WMM and Bluetooth coexistence. I have also checked that the device was on AC power and tried with active traffic and when idle. Does anyone know of a way to force the router to disable PSM? Also, see first link below for the DSLR thread I also have, and see the second link for someone else's thread that was having a simular problem. >
>Stations List
----------------------------------------
idx MAC Associated Authorized RSSI PSM Tx rate Rx rate Connect Time
90:C7:XX:XX:XX:XX Yes Yes -45dBm Yes 1M 6M 02:29:46
08:ED:XX:XX:XX:XX Yes Yes -48dBm No 78M 02:30:09
34:68:XX:XX:XX:XX Yes Yes -81dBm No 58.5M 02:30:09
00:A0:XX:XX:XX:XX Yes Yes -69dBm No 1M 1M 02:30:10
E4:8B:XX:XX:XX:XX Yes Yes -43dBm Yes 1M 02:30:11
54:72:XX:XX:XX:XX Yes Yes -60dBm Yes 1M 24M 02:30:11

90:C7:XX:XX:XX:XX = 192.168.1.30 = Android smart phone (PSM Yes)
08:ED:XX:XX:XX:XX = 192.168.1.156 = Windows 7 laptop (PSM No)

Ping from Router Diagnostics Tools:
PING 192.168.1.30 (192.168.1.30): 56 data bytes
64 bytes from 192.168.1.30: seq=0 ttl=64 time=487.014 ms
64 bytes from 192.168.1.30: seq=1 ttl=64 time=73.032 ms
64 bytes from 192.168.1.30: seq=2 ttl=64 time=295.298 ms
64 bytes from 192.168.1.30: seq=3 ttl=64 time=203.412 ms
64 bytes from 192.168.1.30: seq=4 ttl=64 time=116.066 ms
64 bytes from 192.168.1.30: seq=5 ttl=64 time=336.596 ms
64 bytes from 192.168.1.30: seq=6 ttl=64 time=247.886 ms
64 bytes from 192.168.1.30: seq=7 ttl=64 time=160.912 ms
64 bytes from 192.168.1.30: seq=8 ttl=64 time=69.466 ms
64 bytes from 192.168.1.30: seq=9 ttl=64 time=288.494 ms
--- 192.168.1.30 ping statistics ---
10 packets transmitted, 10 packets received, 0% packet loss
round-trip min/avg/max = 69.466/227.817/487.014 ms

PING 192.168.1.156 (192.168.1.156): 56 data bytes
64 bytes from 192.168.1.156: seq=0 ttl=128 time=1.713 ms
64 bytes from 192.168.1.156: seq=1 ttl=128 time=1.005 ms
64 bytes from 192.168.1.156: seq=2 ttl=128 time=1.002 ms
64 bytes from 192.168.1.156: seq=3 ttl=128 time=0.995 ms
64 bytes from 192.168.1.156: seq=4 ttl=128 time=1.009 ms
64 bytes from 192.168.1.156: seq=5 ttl=128 time=0.992 ms
64 bytes from 192.168.1.156: seq=6 ttl=128 time=0.983 ms
64 bytes from 192.168.1.156: seq=7 ttl=128 time=0.977 ms
64 bytes from 192.168.1.156: seq=8 ttl=128 time=1.081 ms
64 bytes from 192.168.1.156: seq=9 ttl=128 time=0.992 ms

--- 192.168.1.156 ping statistics ---
10 packets transmitted, 10 packets received, 0% packet loss
round-trip min/avg/max = 0.977/1.074/1.713 ms>
>http://www.dslreports.com/forum/r31436172-High-Latency-to-both-Chromebooks-only?r=762#31437005>
>https://www.snbforums.com/threads/powersave-mode-doesnt-work.32278/

  Reply Reply Quote & Reply Quote & Reply
Optimus2357
Member
Posts : 61
Level :
Tech Points : 59
From: USA
Posted:5/30/2017 3:51:00 AM # 2
>::Update:: >
>I think I have isolated the difference between the devices affected and those not affected. The devices affected (2 Chromebooks, 2 different Iphones, 1 Android phone) all have batteries. With the exception of my laptop, all the devices not affected (Printer, Blu Ray player, Network audio streamer, and chromecast) don't have batteries. What makes my laptop a interesting exception is it has problems running on battery, probably because the battery is not removable and is on AC power all the time. It only runs for 5-10 min before it will shut down, even though the battery shows almost full charge. So I wonder if my laptop isn't detecting the battery so isn't assigning the PSM flag to it's Wifi connection. It could also be the laptop has Windows which patched the wireless NIC to not automatically apply the PSM flag.The question is: why isn't the Asus router ignoring that PSM flag when WMM is disabled? My best guess is a bug or design feature in the router's firmware that isn't applying or understanding the router's config for WMM. I looked through the firmware release notes and can't find any modification that would be relevant. Release notes found at link below.>
>https://www.asus.com/us/Networking/RTN66U/HelpDesk_Download/ >
>Does anyone else have a Asus RT-N66U running stock firmware that can see if they can replicate this issue? If so, I think my next step is to try 3rd party firmware. That means reconfiguring my entire network, which is difficult since I have many DHCP reservations and firewall configuration that I will have to set up in the new firmware. If someone with older firmware finds they are not effected, a downgrade would be easier since I would be able to use the same config file. Any assistance would be appreciated. Thanks!

  Reply Reply Quote & Reply Quote & Reply
Optimus2357
Member
Posts : 61
Level :
Tech Points : 59
From: USA
Posted:5/30/2017 10:14:00 PM # 3
::2nd Update:: Anyone want to chime in? Suggestions?>
>So I was doing more research on WMM APSD and found this

"Similar to how beacon interval and DTIM period works together to conserve your device’s power usage."
»routerguide.net/wmm-apsd ··· -or-off/

Which lead me to this Cisco page

"The DTIM Interval notifies a wireless client with power-save mode how many beacons till buffered multicast traffic will be broadcast by the access point. The wireless client keeps track of this DTIM Interval in order to wake-up and be available to receive traffic at the right time."
»supportforums.cisco.com/ ··· -wap200e

I noticed my DTIM was set to 3 and my Beacon was set to 100, and many of my pings were in the 300ms range. I didn't think that was a coincidence. So I lowered my DTIM to 1 and my beacon to 50ms and the latency to the chromebook instantly dropped to below 50ms with the average being around 25ms. I also enabled IGMP snooping as it optimized the multicast traffic between beacons. So now the question is; What is the optimal DTIM and Beacon Interval setting? These are my settings currently:

Band
Enable Radio Yes
Enable wireless scheduler No
Set AP Isolated No
Roaming assistant Disable
Bluetooth Coexistence Disable
IGMP Snooping Enable
Multicast Rate(Mbps) Auto
Preamble Type Short
RTS Threshold 2347
DTIM Interval 1
Beacon Interval 50
TX Bursting Enable
WMM APSD Disable
Enhanced interference management Disable
Optimize AMPDU aggregation Enable
Optimize ack suppression Enable

  Reply Reply Quote & Reply Quote & Reply
1
Previous Topic:: Next Topic
Feedback for our Website ModeratorModerator: New Topic Related FAQ Related FAQ Forum Home Forum Home
Forum Model :
Legal Information | Privacy Policy | Sitemap | Contact Us | ASUS Global
  ©ASUSTeK Computer Inc. All rights reserved.