From your Phonepower Speed Test Advanced readings, your jitter server you of 21.9 ms should be under 5 ms, and packet loss and discards should be Zero. These out of limt values are usually and indiacation of a defective or malfunctioning Modem, or poor Internet provider signal.
Try removing power from your Modem then repowering your Modem. Better yet if your Modem can be reset do that. Reference your Modem owner's Manual for procedure.
Note: If your modem contains a battery, it must be temporarily removed for a successful Modem Reboot or Reset. Reference your Modem owner's manual.
If that doesn't help, replace your Modem. If that doen't help contact your Internet provider to correct their Internet signal coming into your home.
Yuck. Thanks for the metric to aim for.thunderbird wrote:your jitter ... of 21.9 ms should be under 5 ms
Double yuck.thunderbird wrote:packet loss and discards should be Zero
There is no modem ... so that pretty much leaves the wireless ISP as my next step.thunderbird wrote:indication of a defective or malfunctioning Modem, or a poor Internet provider signal
There is no modem; but I have repowered and reset the router to defaults and then reset all the settings to little avail. I've also disconnected & reconnected the POE which powers the rooftop antenna & radio aimed at the line-of-sight WISP access point.thunderbird wrote:Try removing power from your Modem
What worked best (somewhat) were the aforementioned QoS settings in the Linksys WRT54G v5 router - but there's still more to do - if I only knew how to debug the exact cause of the lousy VOIP quality.
I sent my WISP provider a link to this thread!thunderbird wrote:contact your Internet provider to correct their Internet signal coming into your home.
WMM, I find conflicting stories out there. http://www.smallnetbuilder.com/wireless ... s-with-wmm
If it was off before, leave it off (unless enabling is required for QoS to work). But you can probably do some testing with it on and with it off.
Those port priority settings you've added...I'd recommend removing those. I think QoS by MAC is good enough. Dups and redundant entries to achieve the same thing may create more problems than benefits is what I hear.
Thanks for the helpful advice. I have it set to manual, and 400 Kbps.lbmofo wrote:Yes, I think manually specifying upstream is better...
I read and re-read that article but I didn't quite get it. Apparently WMM blocks acknowledgements (ACKs) to enable 802.11n high throughput rates. I'm not sure how that applies to Ooma. Is the Ooma WiFi adapter "n" ?lbmofo wrote:WMM, I find conflicting stories out there. http://www.smallnetbuilder.com/wireless ... s-with-wmm
I can easily turn WMM on and off to test - but the voice quality is a subjective test - so I'm not sure if there is an objective test that will work for us.lbmofo wrote:But you can probably do some testing with it on and with it off.
Good point. I'll remove the port speedup settings.lbmofo wrote: redundant entries to achieve the same thing may create more problems than benefits is what I hear.
What's the best 'objective' test of VOIP quality to test the value of any particular QoS change in the router?
Is it the test sequence that I already ran at the beginning of this thread?
Location based speedtest or line quality test (to see how Ooma would do, pick a location near the Bay area); pick VoIP testing. (example: Current Test: VoIP Test Location: USA: San Jose, California)
Yes, Ooma wifi adapter is N up to 150Mbps.
By way of update ... I called my ISP who switched me to the next tier of service (for extra money!) on the 15th of this month.lbmofo wrote:Current Test: VoIP Test Location: USA: San Jose, California)
So now my new http://www.phonepower.com/speedtest is as follows:
1. download = 8.96 Mbps
2. upload = 3.77 Mbps
3. jitter = 7.1 ms
4. packet loss = 0.0%
5. MOS = 4.1
a. download = 8961 kbps
b. upload = 3771 kbps
c. download consistency = 69%
d. upload consistency = 8%
e. download test type = socket
f. upload test type = socket
g. maximum T delay = 76 ms
h. average download pulse = 2 ms
i. minimum round trip = 36 ms
j. average round trip = 87 ms
k. estimated download bandwidth = 20800 kbps
l. route concurrency = 2.321122
m. download TCP forced idle = 71%
n. maximum route speed = 14563 kbps
o. jitter you--> server = 7.1ms
p. jitter server->you = 2.1ms
q. packet loss you-->server = 0.0%
r. packet loss server->you = 0.0%
s. packet discards = 0.0%
t. packets out of order = 0.0%
u. estimated MOS score = 4.1 I'm not sure what now to set my QoS to on my Ooma and router because I don't know the algorithm for doing so.
Any ideas what the 'new' settings should be for the QoS of the Ooma and of the Linksys Router and of the Ubiquiti antenna/radio on the roof?
I picked San Jose and this is what I got for jitter: Here are the upload speeds: Here are the download speeds: Q: What would be the algorithm for setting the QoS in my Ooma, my home broadband router, and in my rooftop antenna/router?lbmofo wrote:http://myspeed.visualware.com/indexvoip.php
Here are the rest of the graphs (you can only attach 3 at a time). Q: What do I use of the above to run the algorithm that tells me what QoS to set my oooma, my home broadband router, and the router/radio transceiver on my roof?lbmofo wrote:http://myspeed.visualware.com/indexvoip.php
You need to use http://www.phonepower.com/speedtest for the Upload and download speeds.
If you are connected Modem-Ooma-Router set the Upstream and Downstream Quality of Service settings in Ooma Setup, Advance page to 80 % of the Phonepower Speedtest results.
If you are connected Modem-Router-Ooma, set the Upstream and Downstream Quality of Service settings in Ooma Setup, Advance page to ZERO.
thunderbird is right in that if your Ooma is between modem and router, set the QoS in Ooma.
If Ooma is behind the router, set Ooma QoS to all 0. Do QoS in your router instead based on Ooma IP address/MAC address.