The problems we're getting with my 3-person local (within 10km) sessions seem to be more intense since 2 weeks ago - we would never complain about the level of lag (latency) or jitter (sound quality errors) until then.
Looking at the performance dashboards with this lower general performance, it seems clear that my broadband leg (which arrives via microwave to my chimney) is causing bad results on my sending legs to the two others and on the receiving legs from them both. This may have been the relative qualities of links previously but the effect on the session was able to be ignored previously.
Anyone out there have a similar problem apparently caused by microwave broadband, or know any theoretical reason for problems, or anyone else who gets this? All three of us have great kit and broadband downlink and uplink speeds, and always use wired connections direct from the router RJ45 connector.
(11-19-2020, 02:04 PM)mheeley.gm@gmail.com Wrote: The problems we're getting with my 3-person local (within 10km) sessions seem to be more intense since 2 weeks ago - we would never complain about the level of lag (latency) or jitter (sound quality errors) until then.
Looking at the performance dashboards with this lower general performance, it seems clear that my broadband leg (which arrives via microwave to my chimney) is causing bad results on my sending legs to the two others and on the receiving legs from them both. This may have been the relative qualities of links previously but the effect on the session was able to be ignored previously.
Anyone out there have a similar problem apparently caused by microwave broadband, or know any theoretical reason for problems, or anyone else who gets this? All three of us have great kit and broadband downlink and uplink speeds, and always use wired connections direct from the router RJ45 connector.
>>>
Hi, although your LAN may be cabled/eth/RJ45, your WAN is still a wireless transmission, aka WiFi. (WiMAX actually) Cons of this type of connection are e.g. interference & shared bandwidth (at the 'tower'). Interference can be as trivial as high moisture atmosphere.
Thanks for your response - I absolutely agree that wireless replaces fibre leg(s) - the service is beamed from specific dishes on rural properties (whose owners may be Users) , to a set of End Users, but I had thought rather than being a soupde-up WiFi it was more akin to the microwave spec used to take mobile telco traffic to cellsites where fibre is impractical - and therefore like for like with fibre quality. I'll take it further with my provider who claim a superior service (all round in my understanding) to the local fibre provider. Certainly for business and video conferencing, and any weather effects to date, the service is equal to or better than the local fibre provider. However, it does seem that on the specifics of latency and jitter I'm causing problems. I plan to make a change to fibre in the fairly near future anyway, so any visible improvement would settle the issue!
It'd be good to hear from someone who had made that change, or the reverse, to firmly pin the problems on the technology, or not.
I suggest you each run an Internet test to measure latency and it's variation over short intervals of time (this is jitter, as shown on the graph). https://testmy.net/ has been recommended on this forum. Even though the service provider' claims of better service than fiber competitors,, there are few consumer use cases beyond what JamKazam is attempting to do that require very low latency. I would imagine if you called them up and said, "I require sub-20ms latency", they would tell you to look elsewhere. Even in our current world of Zoom/Skype/Google Meet video conferencing, we are all tolerating >100ms of latency, with freezes and stutters. Since these services are not "full duplex" (a telephony terms meaning simultaneous transmit and receive), we subconsciously just deal with it. I had a single person Facebook Messenger audio-only call with someone only a couple hundred miles away the other day, both of us wifi'ed to our high-speed internet with no other traffic, and it felt like I was talking to someone on the moon. The delay was almost a whole second, and we kept speaking over each other as a result.
Sorry for the delay in getting back. I ran the tests 'to the internet' to a node which my provider gave me on his recommended tool 'PingPlotter' and got results for latency at times greater than 20ms and significant jitter. We did actually have the discussion about whether I would like the Very Low latency option for 10x the price!
I fully agree with what you say. I've clarified what it is that I need to discover, and that is whether my microwave broadband (last mile+) performance is poor relative to what fibre would provide. I have the option to change to fibre if so (with a little inconvenience). And more generally whether the slight quality problems I'm getting are the same for everyone else - it did seem from the recent classical festival that I'm only getting a slightly greater problem of pops and occasional distortion than in those excellent sessions.
So I've actually seen these effects which show what JamKazam's trying to overcome in my case, albeit on the poor UI of the free PingPlotter trial version. I'll try https://testmy.net/ and add the results here if they are more readable.
01-07-2021, 10:40 PM
(This post was last modified: 01-07-2021, 10:40 PM by StuartR.)
(01-07-2021, 08:54 PM)mheeley.gm@gmail.com Wrote: Sorry for the delay in getting back. I ran the tests 'to the internet' to a node which my provider gave me on his recommended tool 'PingPlotter' and got results for latency at times greater than 20ms and significant jitter. We did actually have the discussion about whether I would like the Very Low latency option for 10x the price!
I fully agree with what you say. I've clarified what it is that I need to discover, and that is whether my microwave broadband (last mile+) performance is poor relative to what fibre would provide. I have the option to change to fibre if so (with a little inconvenience). And more generally whether the slight quality problems I'm getting are the same for everyone else - it did seem from the recent classical festival that I'm only getting a slightly greater problem of pops and occasional distortion than in those excellent sessions.
So I've actually seen these effects which show what JamKazam's trying to overcome in my case, albeit on the poor UI of the free PingPlotter trial version. I'll try https://testmy.net/ and add the results here if they are more readable. If you have the option to change to fiber, take it. Radio-based Internet service is always going to have unreliably high latency compared to ground-based wire. Also, be sure your local home router isn't being used for streaming or online gaming while you're in a JK session. Most home routers won't prioritize realtime audio over other traffic without QoS policy enabled.
Speed & Latency Tests 20200107.pdf (Size: 445.95 KB / Downloads: 0)
Here are the results of testing just now, via TestMy.com (Red is >35ms latency) , Ookla and PingPlotter (where Amber is >20ms latency, Red is >30 ms).
I've had some great prompt feedback and discussion from JamKazam Support and I'm going to transition to fibre broadband in the short term, meantime not expecting perfection on our sessions as regards my links! Thanks all for your input.
As I keep mentioning, especially considering this limitation, we've been getting very usable results over the last few weeks, and enjoying very good JamKazam sessions.
(01-08-2021, 12:16 AM)mheeley.gm@gmail.com Wrote: I've had some great prompt feedback and discussion from JamKazam Support and I'm going to transition to fibre broadband in the short term, meantime not expecting perfection on our sessions as regards my links! Thanks all for your input.
As I keep mentioning, especially considering this limitation, we've been getting very usable results over the last few weeks, and enjoying very good JamKazam sessions. Good news and hang in there as things should get measurably better over time.
|
