Joined
·
26,721 Posts
No latency remote jamming?
1 - 20 of 44 Posts
S: R&C Half-curved, EM Curvy; A: YAS875EXIIS, PM 67R; T: Eastman 52nd St, Triumpf; B: Nova Bronze
Joined
·
8,211 Posts
Minimal latency remote jamming
Good concept. I'm interested in what protocols/proprietary software is hiding in the back end that allows for the connections. Has to be some data handling, latency offset and other packaging that occurs.
Keilwerth saxes (S/A/T), Selmer clarinets (S/B), Altus Azumi flute
Joined
·
3,676 Posts
The phrase "a few milliseconds" is doing a lot of lifting here.
What this company is promising (realtime collaboration between musicians around the world with no noticeable latency) is physically impossible. We discussed this in this recent thread, in which I calculated that the minimum theoretically attainable round-trip latency from LA to Singapore is about 100ms, and that's without accounting for any latencies added by the equipment or network.
I believe the testimonial from the players who were 450 miles apart, but even that is only doable under the best of conditions.
What this company is promising (realtime collaboration between musicians around the world with no noticeable latency) is physically impossible. We discussed this in this recent thread, in which I calculated that the minimum theoretically attainable round-trip latency from LA to Singapore is about 100ms, and that's without accounting for any latencies added by the equipment or network.
I believe the testimonial from the players who were 450 miles apart, but even that is only doable under the best of conditions.
Joined
·
4,623 Posts
Mhhh.. it is based on 5G wireless technology and I concur that this is pretty much impossible from a technical standpoint. 5G goes into edge nodes and then some intermediate compute nodes before it is sent via high speed pipes to the backbone. And latencies are reasonably low but just having all the hops in there will be noticeable. Of course it will be better than 4G but that really doesn't mean anything.
I have been using Jamkazam app, on my Mac with focusrite 4i4, and find it works pretty well, depending on the connections of the other folks involved.
S: R&C Half-curved, EM Curvy; A: YAS875EXIIS, PM 67R; T: Eastman 52nd St, Triumpf; B: Nova Bronze
Joined
·
8,211 Posts
"When it comes to the law of physics, no one is gonna beat that," Benincaso (One of the founders of Elk Audio) said. "What is very important is not only the latency itself, but the consistency of the latency.That's a very, very important point, because the musician can adapt, but what they cannot do, they cannot adapt if the latency is not stable."
- https://www.forbes.com/sites/dbloom...s-music-collaboration-education/#551fe4585503
Just figured that would be more beneficial to see another link, as I was looking this up.. What they are doing is stabalizing a connection so it isn't jumping in latency - and then post processing/data handling to even out how how multiple users would see/hear the connection simultaneously. Either way, it is impossible to create a perfect real time solution as science shows, and as Elk's founder clearly agrees with here.
Joined
·
3,790 Posts
A friend of mine who's a sound tech is experimenting with ham radio for low-latency jams. Ham-jams. Not private, though, and can run afoul of broadcast regulations. I suspect that a good ol' fashioned land line might work well too, though fewer and fewer folk have them anymore.
Interesting topic. I wonder how far you can stand from someone and still play in sync? If you were 20 feet apart, for instance, it takes about 18 milliseconds for their sound to reach you. So I'd think if latency is under that and consistent, it could work.
Double that time though and you'll probably be as in sync as a marching band without a drum major.
Double that time though and you'll probably be as in sync as a marching band without a drum major.
Joined
·
1,830 Posts
Yes. Broadcasting of music is forbidden on HAM radio frequencies according to international (ITU) regulations.
The only odd exceptions were that one person singing "Happy Birthday" to another was not deemed music and the other is that some recorded music from the Space Shuttle was broadcast. Personally, I think the latter was an ex post facto ruling.
Anyway, this is a very interesting subject!
Joined
·
4,845 Posts
Yes, of course "zero latency" is rubbish. Even standing next to each other if there is signal processing going on.
But no reason that the latencies cannot be balanced so that all listeners can be synchronised.
I am laying down some tracks this week for a local wind ensemble (all of the sax parts? Why not!) to be assembled for a virtual broadcast.
Not being done in real time, but has the same issues.
But no reason that the latencies cannot be balanced so that all listeners can be synchronised.
I am laying down some tracks this week for a local wind ensemble (all of the sax parts? Why not!) to be assembled for a virtual broadcast.
Not being done in real time, but has the same issues.
Joined
·
1,229 Posts
Old school telephones or quantum entanglement? Again?
It would appear that the wavefunctions of these two threads have converged
https://forum.saxontheweb.net/#/topics/375538?page=2
It would appear that the wavefunctions of these two threads have converged
https://forum.saxontheweb.net/#/topics/375538?page=2
Joined
·
1,623 Posts
It all depends on what is meant by the term "collaboration." If remote players are simply comping behind a soloist, or playing written arrangements, I'd agree that a few milliseconds probably fall within the IRL standard margin of error & therefore won't have much noticeable effect -- even though audio round trip doubles the latency.
However, in genres where polyphonic real-time improvisation is the ideal form of collaboration, musicians make instantaneous decisions about what to play based on what other players just played & where they might go with it in the next moment.
The goal is to construct a unique, stylistically appropriate improvised line that will blend (or contrast pleasingly) with the maximum range of possible next-moment outcomes. This requires rapid multi-channel mental processing: deep listening & near-instantaneous decision-making. Keeping the tune & the ensemble on track when they're continually teetering on the brink of chaos -- what an exhilarating ride!
Lose a double handful of milliseconds to latency... chaos might win.
However, in genres where polyphonic real-time improvisation is the ideal form of collaboration, musicians make instantaneous decisions about what to play based on what other players just played & where they might go with it in the next moment.
The goal is to construct a unique, stylistically appropriate improvised line that will blend (or contrast pleasingly) with the maximum range of possible next-moment outcomes. This requires rapid multi-channel mental processing: deep listening & near-instantaneous decision-making. Keeping the tune & the ensemble on track when they're continually teetering on the brink of chaos -- what an exhilarating ride!
Lose a double handful of milliseconds to latency... chaos might win.
Joined
·
1,830 Posts
I recall that Einstein wasn't fond of spooky music at a distance.
Joined
·
1,229 Posts
God doesn't play diatonics with the music of the spheres..
... Not that s/he couldn't, being omnipotent 'nd all.
Joined
·
3,790 Posts
Why not just use an ansible?
Joined
·
4,845 Posts
God not only plays diatonics, sometimes he plays them where they cannot be heard.
Joined
·
1,229 Posts
Base solo? Gravitational waves?
1 - 20 of 44 Posts
