Greetings music makers.
Koord just launched a free Beta private server creation web-service.
You can easily create a session with a couple of clicks and play using Jamulus. https://koord.live/
Try it out and share your experience here.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Having been involved in the Jamulus project for a few months, I get the impression that the idea of public servers isn't really in line with how the majority of people want to use Jamulus. That's only been my general impression from reading the forums, the Facebook group and observing the many empty servers available. It's hard to know if that's accurate without knowing how may private servers are in fact also in use. Koord lends more weight to my hypothesis though.
As to the experience of Koord, it will be interesting to know whether people misinterpret the map to mean that either they can only play with people in those cities (which is probably incorrect), or set up a session in London and play with a friend in New York (which is certainly incorrect). And if people think the latter, then this is going to make Jamulus look a bit bad :-)
Last edit: Gilgongo 2020-12-03
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I think that in the pre-Covid era, the only real reason for playing online was to jam with distant people - not for many. Now a larger population of musicians that cannot play together even if not too far has found Jamulus as solution. Anyway, Koord (and Melomax) are a good idea, although I am not sure it will be a long-lasting business (at least I hope :( ). The issue of latency is clearly exposed in the site.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The interesting part (for me) is the session management you've added - its quite a faff to manage server recordings manually and usually I just don't bother.
In my experience the biggest limitation to greater adoption of jamulus is not simpler session management though - its the lack of support for Android and (to a lesser extent) iOS devices. So many of the people I'd like to be making music with are not IT experts and only have tablets or phones.
There's been some development work done on this, I know, but nothing has been released to date. Jamulus is open source - so its entirely dependent on people giving up their time to develop the application. I wonder if your group could take on and complete that challenge? Then I feel your service would really be of general interest.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
@Gilgongo you are absolutely on point with the ambiguity with the map region chooser :)
We are aware that this is not providing the information necessary, and are in the process of building out a new UI that clarifies things. On a later iteration we will use an automated preliminary ping-check to confirm the best city-region for the session creator, and advise (to some extent) on the supported geographical coverage for that session. This last point is a little hazy as we can often do only a "best guess" using incomplete public data on internet backbone hop latency.
But we will do our best and keep on improving.
@Vincenzo We agree that Jamulus is a core part of a new way of collaboration for lots of musicians. And the effective latency of the internet in many regions is already sufficient to support a huge number of musicians globally (we are massively updating our region support in an upcoming release). The pandemic revealed to us and professional musicians in our circles the lack of services that take out the hard, technical parts of collaborating online. We have only done the most basic part of this so far, but see this as a useful part of a solution for lots of musicians beyond the pandemic - and yes may that be soon, as you say!
@Andrew S Mobile support is definitely an important milestone for Jamulus development. We have already picked this up quite strongly from our early beta feedback. We are tracking the development in this area and are definitely looking to assign some resources to both the Android (alpha stage) and iOS (nonexistent) efforts. We have started to maintain a fully-compatible clone at https://github.com/koord-live/jamulus to work on these clients, but at the moment there is nothing added there other than some very rough early hacks towards creating a running WebAssembly build of Jamulus (a longer-term milestone).
Thanks for confirming our conviction on this.
Currently our dev team is maxed out on bandwidth (and mostly unpaid), but with any luck we will be able to expand and add more resources before too long.
👍
1
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Regarding professionals, be aware that there is Aloha by Elk in beta and likely out in few months. It's a different thing (p2p, no server involved), likely expensive, but from a business point of view it is surely a competitor on such audience (if you in future will be able to provide Jambox-ready boxes...).
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I somehow find their claim of <1 mS roundtrip latency a little hard to believe, unless they've found a way for electrons to flow at faster than the speed of light! JamKazam (who are about to go subscription only) is p2p as well - and they used to produce a box which they ended up discontinuing due to lack of interest. JamKazam is fine for small groups with good Internet connections, but doesn't scale. I expect Aloha to be the same. Perhaps useful to plug in remote collaborators to the studio system?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I somehow find their claim of <1 mS roundtrip latency a little hard to believe
Yeh, after I saw that I only skimmed the rest. Their PR group is out-of-control. Even photons are bound by the speed of light. P2P may be good for small groups of technically interested people with good Internet connectivity. For most people out there Jamulus has found the sweet spot I think.
Getting back to the OT, however, I do not think that we have any lack of access to servers. That is not the problem. The problem is all the hand-holding necessary to get people going, getting their client running and the audio adjusted, even if they don't have any particular technical problems.
I am in the process of getting the amateur orchestra that I play with up and running on Jamulus. We started with a test rehearsal with only 4 people, but including director and 1st violin. Each week we have a couple more members playing with us as word-of-mouth gets out that it really works and we really can play together. Almost all are running MACs using the internal audio. The biggest problem we have had was the tendency of most people to set their microphone too loud causing distortion when we play loud.
I hope that we can get everyone on-board in the next couple of weeks, but it does take a bit of time with each individual to get things up and running.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
At least in some place the <1ms is stated as related to the interface latency, not to the overall latency, which in any case depends on the network. This would be better than USB interfaces.
Anyway, it will good for pros, and I am happy with a server-based solution that also allows to record much much better than our previous smartphone-based, drums-covered recordings :)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Greetings music makers.
Koord just launched a free Beta private server creation web-service.
You can easily create a session with a couple of clicks and play using Jamulus.
https://koord.live/
Try it out and share your experience here.
Interesting.
Having been involved in the Jamulus project for a few months, I get the impression that the idea of public servers isn't really in line with how the majority of people want to use Jamulus. That's only been my general impression from reading the forums, the Facebook group and observing the many empty servers available. It's hard to know if that's accurate without knowing how may private servers are in fact also in use. Koord lends more weight to my hypothesis though.
As to the experience of Koord, it will be interesting to know whether people misinterpret the map to mean that either they can only play with people in those cities (which is probably incorrect), or set up a session in London and play with a friend in New York (which is certainly incorrect). And if people think the latter, then this is going to make Jamulus look a bit bad :-)
Last edit: Gilgongo 2020-12-03
I think that in the pre-Covid era, the only real reason for playing online was to jam with distant people - not for many. Now a larger population of musicians that cannot play together even if not too far has found Jamulus as solution. Anyway, Koord (and Melomax) are a good idea, although I am not sure it will be a long-lasting business (at least I hope :( ). The issue of latency is clearly exposed in the site.
The interesting part (for me) is the session management you've added - its quite a faff to manage server recordings manually and usually I just don't bother.
In my experience the biggest limitation to greater adoption of jamulus is not simpler session management though - its the lack of support for Android and (to a lesser extent) iOS devices. So many of the people I'd like to be making music with are not IT experts and only have tablets or phones.
There's been some development work done on this, I know, but nothing has been released to date. Jamulus is open source - so its entirely dependent on people giving up their time to develop the application. I wonder if your group could take on and complete that challenge? Then I feel your service would really be of general interest.
Thank you guys for your astute comments.
@Gilgongo you are absolutely on point with the ambiguity with the map region chooser :)
We are aware that this is not providing the information necessary, and are in the process of building out a new UI that clarifies things. On a later iteration we will use an automated preliminary ping-check to confirm the best city-region for the session creator, and advise (to some extent) on the supported geographical coverage for that session. This last point is a little hazy as we can often do only a "best guess" using incomplete public data on internet backbone hop latency.
But we will do our best and keep on improving.
@Vincenzo We agree that Jamulus is a core part of a new way of collaboration for lots of musicians. And the effective latency of the internet in many regions is already sufficient to support a huge number of musicians globally (we are massively updating our region support in an upcoming release). The pandemic revealed to us and professional musicians in our circles the lack of services that take out the hard, technical parts of collaborating online. We have only done the most basic part of this so far, but see this as a useful part of a solution for lots of musicians beyond the pandemic - and yes may that be soon, as you say!
@Andrew S Mobile support is definitely an important milestone for Jamulus development. We have already picked this up quite strongly from our early beta feedback. We are tracking the development in this area and are definitely looking to assign some resources to both the Android (alpha stage) and iOS (nonexistent) efforts. We have started to maintain a fully-compatible clone at https://github.com/koord-live/jamulus to work on these clients, but at the moment there is nothing added there other than some very rough early hacks towards creating a running WebAssembly build of Jamulus (a longer-term milestone).
Thanks for confirming our conviction on this.
Currently our dev team is maxed out on bandwidth (and mostly unpaid), but with any luck we will be able to expand and add more resources before too long.
Regarding professionals, be aware that there is Aloha by Elk in beta and likely out in few months. It's a different thing (p2p, no server involved), likely expensive, but from a business point of view it is surely a competitor on such audience (if you in future will be able to provide Jambox-ready boxes...).
I somehow find their claim of <1 mS roundtrip latency a little hard to believe, unless they've found a way for electrons to flow at faster than the speed of light! JamKazam (who are about to go subscription only) is p2p as well - and they used to produce a box which they ended up discontinuing due to lack of interest. JamKazam is fine for small groups with good Internet connections, but doesn't scale. I expect Aloha to be the same. Perhaps useful to plug in remote collaborators to the studio system?
Yeh, after I saw that I only skimmed the rest. Their PR group is out-of-control. Even photons are bound by the speed of light. P2P may be good for small groups of technically interested people with good Internet connectivity. For most people out there Jamulus has found the sweet spot I think.
Getting back to the OT, however, I do not think that we have any lack of access to servers. That is not the problem. The problem is all the hand-holding necessary to get people going, getting their client running and the audio adjusted, even if they don't have any particular technical problems.
I am in the process of getting the amateur orchestra that I play with up and running on Jamulus. We started with a test rehearsal with only 4 people, but including director and 1st violin. Each week we have a couple more members playing with us as word-of-mouth gets out that it really works and we really can play together. Almost all are running MACs using the internal audio. The biggest problem we have had was the tendency of most people to set their microphone too loud causing distortion when we play loud.
I hope that we can get everyone on-board in the next couple of weeks, but it does take a bit of time with each individual to get things up and running.
At least in some place the <1ms is stated as related to the interface latency, not to the overall latency, which in any case depends on the network. This would be better than USB interfaces.
Anyway, it will good for pros, and I am happy with a server-based solution that also allows to record much much better than our previous smartphone-based, drums-covered recordings :)