Thank you all for you input. Please excuse addressing everyone in one email but I think this is for the sake of my sanity. On Thu, Nov 19, 2009 at 3:27 AM, Lorenzo <lsutton@xxxxxxxxx> wrote: > Although some attempts have been made (see for eg. > http://www.computer.org/portal/web/csdl/doi/10.1109/MMCS.1999.778626) it > looks like currently network (even cabled) latency issues are still too much > for performing remotely. Thanks for that article. I should be able to get it through the university. The network latency is not THAT much of an issue to me. I do expect some latency and the composer who will write the music will take the latency into consideration. And the music will probably not be beat oriented in any way. That said, latencies in the range of tens of seconds would be a little extreme. > If I understand correctly what you have in mind, and considering the > relatively small space-range (a bulding) you might be better using wireless > (radio) audio equipment (like radio mics and audio transmitters) which seems > to be more established and reliable, although there will always be 'some' > latency. The problem is that radio communication does not work well in closed spaces and, well, I was more than vague, but we will actually span several buildings (university campus). So radio is going to fail. We will be able to use the university's WiFi infrastructure. Another important consideration is that we want to "orchestrate" routing of signals between the 5 musicians which would be impossible(?) (without much manipulation at least) using radio (or cell phones, which we have taken into consideration). We do not expect the players to be tech savvy so the idea is that we can setup the computers and script as much as possible the players only boot their computers and off they go. > Keep in mind that Ninjam introduces a 'fixed latency' (in musical terms it's > usually 1 beat 'out') to try and sync all palyers, which is clever and suits > some kind of applications like jamming, but not real-time remote playing. Ninjam is something I want to take a look at but when I tried it yesterday I had problems with client not being able to access my sound card. I will look into it because a fixed latency (that I can predict) might be a an interesting compromise. I am also looking at http://www.celt-codec.org/ if I can get something going in that area. 2009/11/19 Jörn Nettingsmeier <nettings@xxxxxxxxxxxxxxxxxxxxxx>: > i'd say you could get <3s roundtrip with careful tuning, but that is > still unacceptable for live playing "on a beat". I don't think there will be "on the beat" happening but 3 secs might still be a lot (unless it is a constant time interval that we can work into the music like ninjam seems to be able to do) > and you need to avoid packet collisions. with 10 (!) real-time audio > streams on one wifi, there will be collisions all over the place. the > only way around that i can see is to use several wifi cells, and keep > each performer on a separate one, if possible. still, risky. it also > means you need to hire thugs that jump on every spectator trying to pull > her/his iphone or other wifi gadget out. Well, the musicians will be in different areas of the campus so they will be using different access points at least. We will not have the budget for the thugs but if the network communication is a real issue we think we may have some special treatment from the IT department... ;) > but: you could try vorbis encoding downstream over http (with re-send in > case of packet loss) at just under a second, and netjack/CELT for the > monitoring, where dropouts are more tolerable. you mean using icecast or something? I will take a second look at netjack, I could not get it to work for some reason, but the problem is probably with my brain. Also, we don't mind dropped packets as long as most of the reach the destination (i.e. you can actually clearly hear that a sax is playing a melody). Even if it stutters a bit. It is part of the aesthetic. On Thu, Nov 19, 2009 at 8:18 AM, Jonathan E. Brickman <jeb@xxxxxxxxxxxxxxx> wrote: > Five musicians plus a server, in high fidelity, realtime, moving around. > That's a lot of data, and you need it minimum latency and minimum loss. > What kind of wireless router are you using, and what kind of wireless do the > laptops have? we don't yet have any specific laptops but were thinking of netbooks with 802.11n type interfaces. We want the gear to be as light as possible and still offer some flexibility. The network is 1GbE, in most areas at least (not in the area where I am testing it though...) > I would start with top-of-the-line Belkin [...] Thanks for the tips on Belkin, will investigate that. Respectfully yours -- ./MiS 514-344-0726 http://www.creazone.ca _______________________________________________ Linux-audio-user mailing list Linux-audio-user@xxxxxxxxxxxxxxxxxxxx http://lists.linuxaudio.org/mailman/listinfo/linux-audio-user
