What is the maximum distance two players can be separated by and still synchronize like they are in the same room? Short Answer: 25ms Long Answer: The maximum one-way delay between two players can be 25 milliseconds. That is the amount of time it should take for the sender to strike the note, send it across the Internet, and be heard by the remote receiver. Since network travel times vary as a result of many factors, not just distance, there is no exact answer to this question. However, the general rule of thumb is approximately 0-1000 miles. Again, this will vary on each connection, even within the same session. It is still possible for players in a session to synchronize their playing when one-way delays exceed 25ms, but the effects become more apparent as the delay gets longer and the players have to work harder to keep their playing together without slowing down. There is no "hard" number for maximum acceptable delay, but most players will find that one-way delays in the 40-50ms range will be very noticeable and will require much effort to subdue the natural tendency to slow down in anticipation of the remote player's sluggish audio stream.
I think there's a wide range of "what's exceptable" as far as latency is required. My band rehearsals feel incredibly tight and we're about 200 miles apart.
My friend Susan in San Francisco collaborates with a writing partner in France. They will be using the JamLink soon. It will be really interesting to see how their sessions go....they are both really excited to get started. Sure there will be some extra latency on a California/France connection, but it sure beats the time and expense of intercontinental flights!
--DAN--
Jeff Porcaro could tell when a track was off by even 2ms. How could anyone, particularly seasoned musicians with great time, play with 25ms delays?
here i am checking out all the new bells and whistles and the first thing i read is about jeff porcaro, who i used to jam with in high school, ...eyeyeyeyeye... is there some disillusionment in your tone? i believe the answer lies in approach and adaptation, in which the very fabric of the piece is imbued with wrinkles in the expected time. perhaps there are some musics not suitable for online, but certainly there is much that is suitable.
well put. For me, I hope to use it as a collabartion tool.
As for the latency...15 ms is a lot to tolorate especially if tempos are in the 180 + and you are playing bebop...
I'm dubious about 2 points: -- that there is a "hard" limit on latency that affects two players' ability to sync and -- that Porcaro was hearing "rhythm sync" between two players and not just a timbre change when sliding two tracks relative to one another For the former, stay tuned to our next paper on the subject. The most recent ones with the best data are at http://ccrma.stanford.edu/~cc/shtml/ensDelay.shtml The new one will maybe refine the data a bit more (in progress). As to Porcaro, can we hear more about what he was listening to? Chris