Guests:
- Ivan Corbett, Producer II
- Benoit Beausjour, Chief Technology Officer
- Jordan Wood, Network Programmer III
TL;DW:
- Server meshing is tested with players because they break stuff.
- High player limits (500+) help figure out bottlenecks.
- Replication Message Queue is one of the results of those high capacity tests.
- RMQ fixed initial problems but revealed a bunch of new ones with high latency and packet losses.
- Lack of paralliziation was another problem not found until RMQ made it in.
- Entity synchronization (binding) also had to be optimized (example provided talks about 3 million entities per shard after a week of running a server).
- Current hope is to improve the network performance 10x or more and find the next set of bottlenecks.
That’s what happens when you leave the public facing part of the project with a skeleton crew and force everyone else to work on big boss’ darling behind closed doors. Not that I’m defending this approach but it is what it is. ╮( ˘ ､ ˘ )╭
It might interest you that Evocati is suppose to test a potential 4.0 release candidate for wave 1. Chance to test SM on live servers within the next quarter is more and more probable.
I have been playing the 4.0 tests, incredibly smooth experience. Surely not super far away now. Problems only arising under heavy load which is pretty typical even in 3.x
That’s good to hear. Lets hope the rest of cut 4.0 features won’t take too long once this one’s out.