Hey, I have some questions related to the production envrinonment of the server.
Are GPUs with CUDA support provided in the production game server in case, some bots are powered by neural networks from libraries such as tensorflow or pytorch?
Will the bot and gamer server run on the same aws instance or will they be on different instances and communicate through some kind of network interface?
I can see the TickerTime interval is set in the config file appsettings.Development.json. Can we know the value of this config in production to get a sense of the response time from the server?
I know this is a lot , Just trying to figure out the performance drawbacks we may incure if we design the bot in a certain way.
Are GPUs with CUDA support provided in the production game server in case, some bots are powered by neural networks from libraries such as tensorflow or pytorch?
No, unfortunately no GPU or CUDA support.
Will the bot and gamer server run on the same aws instance or will they be on different instances and communicate through some kind of network interface?
They will run as different instances on the same cluster.
I can see the TickerTime interval is set in the config file appsettings.Development.json. Can we know the value of this config in production to get a sense of the response time from the server?
We’re adding the tickTimer intaval to .Production.json in the next release. It’s subject to change based on the feedback but other than that the final time 150 is what we are going with for the tournaments. If there are no complaints.
, Just trying to figure out the performance drawbacks we may incure if we design the bot in a certain way.
