Distance - How far from the simpit can clients interact without network latency becoming an issue? What type of network topologies could be implimented to ensure playability and performance?
Well, OMP was designed with internet operation in mind. From my tests, a maximum latency of 500ms is on the edge of playability, especially regarding on-orbit operations. This should be enough to have global location separation for both clients and servers.
Spectating - Hop in your buddy's ship and observe as a passenger? Clients can connect and merely observe what's taking place during the simpit's simulation? If so what's the maximum number of clients that can be supported?
Yes, this is already possible. Unfortunately, not every action/animation/sequence can be transmitted. ATM I experiment with Orbiter's recorder events, so it should be possible in future versions to have everything that gets recorded equally transmitted over the wire. In addition, I experimented with event hooking, too, giving the possibility to transmit MFD clicks and panel operations. This way a spectator can e.g. take seat in the DG and see the MFDs change mode...
The maximum number of clients is only limited by the supported client ports on the server (default is 500) and the network traffic generated. The vessel everyone sits in will have to transmit state vectors to every spectator. Depending on machine load, this has a certain threshold. I think it is around 100 clients for modern machines. OMP is using UDP unicast, not multicast, because the latter brings along a whole new set of NAT troubles
.
Synchronized Orbital Maneuvers - Two shuttles docking with the ISS? If so, what's the maximum number of vessels that can participate in this manner at this time?
While I focussed on this use-case right from start, it is still not done in a satisfying way. ATM the biggest problem is a rotation bug with docked vessel, when 2 remote vessels dock together.
That said, we already had many situations where 2 or more DGs/XRs were docked to one ISS. It is possible, but not stable yet.
The maximum number is just the same as for the spectator scenario, the involved mechanisms are the same.
Collision Detection - Is this on the horizon for OMP? I've seen collision detection for older versions of Orbiter at the Hangar.
No. OMP is a framework. It has no content for gameplay and is no enhancement of Orbiter physics. It just replicates local vessel on remote clients.
That said, Orulex seems to be working in connection with OMP, from what I heard.
I've installed the latest OMP release in Windows XP and Linux and been able to log onto the server. In linux I used Wine and winetricks to install dotnet1.1 and dotnet2.0 to accomplish this...
The OMP server can be run with Mono on Linux. In fact my private server is a Debian Linux UML 2.4 kernel with Mono 1.2.6, running the Visual Studio compiled DLLs directly.
The OMP client OTOH is of course a mixed-mode C++/C# DLL. Running this in Linux is certainly a challenge
.
The OMP team is phenominal. Thank you in advance for reading this post and if possible addressing my concerns.
:lol:
Thanks for the kudos. I hope my reply is answering your questions...
Although there is no official OMP "team", I think deltawing777 is doing a great job hosting his server. Together with Axertans web-page design and artwork by CarmenA, I'd say the OMP "staff" is already bigger than I ever imagined it to be.
regards,
Face