Proposed SIG-Simulation meeting agenda for 2023/03/23

[lgleim]

Meeting Details

• Date/Time: 2023/03/23 @ 16:30 (24-hour) UTC
• Location: Discord SIG-Simulation Voice Room

SIG Updates

What happened since the last meeting?

• @amzn-ahmadkrm added to https://github.com/orgs/o3de/teams/sig-simulation-maintainers
• @smurly @lumberyard-employee-dm added to https://github.com/orgs/o3de/teams/sig-simulation-reviewers

Meeting Agenda

• Review pending RFCs
• Review sig-simulation planning for 2023
• Discuss any other proposed topics

Outcomes from Discussion topics

Discuss outcomes from agenda

Action Items

Create actionable items from proposed topics

Open Discussion Items

List any additional items below

[hultonha]

Follow-up for @lgleim. In the sig-simulation governance meeting we discussed two boards

• https://github.com/orgs/o3de/projects/13
• https://github.com/orgs/o3de/projects/23

Do we know if both are needed or should we focus our efforts on one or the other?

[hultonha]

sig-simulation governance meeting discussion

Review sig-simulation planning for 2023

• One item discussed was whether it would be good to look at a way to support an alternative physics back end (for example MuJoCo) that @adamdbrw has been experimenting with.
  • This would apply to adding a single new physics engine as well as making it more architecturally friendly to add new physics engines in general.
  • There is an example of adding an extra physics engine (Jolt Physics) in this O3DCon presentation - https://www.youtube.com/watch?v=qPvBjliKJb8&list=PLCQwFpnHSZQgzCpMmbxruFkWr3d73ZfEJ&index=35
  • Performance of different backends is a concern when it comes to performance (e.g. MuJoCo is single threaded).
  • Note: This is a complicated thing to do. For example Gazebo Classic 'technically' supports 5 physics engines, but only one supports everything.
• Soft body physics - different requirements in terms of realism (e.g. mobile game vs robotics/simulation). In the simulation case you need the physical accuracy, in the first case it is not as important but can just look cool.
  • Using PhysX 5 for this may be sufficient
  • Target could be Omniverse parity

[lgleim]

Follow-up for @lgleim. In the sig-simulation governance meeting we discussed two boards

• https://github.com/orgs/o3de/projects/13
• https://github.com/orgs/o3de/projects/23

Do we know if both are needed or should we focus our efforts on one or the other?

@hultonha https://github.com/orgs/o3de/projects/13 is actually a private board and not visible to the general public. https://github.com/orgs/o3de/projects/23 is the public roadmap that we are asked to maintain by the TSC and should become part of the O3DE documentation. That being said, we do need https://github.com/orgs/o3de/projects/23 and could deprecate https://github.com/orgs/o3de/projects/13

[hultonha]

Thanks for following-up @lgleim. I propose we close https://github.com/orgs/o3de/projects/13 then if we're not really using it. Thanks!