Archive for October, 2009
Game Engine Architecture and Nvidia PhysX: video lecture
Driven by Joe Andresen from SUN Microsystems, this 51-minute long video lecture is part of Sonoma State University colloquium. Talk is covering some aspects about how PhysX SDK works in standart game engine environment as physics engine, how to organize engine pipeline and rendering system correctly, etc – might be interesting for starting programmers.
Source: CSUSonoma
GTC: PhysX demonstrations from Jen-Hsun Huang keynote
During opening Keynote with Jen-Hsun Huang at GPU Technology Conference, several PhysX applications were used to demonstrate computing capabilities of new Fermi architecture. We are now able provide a little more details.
SPH-fluid simulation with Navier-Stokes solvers demo was built using 128 000 particles to create realistic realtime water behavior.
Update:
PhysX Wall Destruction demo is using APEX Destruction Module to simulate wall being destroyed by the impact of flying ragdolls. Over 10 000 rigid bodies represents debris and wall chunks – all being processed and fully self-shadowed by the GPU.
Update #2:
Video for PhysX fluid demo available
PhysX SDK 2.8.3. available

32-bit and 64-bit builds of newest PhysX SDK 2.8.3., along with 9.09.30 PhysX System Software are available for download at PhysX Developer Support site, but we think it won’t take long until SDK will appear as public download.
Update: Changes in PhysX SDK access procedure
Meanwhile, some highlights from Release Notes:
General
- This release enables clothing simulation. These new features are required to run the APEX clothing module.
- The release also adds a hierarchical solver for cloth (HPBD Solver).
- Introduction of split pair notification for cloth (existed for soft bodies only in previous versions).
- Improved statistics to track GPU memory utilization.
- Updated with 64Bit Windows support.
Performance
- The hierarchical cloth solver runs before the regular solver at each time step. This does not immediately improve cloth simulation performace. However, running the hierarchical solver achieves the same stiffness of the cloth with fewer solver iterations of the regular solver. Thus, the performance is increased by explicitly reducing the number solver iterations.
- Added support to run force field updates on the scene’s worker threads.
Fixed Bugs
- NxArray::insert() did not work properly if elements were inserted anywhere else than at the end of the array.
- Crash when fluid tries to access invalid convexes. Mirrored convexes can get corrupted when a heap allocation fails.
- GPU fluids: Force and deletion updates are performed on wrong particles in case the updates address particles that have been added since the last execution of NxScene::simulate().
- GPU fluid mesh collision kernel crashes on GT200 cards due to heavily unbalanced workload.
- GPU fluids: Releasing static primitive rigid body shapes has no effect on particle collision.
- Concurrency issues with multi compartment scene configurations.




















