mhl_water.usd keep crashing

[Barzol]

Hi, when i open mhl_water.usd in the Ocean_assets folder it keep crashing. It crashes both when I open it normally both when I open it with Ocean sim extension. This is the error message in the console :

[25.905s] [ext: OceanSim-1.0.0] startup
2025-05-17 13:11:42 [25,836ms] [Warning] [omni.isaac.ui.menu] omni.isaac.ui has been deprecated in favor of isaacsim.gui.components. Please update your code accordingly.
User USD scene is loaded.
Module omni.warp.nodes._impl.kernels.grid_create d82652b load on device 'cuda:0' took 1.20 ms (cached)
Module omni.warp.nodes._impl.OgnSampleOceanDeform d02cef8 load on device 'cuda:0' took 1.54 ms (cached)
2025-05-17 13:12:18 [61,783ms] [Warning] [omni.fabric.plugin] getAttributeCount called on non-existent path
2025-05-17 13:12:18 [61,783ms] [Warning] [omni.fabric.plugin] getTypes called on non-existent path
2025-05-17 13:12:21 [64,663ms] [Warning] [omni.usd] Warning: in ComputeFlattened at line 364 of C:\g\122538378\USD\pxr\usd\usdGeom\primvar.cpp -- For primvar attribute 'primvars:normals' on 'Mesh' prim </World/rob/Mesh/Mesh> on stage with rootLayer @anon:000002096A54BA10:World1.usd@, sessionLayer @anon:000002096A5473F0:World1-session.usda@: Found 1067648 invalid indices at positions [279397, 279398, 279399, 279400, 279401, ...] that are out of range [0,275169).

2025-05-17 13:12:21 [64,667ms] [Warning] [omni.hydra] Mesh '/World/scene/Plane' has corrupted data in primvar 'normal': buffer size 4 doesn't match expected size 400000 in facevarying primvars
2025-05-17 13:12:21 [64,683ms] [Warning] [omni.hydra] Mesh '/World/rob/Mesh/Mesh' has corrupted data in primvar 'normal': buffer size 1 doesn't match expected size 1347045 in vertex primvars
2025-05-17 13:12:22 [65,623ms] [Warning] [omni.hydra] Mesh '/World/scene/Plane' update topology/point without updating normal, fallback to smooth normal.
2025-05-17 13:12:23 [0ms] [Warning] [carb.crashreporter-breakpad.plugin] [crash] A crash has occurred. If a debugger should be attached, please set the '/crashreporter/debuggerAttachTimeoutMs' setting to a timeout in milliseconds. This can be used to allow the crash reporter to wait for up to that long for a debugger to attach before processing or sending the crash report.
2025-05-17 13:12:23 [6ms] [Warning] [carb.crashreporter-breakpad.plugin] Launched process 'py-spy.exe' as PID 8316
2025-05-17 13:12:23 [527ms] [Error] [carb.crashreporter-breakpad.plugin] 'py-spy.exe' exited with code 1

[haoyuma2002814]

Regarding this issue, I would recommend trying this usd on a more performant computer.

The caustic simulation is not our contribution but a built in function in RTX renderer. We put together this caustic simulation with warp kernel to simulate water caustic but we can not guarantee the stability as noted in our documentation.

This usd file has been tested on our end, and a crash issue is typically related to the Isaac Sim internal code.

Therefore, I am sorry that using a more performant computer is so far the only solution I can think of. Please share your pc specification if you want me to look at it.

[Barzol]

Hello, thanks for the help, yeah I read about the stability of this feature.
This is my setup :

Processor : 12th Gen Intel(R) Core(TM) i5-12400F 2.50 GHz
Installed RAM : 16.0 GB (15.8 GB usable)
Storage : 466 GB SSD SHPP41-500GM, 466 GB SSD Samsung SSD 860 EVO 500GB, 932 GB HDD WDC WD10EZEX-75WN4A0
Graphics Card : NVIDIA GeForce RTX 4060 Ti (8 GB)
System Type : 64-bit OS, x64-based processor

Probably Isaac Sim requires more RAM? I'm not so familiar with the software, i'm using it for my bachelor thesis on underwater robotics simulation

[haoyuma2002814]

I think yours is also a laptop. Your GPU memory should also be larger.

Simulating Caustic and sonar do take many GPU resources.

[Barzol]

This one i sent you it's a desktop

Ok then I'll try with a different PC, thank you!