Daily Digest
Daily brief
Loading brief…
Issues & signals
- Kernel requests too little shared memory when it moves a tile returned by a `wp.func` to shared storagegithub:NVIDIA/warp11/11/2026crashes-stabilitywarpcudashared-memorytilekernelillegal-memory-access
Warp CUDA kernels that call a wp.func returning a tile can under-request dynamic shared memory when the tile is later moved to shared storage (e.g., via wp.tile_broadcast). This can silently corrupt results or trigger illegal memory accesses depending on other
⚡ Correct CUDA shared-memory sizing in Warp kernels is critical for trustworthy high-performance simulation and differentiable workloads in Physical AI pipelines.
- [BUG] Actuator gradients are wrong with several steps on one tapegithub:newton-physics/newton11/11/2026crashes-stabilitynewton-physicswarpautodiffgradientstapeactuator
Newton Actuator steps reuse internal buffers, so recording multiple steps on one wp.Tape causes later steps to overwrite data needed for earlier backward passes. This produces incorrect gradients for multi-step tapes, especially when nonlinear components are i
⚡ Correct differentiable physics gradients are essential for training and optimization loops that underpin Physical AI control and design workflows.
- Custom AGX Orin Carrier board stuck at UEFInvidia-forum:robotics-edge-computing11/11/2026hardware-integrationjetson-agx-orincarrier-boarduefibootbringuphardware-debug
A custom Jetson AGX Orin carrier board is stuck at UEFI. This prevents booting into the OS and blocks any further software setup or deployment work.
⚡ Robust bring-up on custom carrier boards is key to making NVIDIA edge platforms viable for real-robot deployments.
- [BUG] wp.Mesh crashes on out-of-range triangle indicesgithub:NVIDIA/warp11/10/2026crashes-stabilitywarpmeshsegfaultbounds-checkinggeometrycpu
Warp wp.Mesh() does not bounds-check triangle indices and can segfault on CPU with out-of-range or negative indices. Mesh.refit() and assigning Mesh.points can also crash because they read through indices without validation.
⚡ Hard crashes from invalid geometry inputs erode developer confidence and complicate productionizing simulation and geometry pipelines.
- [REQ] Make wp.sparse.bsr_mm differentiable (tape.backward currently crashes)github:NVIDIA/warp11/10/2026training-infrawarpsparsebsr-mmautodifftapecrash
Recording warp.sparse.bsr_mm on a Tape causes tape.backward() to crash due to indexedarray lacking .grad. Even without the crash, kernels are compiled with enable_backward=False, so gradients would be missing.
⚡ Differentiable sparse linear algebra is foundational for scalable optimization and learning pipelines that rely on structured physics and constraints.
- Lossless H264/H265 (NV24/4:4:4) fails on high-entropy frames due to hardcoded 1.5*W*H buffer cap in libtegrav4l2/libnvvideonvidia-forum:robotics-edge-computing11/10/2026crashes-stabilitynvidia-forumrobotics-edgeh264h265lossless-videonv24
Lossless H264/H265 in NV24/4:4:4 fails on high-entropy frames due to a hardcoded 1.5*W*H buffer cap in libtegrav4l2/libnvvideo. This suggests certain frames exceed the cap, causing encode/decode failure for otherwise valid lossless use cases.
⚡ Hardcoded multimedia limits in edge platform libraries can break critical perception/video workflows and block deployment of high-fidelity sensor streams.
- Newton backend: no supported path to run-to-run determinism with contacts (6.2.0.1)github:isaac-sim/IsaacSim11/10/2026sim2realv6.2.0.1isaac-simisaac-sim-6-2-0-1newtondeterminismcontactswarp
Isaac Sim 6.2.0.1 bundles Newton 1.6.1, which predates deterministic collision pipeline and solver APIs, leaving no supported way to get run-to-run determinism with contacts enabled. The issue notes that setting `wp.config.deterministic = RUN_TO_RUN` fails to
⚡ Lack of determinism with contacts blocks reliable regression testing and reproducible training/evaluation, weakening the platform’s credibility for physics-based Physical AI.
- [Bug Report] A depth-only RTX camera makes headless RGB cameras render blackgithub:isaac-sim/IsaacLab11/10/2026renderingisaac-labheadlessrtx-camerasynthetic-datargbdepth
In headless runs, creating a depth-only RTX camera causes all RTX RGB cameras to return black frames in the same process. The report attributes this to `IsaacRtxRenderer.create_render_data` setting a process-wide `/rtx/sdg/force/disableColorRender=True` and it
⚡ Process-wide renderer state leaks in headless mode can silently corrupt synthetic data generation and perception training pipelines at scale.
- ParticleField3DGaussianSplat turns one flat colour after Stop, then ERROR_DEVICE_LOST (Isaac Sim 6.0.1)github:isaac-sim/IsaacSim11/10/2026crashes-stabilityv6.0.1isaac-simgaussian-splattingparticlefield3dgaussiansplatrenderingdevice-lostplay-stop-cycle
ParticleField3DGaussianSplat without a bound material can turn into a flat color after repeated Play/Stop cycles and later crash with ERROR_DEVICE_LOST. The issue reproduces using the public flowers_1 asset and appears tied to repeated Gaussian prim reload beh
⚡ Rendering stability issues that lead to device loss can halt simulation workflows and reduce confidence in Isaac Sim as a production-grade platform.
- [MJX] mjOption.disableactuator is ignored: actuators in a disabled group still produce forcegithub:google-deepmind/mujoco11/10/2026othermujocomjxactuatorsoptions-bitfieldphysics-parityjax
MJX ignores mjOption.disableactuator, so actuators in disabled groups still produce forces and update activation. This deviates from MuJoCo C behavior where disabled actuators are skipped in multiple engine stages.
⚡ Parity gaps between reference physics engines and accelerated backends can compromise reproducibility for Physical AI training and evaluation pipelines.
- Loaded APIC graphs crash on CPU and return zeros on CUDA when a kernel samples a texturegithub:NVIDIA/warp11/10/2026crashes-stabilitywarpapicgraph-captureserializationtexture2dcuda
APIC graphs that include texture sampling save successfully but fail after reload in a new process: CUDA returns zeros and CPU crashes. The loaded graph only works when the original texture is still alive in the same process.
⚡ Graph serialization correctness is key for reproducible, portable compute pipelines that underpin Physical AI simulation and training workloads.
- Loaded APIC graph crashes when a kernel reads through a struct array's `wp.array` fieldgithub:NVIDIA/warp11/10/2026crashes-stabilitywarpapicgraph-captureserializationstruct-arraysillegal-memory-access
APIC graphs that read through a struct array's wp.array field save but crash after reload in a new process (CUDA illegal memory access, CPU segfault). The graph only behaves correctly when loaded in the same process while original arrays remain alive.
⚡ Reliability of saved/loaded compute graphs affects deployability of accelerated components used across Physical AI workflows.
- Jetson orin nano 8gb SDRAM broken,hardware problemnvidia-forum:robotics-edge-computing11/10/2026hardware-integrationjetson-orin-nanosdramhardware-failurermarobotics-edge
User reports Jetson Orin Nano 8GB SDRAM is broken and frames it as a hardware problem, with no further details. This indicates a potential device failure preventing use.
⚡ Hardware reliability incidents on Jetson platforms can directly block edge deployment of Physical AI applications.
- Register and indexed tile operations produce wrong results for padded vector and matrix arraysgithub:NVIDIA/warp11/10/2026crashes-stabilitywarpcudacpu-backendmemory-alignmentvec3mat33
Register and indexed tile load/store operations return incorrect values for padded vec3 and mat33 arrays on both CPU and CUDA. This breaks correctness when data uses common alignment padding (e.g., 16-byte stride for vec3).
⚡ Warp correctness bugs in core memory/tile primitives directly undermine trust in GPU-accelerated simulation and learning workloads built on NVIDIA’s Physical AI stack.
- Gradients of wp.frac() and vector wp.mod() are always zerogithub:NVIDIA/warp11/9/2026training-infrawarpautogradgradientsfracmoddifferentiable-sim
Backward gradients through wp.frac() and vector wp.mod() are always zero, with no error or warning. This silently produces incorrect gradients away from discontinuities where the derivative should be non-zero.
⚡ Broken autodiff for common math ops reduces Warp’s usefulness for differentiable simulation and gradient-based training central to Physical AI.
- Xavier NX system crush and auto rebootnvidia-forum:robotics-edge-computing11/9/2026crashes-stabilityxavier-nxcrashauto-rebootstabilityedge-deployment
A Xavier NX system is reported to crash and auto reboot. The lack of details indicates an early-stage but potentially blocking reliability issue for that platform.
⚡ Unexplained crashes and reboots on deployed Jetson hardware directly threaten uptime for Physical AI edge systems.
- [BUG] Box contacts against mesh triangles can be centimetres deep and sideways since #4974github:newton-physics/newton11/9/2026crashes-stabilitynewtoncollisioncontactsregressionmesh-trianglebox-contact
After a specific contact-normal change, mesh triangle vs box contacts can be reported as centimeters deep with sideways normals when they should be shallow and upward. The regression is reported on both main and release branches and changes contact depth/norma
⚡ Physics contact regressions create non-physical behavior that can invalidate simulation-based testing and training in NVIDIA’s Physical AI workflows.
- NVIDIA Jetson AGX Thor – Boot Failure at Elevated Temperature Despite Operating Within Specified Thermal Limits (JetPack 7.2)nvidia-forum:robotics-edge-computing11/9/2026crashes-stabilityjetsonagx-thorboot-failurethermalstabilityjetpack-7-2
Jetson AGX Thor reportedly fails to boot at elevated temperature even while operating within specified thermal limits, on JetPack 7.2. This prevents reliable field operation and may indicate a platform stability or firmware issue.
⚡ Thermal-robust boot reliability is foundational for Physical AI edge deployments in real-world environments.
- Integrating a hierarchical cognitive controller with Unitree A1 locomotiongithub:isaac-sim/IsaacLab11/9/2026training-infrarllocomotionintegrationfeature-requestisaac-simisaac-lab
Feature request to integrate a hierarchical cognitive controller with Unitree A1 locomotion in Isaac Lab. The issue template indicates a proposal rather than a bug report, with missing details like versions to be filled in.
⚡ Supporting modular hierarchical controllers alongside locomotion stacks would broaden Isaac Lab’s role as a platform for higher-level Physical AI decision-making integrated with motor control.
- Orin Nano Super USB-C only enumerates in recovery, with INA3221/FUSB301 I2C probe errorsnvidia-forum:robotics-edge-computing11/9/2026hardware-integrationjetson-orin-nanousb-crecovery-modei2cina3221fusb301
Orin Nano Super USB-C reportedly only enumerates in recovery mode, accompanied by INA3221/FUSB301 I2C probe errors. If accurate, this blocks normal USB-C operation and potentially device provisioning/debug workflows.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Jetson Orin Nano Developer Kit No-Boot / Dead Board (P/N 945-13766-0000-000)nvidia-forum:robotics-edge-computing11/8/2026hardware-integrationjetson-orin-nanodeveloper-kitno-bootdead-boardhardware-failurerma
User reports a Jetson Orin Nano Developer Kit that will not boot / appears to be a dead board, referencing a specific part number. This prevents any progress on setup or development.
⚡ Reliability of Jetson dev kits is a gating factor for getting Physical AI stacks into developers’ hands and into real-world prototyping loops quickly.
- Orin NX 16GB gets stuck on the system boot screennvidia-forum:robotics-edge-computing11/8/2026hardware-integrationjetsonorin-nxboot-failuresystem-bootrobotics-edgedevice-bringup
Orin NX 16GB reportedly gets stuck on the system boot screen. This prevents the device from booting normally and blocks any development or deployment on the platform.
⚡ Edge device reliability is foundational; boot failures on Jetson-class systems directly block deploying Physical AI applications and running on-device stacks.
- [Bug Report] Startup loading screen raises UnicodeEncodeError (cp932) when stdout is redirected on Japanese Windowsgithub:isaac-sim/IsaacLab11/8/2026renderingv3.0.0renderinghardwareisaac-simisaac-labmujoconewton
Isaac Lab training aborts on Japanese Windows when stdout is redirected due to a UnicodeEncodeError in the loading screen output. The middle dot character in the run-summary title cannot be encoded with cp932, so training never starts.
⚡ Developer experience issues in common OS/locale setups can stop training runs outright, reducing Isaac Lab adoption in global teams and CI environments.
- Gstreamer failing to launch with Nvcompositornvidia-forum:robotics-edge-computing11/6/2026crashes-stabilitygstreamernvcompositormultimediavideo-pipelineruntime-failure
GStreamer fails to launch when using Nvcompositor. This blocks building or running a media pipeline that depends on NVIDIA's compositor element, implying a driver/plugin/config compatibility issue.
⚡ Stability and reliability of the multimedia stack are foundational for edge Physical AI applications that depend on real-time video composition and streaming.
- [BUG] SolverImplicitMPM: yielding material loses volume under shear (Q1/P0), independent of max_iterations; P1d/pic8 lockgithub:newton-physics/newton11/5/2026renderingrenderinghardwarenewtonwarp
SolverImplicitMPM shows unphysical volume loss under plastic shear for a yielding material with the reported Q1/P0 configuration, and it does not improve with more iterations. The reporter notes this appears distinct from a known convergence issue referenced a
⚡ Physics correctness in deformables/MPM is essential for simulation credibility in downstream robotics and autonomy workflows.
- Let `tile_cholesky_inplace()` skip zeroing the opposite trianglegithub:NVIDIA/warp11/5/2026renderingrenderinghardwaremujoconewtonwarp
Feature request for `wp.tile_cholesky_inplace()` to optionally skip zeroing the opposite triangle after factorization, since downstream solves read only the selected triangle. Current behavior adds an extra pass over the whole tile for cleanliness, potentially
⚡ Kernel-level performance knobs in Warp can materially improve end-to-end training and optimization throughput for Physical AI workloads.
- Isaac Sim 6.2 Kit fast_importer selects mixed charset_normalizer Python/native modules, causing 41 extension startup failuresgithub:isaac-sim/IsaacSim11/5/2026crashes-stabilityv6.2crashrlrenderinghardwaredeploymentintegration
On Linux pip installs of Isaac Sim 6.2.0.0, Kit fast_importer can load a mixed Python and native implementation of charset_normalizer depending on directory enumeration order. This triggers binary-layout exceptions and cascades into 41 extension startup failur
⚡ Python packaging determinism and module loading robustness directly impact Isaac Sim reliability as a core dev platform for Physical AI.
- L4T R39.2 NVDEC: SIGSEGV in libnvvideo.so when current picture has no frame buffer (NvVideoDecoderDecode called with NULL+0x310)nvidia-forum:robotics-edge-computing11/5/2026crashes-stabilityl4t-r39-2nvdecvideo-decodesigsegvlibnvvideojetson
On L4T R39.2, NVDEC triggers a SIGSEGV in libnvvideo.so when the current picture has no frame buffer. The crash occurs in NvVideoDecoderDecode when called with NULL+0x310, breaking decode reliability.
⚡ Stable hardware-accelerated video decode on Jetson is critical for edge perception pipelines in Physical AI deployments.
- Issue where HSB multi-camera transmission terminates unexpectedly with an errornvidia-forum:robotics-edge-computing11/4/2026sensors-perceptionrobotics-edge-computingmulti-cameracamera-streaminghsbtransport-errors
HSB multi-camera transmission is terminating unexpectedly with an error. This blocks sustained multi-camera streaming needed for perception pipelines.
⚡ Unreliable multi-camera transport breaks the sensor stack needed for perception and closed-loop autonomy on edge deployments.
- Timeout issue when executing hololink.start()nvidia-forum:robotics-edge-computing11/4/2026integrationrobotics-edge-computinghololinkstartuptimeoutsintegration
Calling hololink.start() hits a timeout. This prevents bringing up the link/streaming path needed to run the system.
⚡ Startup timeouts in key link/stream APIs increase integration risk for end-to-end Physical AI sensor-to-compute pipelines.
- SolverSteel with deterministic=RUN_TO_RUN fails inside a CUDA graph capture that Warp didn't startgithub:newton-physics/newton11/4/2026integrationnewtonwarpcuda-graphsdeterminismtorch-integrationsolversteel
With deterministic=RUN_TO_RUN, SolverSteel cannot be captured into a CUDA graph unless Warp initiated the capture. External capture mechanisms (e.g., cuStreamBeginCapture or torch.cuda.graph) get invalidated mid-step.
⚡ Incompatibility with external CUDA graph capture limits composability of Warp-based physics with broader PyTorch and CUDA ecosystem workflows.
- [BUG] add_mjcf silently drops <sensor> elements (e.g. <touch>)github:newton-physics/newton11/4/2026sensors-perceptionnewtonmjcfmodel-importsensorstouch-sensorsolver-mujoco
ModelBuilder.add_mjcf silently ignores the MJCF <sensor> section, so sensors like <touch> never appear in the built model. The same XML loaded directly with MuJoCo includes the sensor, indicating a loss during import without warnings.
⚡ Silent drops in sensor import pipelines erode trust in simulation-to-control parity, weakening sim-based validation and data generation workflows.
- Jetson Thor (L4T 39.2.1): Binding MGBE to vfio-platform for DPDK fails (-EINVAL / Error 22) – Missing dma-coherent on SMMU0nvidia-forum:robotics-edge-computing11/5/2026▲ 4hardware-integrationhardware
On Jetson Thor with L4T 39.2.1, binding the MGBE device to vfio-platform for DPDK fails with -EINVAL (Error 22). The post indicates a missing dma-coherent setting on SMMU0, blocking the intended userspace driver path.
⚡ DPDK/vfio enablement on Jetson is key to deterministic, high-throughput edge pipelines that Physical AI applications depend on.
- IGX700 BMC – Account Lockout After Multiple Failed Login Attemptsnvidia-forum:robotics-edge-computing11/10/2026▲ 1hardware-integrationigx700bmcaccount-lockoutremote-managementedge-deployment
IGX700 BMC accounts can lock out after multiple failed login attempts, potentially blocking remote management access. The issue suggests an operational friction point around access policy and recovery procedures.
⚡ Manageability and recoverability issues in IGX platform administration increase operational risk for edge Physical AI deployments.
- Hardware timestamping on RTL8126 failsnvidia-forum:robotics-edge-computing11/11/2026hardware-integrationjetsonnicrtl8126hardware-timestampingptpedge-computing
Hardware timestamping fails on the RTL8126 network interface. This blocks or degrades time-synchronized networking workflows that depend on NIC timestamp support.
⚡ Reliable hardware timestamping is foundational for synchronized sensing and control loops across edge robotics deployments.
- Jetson Thor USB Host Controller Resource Limit with Multiple CAN Adaptersnvidia-forum:robotics-edge-computing11/11/2026hardware-integrationjetson-thorusbhost-controllercan-buscan-adapterresource-limit
Jetson Thor USB host controller hits a resource limit when using multiple CAN adapters. This prevents scaling to multiple CAN devices on the same system without changes or workarounds.
⚡ I/O scalability limits on edge platforms directly constrain real-robot sensor/actuator integration for Physical AI applications.
- HSB SDK v3.0.0: How to access data in a custom Python operator?nvidia-forum:robotics-edge-computing11/11/2026docs-onboardinghsb-sdkpython-operatorcustom-operatordata-accesssdk-usagerobotics-edge
A user asks how to access data in a custom Python operator in HSB SDK v3.0.0. This suggests missing or unclear documentation/API patterns for operator I/O and data handling.
⚡ Better operator data-access guidance reduces friction when extending NVIDIA edge/robotics SDK pipelines used in Physical AI products.
- multicontact: GJK/MPR manifold scratch buffers rely on CUDA-only fixedarray idiomsgithub:newton-physics/newton11/11/2026integrationnewton-physicswarpmulticontactcuda-specificmetal-backendportability
Newton multicontact manifold scratch buffers rely on CUDA-only fixedarray and pointer aliasing behaviors. This breaks on backends with strict address spaces (e.g., Apple Metal) because it treats thread-local stack storage as a device pointer.
⚡ Reducing backend-specific assumptions in core physics/contact code increases portability across GPU backends needed by diverse Physical AI developer environments.
- Nvidia,disable-power-downnvidia-forum:robotics-edge-computing11/11/2026hardware-integrationjetsonpower-managementdevice-treedisable-power-downplatform-configedge-computing
The topic references "Nvidia,disable-power-down", indicating a need to disable a power-down behavior via NVIDIA-related configuration. Missing context suggests unclear device-tree or platform settings for power management.
⚡ Power-management controls that are hard to configure can destabilize real-time robotics behavior and complicate deployment on NVIDIA edge devices.
- How can we run inference of Alpamayo 2 super on jetson AGX thor?nvidia-forum:robotics-edge-computing11/11/2026deploymentjetson-agx-thorinferencedeploymentmodel-compatibilityedge-airuntime
A user asks how to run inference for "Alpamayo 2 super" on Jetson AGX Thor. This indicates uncertainty about model support, optimization, or runtime deployment steps on that platform.
⚡ Running modern model inference reliably on Jetson-class hardware is central to deploying Physical AI capabilities on robots.
- <model> asset in an included file is looked up with the directory twice when the main file is loaded as dir/scene.xmlgithub:google-deepmind/mujoco11/10/2026asset-pipelinemujocomjcfasset-resolutionincludepath-bugrobot-models
MuJoCo incorrectly resolves a <model> asset file path in an included file when the main MJCF is loaded from a subdirectory, duplicating the directory in the resolved path. This causes model loading to fail with a file-not-found error like 'case2/case2/sub/chil
⚡ Consistent asset path resolution is important for reliable robot model composition workflows that feed simulation-to-training pipelines.
- to_xml() drops mass="0" density="0" from a geom default class, so the saved XML compiles to a heavier bodygithub:google-deepmind/mujoco11/10/2026asset-pipelinemujocomjcfxml-exportmodel-serializationmass-propertiesrobot-arms
MuJoCo's to_xml() drops explicit mass="0" and density="0" on geom defaults, changing the exported model's physical properties. The re-saved XML compiles with default density and yields a much heavier body than the original.
⚡ Simulation-to-asset export fidelity issues directly undermine trust in model interchange and sim2real validation workflows.
- Support wp.handle in framework and DLPack convertersgithub:NVIDIA/warp11/10/2026integrationwarpdlpackjaxpytorchpaddlehandle-dtype
Warp converters (from_torch/from_jax/from_paddle/from_dlpack) do not support explicit dtype=wp.handle imports/exports using 64-bit integer storage. Users want handle arrays to round-trip through DLPack and framework converters while preserving underlying bits
⚡ Robust interop (Torch/JAX/Paddle/DLPack) is a key enabler for Physical AI pipelines that mix simulation, learning, and GPU compute libraries.
- wp.sqrt() gradient is inf at x == 0, unlike wp.length()github:NVIDIA/warp11/10/2026training-infrawarpautodiffgradientsnumerical-stabilitysqrtnan
Warp's wp.sqrt() adjoint computes 0.5/sqrt(x) with no guard, producing inf gradients at x==0 that often become NaNs downstream. This differs from wp.length() behavior and can break common patterns like sqrt(dot(v,v)) at v==0.
⚡ Numerical stability in autodiff primitives is critical for reliable gradient-based optimization and learning workloads in Physical AI.
- [REQ] Make wp.sparse.bsr_mv differentiablegithub:NVIDIA/warp11/10/2026training-infrawarpsparsebsr-mvautodifftapegradients
warp.sparse.bsr_mv is not differentiable with wp.Tape because kernels are built with enable_backward=False. tape.backward() completes but returns all-zero gradients for x and A.values with only a warning, requiring users to implement their own adjoint.
⚡ Missing gradients in sparse ops force users into custom adjoints, slowing adoption of Warp as a learning/optimization backend for Physical AI.
- Deterministic contacts and contact matching are limited to 2**20 shapes with meshes, heightfields or hydroelastic shapesgithub:newton-physics/newton11/10/2026crashes-stabilitynewton-physicsdeterminismcontactscollisionmeshesheightfields
Newton's deterministic contacts/contact matching use an int64 sort key that limits models with mesh/heightfield/hydroelastic contacts to at most 2**20 shapes. Exceeding that now raises ValueError when deterministic mode is enabled (directly or via contact_matc
⚡ Determinism limits at large scale constrain reproducible simulation and evaluation for big scene generation and multi-object Physical AI workloads.
- Box-triangle contact is 1 m deep for a triangle nearly parallel to a thin box face (after #5241)github:newton-physics/newton11/10/2026crashes-stabilitynewton-physicscollisioncontact-depthregressionbox-trianglethin-geometry
After a fix in #5241, Newton now produces an incorrect ~1 m deep box-triangle contact when a triangle is nearly parallel to a thin box face. The same configuration previously produced a ~1.4 mm contact, indicating a regression in narrowphase contact depth hand
⚡ Contact robustness regressions can destabilize physics-based training and evaluation, especially for thin geometries common in real-world assets.
- OVRTX periodic cables do not follow Newton simulationgithub:isaac-sim/IsaacLab11/10/2026crashes-stabilityisaac-labnewtonovrtxcablesusdgeombasiscurves
Periodic `UsdGeom.BasisCurves` cables render but stay at their authored pose instead of following the simulated cable when using Newton + OVRTX. Open cables update correctly, indicating a specific issue with periodic curves not updating with simulation.
⚡ Simulation/visualization mismatches undermine trust in physics-based workflows and reduce the value of Isaac Lab as a debugging and development environment.
- [BUG] SolverXPBD zeroes body velocities below 1e-4, so slow bodies stop or never start movinggithub:newton-physics/newton11/10/2026othernewton-physicsxpbdrigid-body-dynamicssolver-stabilitynumerical-thresholdsvelocity-clamping
SolverXPBD clamps small linear/angular velocities to zero every iteration, causing slow free bodies to stop and preventing bodies from ever starting to move when per-step changes are below the threshold. This breaks expected behavior versus other solvers and r
⚡ Physics-solver correctness issues at small velocity scales can undermine trust in simulation fidelity used for Physical AI evaluation and training.
- [BUG] Hollow mesh mass properties ignore the requested shell thicknessgithub:newton-physics/newton11/10/2026othernewton-physicsmass-propertiesinertiameshasset-physicshollow-shell
Hollow mesh mass properties ignore the requested shell thickness and reuse cached defaults. The scaling uses volume (s³) instead of thin-shell area (s²), producing incorrect inertia/mass properties for scaled hollow meshes.
⚡ Accurate mass/inertia computation for assets is foundational for reliable simulation behavior that Physical AI stacks depend on.
- Orin Nano Super, JetPack 6.2.1: CPU at 95 % while GPU clock stays below max during TensorRT inference on 4 RTSP streamsnvidia-forum:robotics-edge-computing11/10/2026deploymentjetson-orin-nanojetpack-6-2-1tensorrtrtspvideo-inferenceperformance-bottleneck
On Orin Nano Super with JetPack 6.2.1, TensorRT inference on four RTSP streams shows CPU near 95% while GPU clocks stay below max. This indicates a bottleneck outside GPU compute (e.g., decode/IO/scheduling) that limits throughput.
⚡ Edge inference performance predictability (CPU/GPU utilization) is critical for scalable Physical AI deployments on Jetson/Orin.
- [BUG] ModelBuilder.collapse_fixed_joints gives the merged body a wrong inertiagithub:newton-physics/newton11/10/2026othernewton-physicsmodelbuilderfixed-jointsinertiarigid-body-dynamicsarticulation
collapse_fixed_joints merges masses and COM correctly but computes a wrong composite inertia about the merged COM. The child inertia is shifted using the child frame origin instead of its COM, and the parent inertia is not shifted to the merged COM.
⚡ Incorrect inertia composition in model-building tools can silently corrupt dynamics, reducing reliability of simulation-based Physical AI development.
- [BUG] Aggressive viewer picking response on a lightweight branched harness with SolverVBDgithub:newton-physics/newton11/10/2026tooling-dxnewtonwarpviewerpickingwiring-harnesssolvervbd
Viewer picking causes overly aggressive forces near branch junctions, making a lightweight wiring harness fold, tumble, and oscillate excessively. The behavior is reported with SolverVBD and takes a long time to settle after release.
⚡ Unstable interactive manipulation in physics viewers slows iteration and debugging for robotics and digital-twin workflows on NVIDIA’s platform.
- Bazel: downstream modules fail on mujoco_pip for Python 3.15; please bump rules_python to 2.5.0github:google-deepmind/mujoco11/9/2026integrationmujocobazelrules-pythonpython-3-15build-integrationmodule-deps
Downstream Bazel modules fail when using mujoco_pip for Python 3.15 due to rules_python version pinning. The request is to bump rules_python to 2.5.0 to restore pip toolchain/interpreter resolution.
⚡ Build-system breakages in popular simulators compound ecosystem fragmentation and raise the cost of integrating Physical AI components into Bazel-based stacks.
- Why is π0.5 inference about 5× slower under T3000 software emulation on Jetson Thor?nvidia-forum:robotics-edge-computing11/9/2026deploymentjetson-thorsoftware-emulationinferenceperformancet3000edge-ai
π0.5 inference is reported to run about 5× slower under T3000 software emulation on Jetson Thor. This raises questions about expected performance characteristics and emulation fidelity.
⚡ Large performance gaps under software emulation complicate benchmarking and capacity planning for running Physical AI models on Jetson Thor.
- Intermittent Slow Suspend (~1 in 500 cycles) on Jetson Orin Nano (JetPack 7.2)nvidia-forum:robotics-edge-computing11/9/2026crashes-stabilityjetson-orin-nanojetpack-7-2suspendpower-managementintermittentstability
Jetson Orin Nano shows intermittent very slow suspend behavior occurring roughly 1 in 500 cycles. The report is tied to JetPack 7.2 and suggests a sporadic stability/latency problem in power management.
⚡ Rare-but-severe power-state latency issues erode reliability expectations for always-on robotics systems deployed on Jetson.
- wp.autograd.jacobian() returns zero gradients for kernels whose adjoint reads an output (atomic_min/atomic_max)github:NVIDIA/warp11/9/2026training-infrawarpautogradjacobiangradcheckatomic-minatomic-max
wp.autograd.jacobian() returns zero gradients for kernels using atomic_min/atomic_max when the adjoint reads an output, while wp.Tape produces correct gradients. gradcheck fails without Warp reporting an error or warning.
⚡ Inconsistent gradient tooling results weaken developer confidence in Warp-based differentiable pipelines for Physical AI training and verification.
- Jetson Thor T5000 PCIe C2 width confirmation: Data Sheet / Design Guide say x1, Linux Developer Guide says x2 (Lane4+Lane5) — which is correct?nvidia-forum:robotics-edge-computing11/9/2026docs-onboardingjetson-thort5000pcielane-widthdocumentationhardware-design
Jetson Thor T5000 PCIe C2 lane width is inconsistent across documents (data sheet/design guide vs Linux Developer Guide). The user is asking which PCIe width is correct (x1 vs x2).
⚡ Conflicting hardware interface documentation increases integration errors and slows partner designs for Jetson Thor-based Physical AI devices.
- [REQ] Reserve less memory by default for rebuildable volumesgithub:NVIDIA/warp11/9/2026feature-requestswarpvolumesnanogridmemory-usagesparse-gridsfem
Rebuildable Warp volumes reserve worst-case memory by default, leading to very large per-point allocations even for dense inputs. The request is to reserve less memory by default for allocate_by_voxels/allocate_by_tiles and related paths (e.g., fem.Nanogrid.fr
⚡ Excess default memory reservation limits scale and density of 3D world representations, impacting Physical AI simulation and data generation throughput.
- [BUG] Rebuildable volumes crash when their reserved memory can't be allocatedgithub:NVIDIA/warp11/9/2026crashes-stabilitycrashrenderinghardwarewarp
Warp rebuildable volumes can crash when reserved memory allocation fails because allocation success is not checked. Failures lead to CPU segfaults or CUDA illegal memory access that breaks the context, and rebuildable volumes can hit this due to extreme defaul
⚡ Hard crashes from allocation failures undermine reliability for large-scale synthetic worlds and geometry processing pipelines in Physical AI workloads.
- omni.replicator.core: orchestrator.step() drives four full renders per capture and keeps onegithub:isaac-sim/IsaacSim11/9/2026renderingv6.1.0renderinghardwaresensorsperceptiondocsintegration
In Isaac Sim Replicator, a single orchestrator.step() triggers four full application updates and renders per capture while only keeping one image. With path tracing where each update converges, this wastes most render time as discarded pixels.
⚡ Inefficient capture orchestration directly increases synthetic data generation cost and slows dataset throughput for training Physical AI models.
- [BUG] capture_debug_dot_print() fails on CPU graphs and on loaded graphs before their first launchgithub:NVIDIA/warp11/9/2026renderingrenderinghardwarewarp
wp.capture_debug_dot_print() throws an unrelated CUDA invalid-argument error when called on CPU-captured graphs or on graphs loaded before first launch. The function should either support these cases or fail with a clear, accurate error message.
⚡ Warp graph tooling reliability directly impacts developer trust and iteration speed for physics/robotics workloads that depend on capture/replay.
- [BUG] bsr_compress() isn't replayed by CPU graph captures or kept in saved graphsgithub:NVIDIA/warp11/9/2026crashes-stabilitycrashhardwareintegrationwarp
Graph capture does not record wp.sparse.bsr_compress() for CPU captures, so it runs during recording but not during replay, producing wrong or uncompressed matrices and can crash when capturing fem.integrate() with row_compress. CUDA graphs run it, but capture
⚡ Correct capture/replay semantics for sparse/FEM operations are critical to deterministic simulation and scalable training pipelines built on Warp graphs.
- Retain allocations made by callable CUDA conditional graph bodiesgithub:NVIDIA/warp11/9/2026hardware-integrationhardwareintegrationnewtonwarp
Callable CUDA conditional graph bodies can allocate arrays or internal scratch during capture, but CUDA disallows allocation nodes inside these bodies and the memory must remain valid for replay even if Python arrays are released. The proposal is to retain cap
⚡ Robust CUDA graph capture for dynamic control flow and memory lifetimes is key to high-performance simulation/training loops that must avoid host-side overhead.
- USB OTG Compatibility Issue with USB 3.0 Hubs on NVIDIA Jetson AGX Orin – JetPack 6.2nvidia-forum:robotics-edge-computing11/9/2026hardware-integrationjetsonagx-orinusb-otgusb-3-hubperipheralsjetpack-6-2
User reports USB OTG compatibility issues with USB 3.0 hubs on Jetson AGX Orin running JetPack 6.2. This can break device connectivity assumptions for cameras, storage, or debug peripherals.
⚡ Peripheral compatibility on edge platforms directly affects integration velocity for robotics prototypes and deployments.
- NVMe SSD Flashing Issue on NVIDIA Jetson AGX Xaviernvidia-forum:robotics-edge-computing11/9/2026deploymentjetsonagx-xaviernvmeflashingprovisioningdeployment
User reports an NVMe SSD flashing issue on Jetson AGX Xavier. This blocks or complicates installing and provisioning the target system on NVMe storage.
⚡ Reliable flashing and storage provisioning are prerequisites for scaling edge deployment of Physical AI systems.
- [Thor]SW Shutdown Case & Resetnvidia-forum:robotics-edge-computing11/9/2026hardware-integrationjetsonthorshutdownresetpower-managementembedded
User asks about a software shutdown case and reset behavior for Jetson Thor. This indicates uncertainty or unexpected behavior around power management and reset flows.
⚡ Well-defined power/reset/shutdown behavior is essential for robust embedded robotics systems using NVIDIA edge modules.
- NvBufSurfTransform (VIC) alters chroma on same size NV12 to NV12 copy, luma kept exact, NV16/NV24 and GPU compute kept exactnvidia-forum:robotics-edge-computing11/9/2026sensors-perceptionjetsonnvbufsurftransformvicnv12video-processingcolor-chroma
User reports that NvBufSurfTransform (VIC) changes chroma on an NV12-to-NV12 copy at the same size, while luma remains exact and other formats/GPU paths remain exact. This implies a potential correctness issue in a common video processing path.
⚡ Deterministic, correct pixel transformations are critical to trustworthy vision pipelines in Physical AI deployments.
- 【Jetson Thor】Troubleshooting for OC Alarm Issuesnvidia-forum:robotics-edge-computing11/9/2026crashes-stabilityjetsonthoroc-alarmpowerstabilitytroubleshooting
User requests troubleshooting guidance for OC alarm issues on Jetson Thor. This suggests over-current or related alarm events impacting system stability or operation.
⚡ Clear diagnostics and mitigations for platform alarms are necessary to maintain system reliability in edge Physical AI deployments.
- Orin NX；hardware watchdognvidia-forum:robotics-edge-computing11/9/2026hardware-integrationjetsonorin-nxwatchdogreliabilityembeddedrecovery
User asks about a hardware watchdog on Jetson Orin NX. This implies a need for system-level reset/recovery mechanisms and clarity on hardware watchdog availability or configuration.
⚡ Reliable watchdog support is essential for autonomous Physical AI systems that must self-recover from hangs in production.
- Unable to Reproduce DROID-W Benchmark Performance from NVIDIA Workshopnvidia-forum:robotics-edge-computing11/9/2026deploymentdroid-wbenchmarkingperformance-regressionworkshopreproducibility
User reports they cannot reproduce DROID-W benchmark performance from an NVIDIA workshop. The post body is empty, so the setup, hardware/software stack, and measured deltas are unknown.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Request for USB PD VIF file for CYPD8225 (CCG8) firmware on Jetson Thor reference designnvidia-forum:robotics-edge-computing11/9/2026docs-onboardingjetson-thorusb-pdvifcypd8225ccg8reference-design
Request for a USB PD VIF file for CYPD8225 (CCG8) firmware on the Jetson Thor reference design. Lack of this artifact may block USB-PD compliance/validation or bring-up workflows for partners.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- How can OrinAGX sipl framework support 8lane configurationnvidia-forum:robotics-edge-computing11/8/2026hardware-integrationjetson-orin-agxsipl8-lanecamera-pipelineconfiguration
User asks how the Orin AGX SIPL framework can support an 8-lane configuration. The post body is empty, so the specific interface (e.g., camera/SerDes) and current limitation are unclear.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Jetson Orin NX – 3.3V GPIO Pull-up Safety During Initial Flashingnvidia-forum:robotics-edge-computing11/8/2026hardware-integrationjetson-orin-nxgpio3v3flashingbring-uphardware-safety
User is concerned about 3.3V GPIO pull-up safety during initial flashing on Jetson Orin NX. This suggests risk of improper electrical states or potential damage/misconfiguration during provisioning.
⚡ Hardware bring-up safety guidance directly impacts developer confidence and time-to-first-success on NVIDIA edge platforms used for Physical AI deployments.
- Does the SMCU (RH850) work without issues on the IGX-T5000?nvidia-forum:robotics-edge-computing11/8/2026hardware-integrationigx-t5000smcurh850compatibilityedge-computingplatform-validation
User asks whether the SMCU (RH850) works without issues on the IGX-T5000. This indicates uncertainty about component compatibility or known issues on that platform.
⚡ Verified compatibility of safety/control microcontrollers with IGX platforms is important for dependable edge robotics deployments and compliance workflows.
- Remove legacy rigid AVBD and start-origin Rod constructiongithub:newton-physics/newton11/8/2026training-infrarldxdocsnewton
Newton proposes completing deprecations by removing a legacy rigid-body formulation and making compliant ALM the only rigid-body mode in `SolverVBD`. It also removes a legacy rod body-origin convention (`start`) in favor of COM-centered frames.
⚡ Reducing legacy physics paths while keeping authored assets compatible is key to stable simulation foundations for training and testing Physical AI behaviors.
- [BUG] wp.utils array helpers run on the current CUDA device instead of their arrays' devicegithub:NVIDIA/warp11/8/2026renderingrenderinghardwarewarp
Warp `wp.utils` array helpers run CUDA work on the current device rather than the device where the arrays reside, using the wrong stream and allocating scratch memory on the wrong device. This can break graph captures or cause operations to be captured into th
⚡ Correct multi-GPU device semantics in core array utilities is essential for robust scaling and predictable CUDA graph capture in Physical AI simulation/training pipelines.
- Support `capture_launch()` inside an active CUDA graph capturegithub:NVIDIA/warp11/8/2026renderingrenderinghardwarewarp
Request that `wp.capture_launch(graph)` work inside an active CUDA graph capture on the same device, enabling nested graph replay. Currently it raises an error even though the graph can be used in `capture_if()`/`capture_while()` branches.
⚡ Composable CUDA graph execution improves performance engineering and modularity for simulation/training kernels used across NVIDIA’s Physical AI stack.
- Thor uses DP1 for display cannot worknvidia-forum:robotics-edge-computing11/8/2026hardware-integrationthordisplayportdp1video-outputrobotics-edgehardware-bringup
User reports that Thor using DP1 for display does not work. This suggests a display output/port routing or driver/firmware compatibility problem that prevents using the device with a monitor.
⚡ Display bring-up and IO compatibility issues on target hardware can block end-to-end Physical AI workflows from sim to deployed edge systems.
- [BUG] USD import silently drops `physxRigidBody:linearDamping` / `physxRigidBody:angularDamping`github:newton-physics/newton11/8/2026crashes-stabilitycrashusdhardwaremujoconewtonwarp
USD import maps PhysX damping attributes to internal keys that are never consumed, so linear/angular damping are effectively dropped without warning. This changes dynamics (e.g., damped bodies keep velocity) and can invalidate simulations when importing PhysX-
⚡ USD/PhysX fidelity gaps during import undermine cross-tool asset portability, making sim results diverge from authored physics properties.
- `v.y += x` on a local vector double-counts the gradientgithub:NVIDIA/warp11/7/2026otherwarp
Warp produces wrong gradients when using += or -= through attributes of local vectors/quaternions/transforms (e.g., w.y += x), effectively double-counting in backprop. The subscript form (w[1] += x) behaves correctly, indicating a codegen/autodiff bug.
⚡ Incorrect gradients in Warp’s autodiff can invalidate optimization and learning pipelines that underpin differentiable simulation and model-based Physical AI.
- LEAPP export disables TorchScript after external task importsgithub:isaac-sim/IsaacLab11/7/2026training-infrarlisaac-simisaac-lab
LEAPP export imports external task entry points before disabling TorchScript, so scripted helpers may compile too early. This breaks tracing tensors and can make exported targets constant, causing re-entry validation to fail unless PYTORCH_JIT=0 is set.
⚡ Export determinism and correct TorchScript/tracing behavior are key to taking trained policies from Isaac Lab into deployable Physical AI runtimes.
- `ptxas` fails on `.ptx` files written by `wp.compile_aot_module()` and the kernel cachegithub:NVIDIA/warp11/7/2026hardware-integrationhardwarewarp
Warp-written .ptx files include a trailing NUL byte that causes ptxas/nvcc to fail with 'Unexpected EOF', preventing external assembly to cubin. The same issue affects kernel-cache PTX outputs and even basic text tooling like grep.
⚡ Interoperable compilation artifacts (PTX/cubin) are important for integrating Warp kernels into larger Physical AI build and deployment toolchains.
- [BUG] SolverImplicitMPM: with the default sparse grid, a viscous film on the ground plane flows ~15 % slower than with grid_type="dense"github:newton-physics/newton11/6/2026renderingrenderinghardwarenewtonwarp
Implicit MPM produces materially different viscous flow speeds on a ground plane depending on whether the sparse or dense grid is used. The discrepancy appears tied to sparse-grid node plane offsets/translation behavior, leading to ~15% slower flow with the de
⚡ Physics solvers that change outcomes based on internal grid representations undermine trust and block adoption for high-fidelity Physical AI simulation workflows.
- wp.from_torch() and wp.from_dlpack() fail on 0-d tensors and mishandle single-vector tensorsgithub:NVIDIA/warp11/6/2026otherwarp
Warp's PyTorch and DLPack converters fail on 0‑D tensors (raising IndexError) and mishandle single-vector tensors with vector dtypes, producing unexpected strides/contiguity and gradient shape assignment errors. This breaks common interop patterns that work wi
⚡ Robust tensor interop is foundational for integrating Warp into PyTorch-based Physical AI training stacks without edge-case failures.
- [Bug Report] OvPhysX: deformable object rows are ordered lexicographically, so envs ≥ 10 read/write another env's deformablegithub:isaac-sim/IsaacLab11/6/2026crashes-stabilitycrashrenderingisaac-sim
On the OvPhysX backend, deformable object view rows are ordered lexicographically rather than by environment index once the scene has 10+ environments. As a result, reads/writes/resets for deformables can target the wrong environment, corrupting simulation sta
⚡ Multi-environment correctness bugs break scalable RL/simulation training, directly impacting throughput and reproducibility for Physical AI development.
- Thor yocto构建的A/B系统切换问题nvidia-forum:robotics-edge-computing11/6/2026deploymentthoryoctoa-b-updatesotaembedded-linuxsystem-switching
User reports an issue with A/B system switching in a Thor Yocto build. This indicates deployment/update reliability problems in an embedded Linux workflow where slot switching should be dependable.
⚡ Robust OTA and A/B update workflows are critical for scalable, safe field deployment of Physical AI systems on NVIDIA edge platforms.
- [BUG] MJCF `<adhesion>` actuator is silently dropped on importgithub:newton-physics/newton11/6/2026asset-pipelineusdmujoconewtonwarp
Newton's MJCF importer silently drops MuJoCo `<adhesion>` actuators during `add_mjcf`, leading to `nu` becoming 0 with no warning. This breaks model behavior and makes debugging difficult because the import succeeds but functionality is missing.
⚡ Faithful MJCF import is important for interoperability across simulators and toolchains feeding Physical AI training and evaluation.
- [BUG] Thin convex hull on a triangle mesh: downward normals at a seam between coplanar triangles, and no contacts past the volume centroid (#4766 follow-up)github:newton-physics/newton11/6/2026crashes-stabilitycrashusdrenderinghardwaredeploymentmujoco
A thin convex hull resting on a closed triangle mesh loses support on coplanar seams due to downward-facing contact normals and also loses all contacts after sinking past its volume centroid. The object can fall through the mesh in scenarios where analytic pla
⚡ Contact robustness on meshes is foundational for reliable manipulation/locomotion simulation and training stability across diverse assets.
- Gpu errornvidia-forum:robotics-edge-computing11/6/2026other
User reports a generic "Gpu error" with no additional details in the corpus. The problem indicates an unstable or failing GPU path that needs investigation and clearer error reporting.
⚡ GPU error visibility and actionable diagnostics are required to keep simulation, training, and edge inference workloads dependable.
- [BUG] Rigid contact writing loses precision far from the origingithub:newton-physics/newton11/6/2026crashes-stabilitycrashnewtonwarp
Rigid contact writing loses precision far from the origin: translating a rigid pair by 128 m changes body-local contact points by up to 9.76 µm even when relative pose is unchanged. The writer rounds world endpoints before transforming back, which can also mis
⚡ Large-world precision issues can undermine physically correct contacts for warehouse/factory-scale simulation and long-horizon autonomy evaluation.
- MathDx calls from a @wp.func fail to link in CUDA 13 debug buildsgithub:NVIDIA/warp11/5/2026renderingrenderinghardwarewarp
In Warp debug mode with CUDA Toolkit 13, calling a MathDx-backed tile builtin through a `@wp.func` in a backward-enabled kernel fails to link with an nvJitLink/NVVM compilation error. Calling the builtin directly from the kernel works, indicating a specific li
⚡ Toolchain compatibility (CUDA 13 + debug) is crucial for developer velocity when building differentiable simulation and learning kernels with Warp.
- CPU-block lanes use stale floating-point control state (FTZ, rounding)github:NVIDIA/warp11/5/2026otherwarp
When CPU cooperative blocks are enabled, Warp CPU fibers can run with stale floating-point control state (FTZ, rounding, exceptions) captured at fiber creation, not the launching thread’s current state. This can cause cooperative CPU-block results to differ fr
⚡ Deterministic numerics across execution paths are critical for trust in simulation, robotics optimization, and reproducible training results on NVIDIA’s Physical AI platform.
- [BUG] SolverMuJoCo freezes slow joint motion: sub-spacing position increments are rounded away in float32 joint_q while qvel persistsgithub:newton-physics/newton11/5/2026crashes-stabilitycrashmanipulationdxintegrationmujoconewton
SolverMuJoCo can freeze slow joint motion because joint positions are stored as float32 and sub-spacing increments get rounded away while velocity persists. This can cause position-driven joints to stop converging once motion slows below a threshold set by dt
⚡ Numerical precision pitfalls in solver state plumbing can silently invalidate slow-motion control behaviors that Physical AI training relies on.
- `array_scan` that works inside conditional bodies and foreign captures (caller-provided temp storage, or a capture query)github:NVIDIA/warp11/5/2026renderingrenderinghardwarewarp
Warp’s array_scan allocates and frees CUB temp storage inside the call, which breaks CUDA graph capture in conditional capture bodies and in captures started by other frameworks. The request is for caller-provided temp storage or a capture-query mechanism to m
⚡ Graph-capture-safe primitives are key to high-performance training and pipeline composability across CUDA frameworks in NVIDIA’s Physical AI platform.
- ParticleField (3DGS) renders blurred when added to an already open stagegithub:isaac-sim/IsaacSim11/5/2026asset-pipelinev6.2.0usdrenderinghardwaresensorsintegrationisaac-sim
ParticleField3DGaussianSplat renders significantly blurrier when referenced into an already-open stage than when present at stage open. Rendering a few frames immediately after adding the reference is reported as a workaround, implying initialization/order-dep
⚡ Rendering consistency when building stages programmatically is important for scalable synthetic data and perception validation in Physical AI workflows.
- [BUG] Kamino contact examples drop a substep per frame under graph capturegithub:newton-physics/newton11/5/2026renderingrenderinghardwarenewton
When running certain contact examples with the Kamino solver under CUDA graph capture, a Python-level state buffer swap is captured incorrectly, causing one substep to be effectively skipped each frame. With an odd substep count, each replay starts from the wr
⚡ CUDA Graph capture correctness issues in physics stepping can silently corrupt sim time and undermine deterministic, high-performance simulation pipelines critical for Physical AI training and testing
- Orin Nano: ~20 us memory stall on 1 s BPMP EMC tick. Can it be disabled?nvidia-forum:robotics-edge-computing11/5/2026deploymentorin-nanolatencymemory-stallbpmpemcreal-time
User reports ~20 µs memory stalls on a 1-second BPMP EMC tick on Orin Nano and asks whether it can be disabled. This periodic stall may violate low-latency requirements and needs a mitigation or configurability option.
⚡ Deterministic latency on Jetson platforms is crucial for real-time Physical AI control and perception workloads at the edge.
- ArticulationView.get_generalized_gravity_forces() overwrites link accelerations and velocities, and returns wrong torques for floating-base articulationsgithub:isaac-sim/IsaacSim11/5/2026crashes-stabilitycrashrlusdrenderinghardwarelocomotion
In the PhysX tensor API on GPU, ArticulationView.get_generalized_gravity_forces() has side effects that overwrite later reads of link accelerations and velocities until the next simulate(). It also returns wrong torques for floating-base articulations, comprom
⚡ Correct and side-effect-free physics tensor APIs are necessary for trustworthy learning, control, and analytics on top of Isaac Sim/PhysX.
- CPU &GPU Frequency unnormal under super modenvidia-forum:robotics-edge-computing11/4/2026hardware-integrationrobotics-edge-computingcpu-frequencygpu-frequencypower-modesperformance-stability
CPU and GPU frequency behavior is reported as abnormal when running in "super mode". This can impact performance consistency and thermal/power management during robotics workloads.
⚡ Power/performance mode instability on edge platforms threatens predictable real-time behavior for deployed Physical AI systems.
- [BUG] warp.optim.linear solvers are slow on CPU and ignore use_cuda_graph=Truegithub:NVIDIA/warp11/4/2026hardware-integrationhardwarewarp
warp.optim.linear solvers are slow on CPU due to a dot-product tiling strategy that redundantly computes sums. The use_cuda_graph=True option has no effect on CPU, forcing repeated Python overhead each iteration.
⚡ Inefficient CPU solver paths and missing graph replay reduce throughput and raise iteration cost for optimization-heavy Physical AI workloads.
- [BUG] SolverVBD: particle-body attachments are ignored in deterministic mode when contacts are passed to step()github:newton-physics/newton11/4/2026crashes-stabilitynewton-physicsvbddeterministic-modecontactsforce-accumulationwarp-kernels
In deterministic mode, SolverVBD ignores particle-body attachments when contacts are passed to step(), due to a kernel doing a plain write that overwrites accumulated forces/hessians. This breaks expected force accumulation behavior under specific stepping con
⚡ Deterministic simulation correctness is essential for reliable regression testing and reproducible training in Physical AI workflows.
- [BUG] SolverVBD: with particle self-contact on, a cloth pile sinks through a static ground planegithub:newton-physics/newton11/4/2026crashes-stabilitynewtonsolver-vbdclothself-contactground-collisiondeformables
With particle self-contact enabled, SolverVBD truncates particle motion against self-contact planes but does not robustly truncate motion into static collision shapes by default. This makes cloth piles sink through a static ground unless additional collision p
⚡ Accurate, stable cloth–rigid contact handling is a prerequisite for trustworthy deformable simulation in Physical AI training and testing loops.
- [BUG] SolverSemiImplicit and SolverXPBD bending: the dihedral-angle gradient is scaled by twice the triangle area, so bending depends on the mesh size (and is skipped below ~1 mm)github:newton-physics/newton11/4/2026crashes-stabilitynewtonsolver-xpbdsolver-semi-implicitcloth-bendingmesh-scalingdihedral-angle
Bending in SolverSemiImplicit and SolverXPBD is incorrectly scaled (by twice triangle area), making bending behavior depend on mesh size. For very small triangles (~1 mm and below), bending can effectively be skipped.
⚡ Mesh-resolution-invariant deformable behavior is critical for consistent simulation fidelity across assets and scales.
- GR00T 2hn:gr00t11/9/2026▲ 3othergroot
A Hacker News post titled "GR00T 2" indicates community discussion/interest in the GR00T 2 topic. No additional technical details are provided in the corpus entry.
⚡ External community attention to robot foundation models is a leading indicator for ecosystem momentum around NVIDIA’s Physical AI strategy.
- Nvidia GR00T 2: Scaling Robot Foundation Models [pdf]hn:gr00t11/9/2026▲ 1othergroot
A Hacker News post references "Nvidia GR00T 2: Scaling Robot Foundation Models [pdf]". No additional content is included in the corpus entry.
⚡ Public discussion of GR00T 2 research outputs signals demand for scalable robot foundation modeling tied to NVIDIA’s platform narrative.
- Orin Nano Super Dev Kit dead: 19.35 V at the jack, 0.000 A draw, nothing lights up (in warranty)nvidia-forum:robotics-edge-computing11/9/2026▲ 1hardware-integrationhardware
User reports an Orin Nano Super Dev Kit that appears dead while under warranty: 19.35V at the power jack, 0.000A draw, and no lights. The corpus includes no troubleshooting steps or outcome.
⚡ Hardware RMA/diagnostic clarity for Jetson dev kits is foundational for developer success and downstream Physical AI deployment confidence.
- Xavier NX software no longer supportednvidia-forum:robotics-edge-computing11/9/2026▲ 1other
User reports that Xavier NX software is no longer supported, implying they are blocked or forced into upgrades/migrations to keep systems maintained.
⚡ Support lifecycle clarity and migration paths are critical to keep deployed robotics fleets on NVIDIA platforms without churn.
- Deprecate use_cuda_graph in the warp.optim.linear solvers in favor of use_graphgithub:NVIDIA/warp11/8/2026▲ 1hardware-integrationhardwarenewtonwarp
Warp is proposing to add a `use_graph` argument to linear solvers and deprecate `use_cuda_graph`. The goal is to reflect behavior changes where capture can apply beyond CUDA and to provide a clearer, future-proof API.
⚡ API consistency around CUDA graph capture controls matters for keeping Warp-based simulation/optimization code maintainable as graph usage expands.
- Subtree mirroring with negative scale on framesgithub:google-deepmind/mujoco11/8/2026▲ 1other
Feature request for MuJoCo to mirror subtree elements when a frame has negative scale, so symmetric subtrees can be authored once. The proposal includes mirroring geometry and inertia and handling joint axis/velocity conventions under mirroring.
⚡ Better mirroring primitives in physics model authoring reduces asset effort and improves scalable environment/robot creation for simulation-based Physical AI.
- Implausible authored physics:mass on three nvidia/simready-assets packages (air hose reel, utility knife, storage shelf)nvidia-forum:simulation11/7/2026▲ 1other
Three SimReady asset packages are reported to have implausible authored physics:mass values (air hose reel, utility knife, storage shelf). This can break realism and downstream training/calibration that depends on correct masses.
⚡ Trustworthy, physically plausible asset metadata is necessary to scale synthetic environments and avoid hidden sim fidelity failures in Physical AI training.
- Can I use an event camera with GStreamer?nvidia-forum:robotics-edge-computing11/6/2026▲ 1sensors-perceptionsensors
User is asking whether an event camera can be used with GStreamer. This indicates uncertainty or lack of clear guidance/tooling for integrating event-based vision streams into the standard NVIDIA multimedia pipeline.
⚡ Event-camera ingestion and middleware compatibility are key to broadening sensor support and accelerating edge robotics adoption on NVIDIA platforms.
- Agility's Digit 5 humanoid robot will run on Nvidia as it heads for Nasdaqhn:humanoid robot nvidia11/5/2026▲ 1otherhumanoidagility
Link-style discussion item about Agility Robotics’ Digit 5 humanoid robot running on NVIDIA as it heads toward Nasdaq. No specific technical issue is described, but it is a market signal relevant to platform adoption.
⚡ Humanoid OEM alignment with NVIDIA compute strengthens the platform’s position as the default stack for real-world humanoid deployment and scaling.
- Where should image processing happen in a low-latency NVIDIA Holoscan camera pipeline?nvidia-forum:robotics-edge-computing11/5/2026▲ 1deploymentholoscancamera-pipelinelow-latencyimage-processingedgepipeline-architecture
User asks where image processing should occur in a low-latency NVIDIA Holoscan camera pipeline. This indicates uncertainty about best practices for placing processing stages to meet latency targets.
⚡ Reference guidance for low-latency pipeline partitioning is essential to make Holoscan a predictable building block for Physical AI perception systems.
- Unaddressed issue in the Isaac Ros reponvidia-forum:isaac11/4/2026▲ 1other
Forum post reports an "Unaddressed issue in the Isaac Ros repo" but provides no details in the body. The lack of context blocks triage and suggests a support/visibility gap.
⚡ Perceived unresponsiveness on Isaac ROS issues harms developer trust and slows ecosystem adoption.
- Unaddressed issue in the Isaac Ros reponvidia-forum:isaac-ros11/4/2026▲ 1other
Forum post reports an "Unaddressed issue in the Isaac Ros repo" but provides no details in the body. The lack of context blocks triage and suggests a support/visibility gap.
⚡ Perceived unresponsiveness on Isaac ROS issues harms developer trust and slows ecosystem adoption.
- [Python] rollout rejects a tuple of MjData although its signature accepts a Sequencegithub:google-deepmind/mujoco11/10/2026tooling-dxmujocopython-apirollouttypingsequencewindows
MuJoCo's Python rollout API claims to accept a Sequence[MjData] but only recognizes list, so tuples get wrapped incorrectly and then fail in compiled code. The resulting error message is unhelpful and doesn't explain the real cause.
⚡ API contract mismatches in core rollout tooling increase user confusion and slow iteration in simulation-driven learning loops.
- 30/5000 Hello, NVIDIA team. How does Yocto support the overlays function?nvidia-forum:robotics-edge-computing11/10/2026docs-onboardingnvidia-forumsyoctodevice-tree-overlaysrobotics-edgeembedded-linuxos-integration
A forum user asks how Yocto supports the overlays function, with no additional context provided. The lack of details suggests a documentation or onboarding gap around overlays in NVIDIA's Yocto-related support.
⚡ Clear guidance for embedded OS customization (like Yocto overlays) reduces integration risk for edge robotics deployments.
- Official pkg-config recipes on Linux (Ubuntu, Fedora, etc.) and macOSgithub:google-deepmind/mujoco11/10/2026tooling-dxmujocopkg-configlinuxmacospackagingbuild-linking
Users want official pkg-config files and standard Linux/macOS packaging to avoid manual environment setup when installing or upgrading MuJoCo. They also request hosted prebuilt packages via common package managers (APT/DNF).
⚡ Packaging and discovery gaps in core sim dependencies increase setup friction and slow adoption of robotics stacks that need reproducible builds.
- SMCU reset AND gate (U184): needed without the debug MCU?nvidia-forum:robotics-edge-computing11/10/2026hardware-integrationnvidia-forumrobotics-edgesmcuhardware-designdebug-mcu
A user asks whether an SMCU reset AND gate (U184) is needed when the debug MCU is not present. The post indicates uncertainty about required hardware design components for a given system configuration.
⚡ Unclear hardware requirements and reference design assumptions create integration risk for edge/robotics products built on NVIDIA platforms.
- Programming the sMCU and FSI on IGX Thor systemsnvidia-forum:robotics-edge-computing11/10/2026hardware-integrationnvidia-forumrobotics-edgeigx-thorsmcufsibringup
A user asks about programming the sMCU and FSI on IGX Thor systems. The post indicates uncertainty or missing documentation/tools for programming these components.
⚡ Clear programming and provisioning workflows for embedded controllers are necessary to shorten bring-up cycles for IGX-based Physical AI systems.
- Inquiry about unexpected GPIO behavior on Jetsonnvidia-forum:robotics-edge-computing11/10/2026hardware-integrationjetsongpiorobotics-edgehardware-ioembedded
User reports unexpected GPIO behavior on Jetson but provides no details in the body. This suggests a potential hardware/driver/configuration mismatch impacting basic I/O.
⚡ Hardware I/O reliability on Jetson-class devices directly affects deployability of Physical AI systems at the edge.
- Inquiry Regarding MAC Address Assignment on Thor T5000nvidia-forum:robotics-edge-computing11/10/2026hardware-integrationthor-t5000mac-addressnetworkingrobotics-edgedevice-provisioning
User asks how MAC address assignment works on Thor T5000, but no further context is provided. This points to uncertainty around networking configuration and device identity management.
⚡ Clear network identity provisioning on NVIDIA edge/robotics platforms reduces integration friction for deployed Physical AI systems.
- Question about replacing FSI GPIO/SGMII pins for UFS interface on IGX T5000nvidia-forum:robotics-edge-computing11/10/2026hardware-integrationigx-t5000ufssgmiigpiopinmuxhardware-bringup
User asks about replacing FSI GPIO/SGMII pins for a UFS interface on IGX T5000, with no additional details. This suggests potential constraints or confusion in available pin/interface multiplexing options.
⚡ Pin-mux and interface flexibility questions indicate integration complexity that can slow Physical AI hardware bring-up.
- [Blog] Turn Human Demonstrations Into Robot Training Data With RLWRLD and NVIDIA Cosmosnvidia-forum:robotics-edge-computing11/9/2026synthetic-datasynthetic-datacosmos
A forum post links a blog about turning human demonstrations into robot training data using RLWRLD and NVIDIA Cosmos. It indicates interest in a workflow that converts demonstrations into usable training datasets.
⚡ Converting human demonstrations into scalable training data is a key lever for accelerating robot policy development on NVIDIA’s Physical AI stack.
- Avoid intermediate overflow in tile_ldlt solve and adjointgithub:NVIDIA/warp11/9/2026hardware-integrationhardwarewarp
The tile_ldlt solve and its adjoint can overflow intermediates, producing inf even when the true result is representable. The issue requests numerical hardening in specific code paths to avoid overflow.
⚡ Numerical robustness in differentiable solvers is critical for reliable optimization and learning pipelines built on Warp.
- Extend shared-diagonal tile_cholesky path to FP32github:NVIDIA/warp11/9/2026otherwarp
A faster shared-diagonal tile_cholesky path is currently only selected for FP64 and the request is to extend it to FP32. Preliminary measurements suggest meaningful speedups across multiple configurations, but selection rules and tests need updates.
⚡ Improving FP32 performance paths in core math kernels helps scale real-time simulation and learning workloads efficiently.
- [NEWS] NVIDIA at CoRL 2026: Advancing Robot Learningnvidia-forum:robotics-edge-computing11/9/2026other
A forum news post announces NVIDIA activity at CoRL 2026 focused on advancing robot learning. The entry is informational and does not describe a problem.
⚡ Conference visibility and research positioning influence developer adoption and partnership opportunities for the Physical AI platform.
- [REQ] Speed up building NanoVDB volumes on the CPUgithub:NVIDIA/warp11/9/2026otherwarp
Building NanoVDB volumes on the CPU is slow for allocate/rebuild paths, affecting Warp Volume APIs and warp.fem Nanogrid geometries. The issue requests performance improvements, with a timing example showing ~137 ms for activating 2^20 random points.
⚡ Faster CPU-side volume construction broadens where Warp can be used (including CPU-only and mixed pipelines) and reduces iteration time for simulation and geometry workflows.
- [REQ] Export CPU graph captures to DOT with capture_debug_dot_print()github:NVIDIA/warp11/9/2026hardware-integrationhardwarewarp
capture_debug_dot_print() only exports DOT for CUDA graphs because it relies on cudaGraphDebugDotPrint(). The request is to export CPU-captured graphs to DOT using the already-recorded CPU operation list so the API behaves consistently across devices.
⚡ Parity in graph introspection across CPU and CUDA improves debugging workflows and lowers friction adopting Warp capture for diverse deployment targets.
- Olli – Autonomous Drink-Serving Robot with Local Voice AI (Hackathon Project, Germany)nvidia-forum:isaac11/9/2026other
Isaac forum post titled an autonomous drink-serving robot with local voice AI from a hackathon project. The body is empty in the corpus, so no concrete issue or request can be derived.
⚡ General Isaac community projects can guide prioritization toward real deployments, but actionable signals require technical specifics.
- Olli – Autonomous Drink-Serving Robot with Local Voice AI (Hackathon Project, Germany)nvidia-forum:isaac-ros11/9/2026other
Isaac ROS forum post titled an autonomous drink-serving robot with local voice AI from a hackathon project. The corpus contains no body content to extract concrete problems, integrations, or feature gaps.
⚡ Surface-level community demos in the Isaac ROS area can indicate demand for packaged perception-to-action stacks, but need details to drive product work.
- Olli – Autonomous Drink-Serving Robot with Local Voice AI (Hackathon Project, Germany)nvidia-forum:robotics-edge-computing11/9/2026other
Forum post titled an autonomous drink-serving robot with local voice AI from a hackathon project. No technical details are provided in the corpus body to classify specific issues or requests.
⚡ Community project showcases highlight emerging integration patterns (robot + local voice AI) that can inform end-to-end Physical AI reference solutions.
- Need help to reconcile measured Orin NX junction-to-case temperature differences with the preliminary values in the TDGnvidia-forum:robotics-edge-computing11/9/2026hardware-integrationhardware
User needs help reconciling measured Orin NX junction-to-case temperature differences with preliminary values in the TDG. The corpus provides no additional measurement details or resolution.
⚡ Clear thermal characterization guidance for Jetson modules reduces deployment risk and support load for edge Physical AI systems.
- Camera benchmark regressions linked to Warp upgradegithub:newton-physics/newton11/9/2026sensors-perceptionsensorswarp
Newton reports camera benchmark regressions that are linked to a Warp upgrade. The corpus provides no additional details on the magnitude, versions, or repro steps.
⚡ Performance regressions tied to dependency upgrades can erode confidence in the stack and require tighter benchmarking/compatibility gating across components like Warp.
- CPU shared tile loads inhibit vectorization and slow tile matmulgithub:NVIDIA/warp11/9/2026training-infrarlrenderinghardwarenewtonwarp
On CPU, shared-tile loading in Warp tile_matmul() inhibits vectorization and becomes a bottleneck in the default execution mode, making Newton/Kamino CPU solves ~9–20% slower versus a custom path. The issue is attributed to scalar shared-load address iteration
⚡ CPU performance bottlenecks in core math primitives reduce the cost-effectiveness of simulation and optimization workloads that must run off-GPU or at scale.
- Raspberry Pi or Jetson for an event-camera prototype?nvidia-forum:robotics-edge-computing11/9/2026hardware-integrationhardwaresensors
User is deciding between Raspberry Pi and Jetson for an event-camera prototype and likely needs clarity on performance, software support, and sensor pipeline compatibility.
⚡ Clear guidance on edge hardware choices for novel sensors reduces time-to-prototype and drives adoption of NVIDIA edge stacks in Physical AI workflows.
- BAM friction models for actuatorsgithub:newton-physics/newton11/9/2026integrationintegrationmujoco
Request/proposal to compute and apply BAM friction using MuJoCo solver observables (e.g., friction loss/constraint/bias forces) to avoid chatter and stiffness issues. The approach is explicitly limited to MuJoCo and depends on notifying model changes for joint
⚡ High-fidelity actuator/friction modeling across simulators improves sim-to-real transfer for Physical AI control policies.
- Clarification on Flashing Jetson Orin Nano Production Module to External NVMe SSD Using a Custom Carrier Boardnvidia-forum:robotics-edge-computing11/9/2026hardware-integrationjetsonorin-nanocustom-carrier-boardnvmeflashingoem-integration
User seeks clarification on flashing a Jetson Orin Nano production module to an external NVMe SSD when using a custom carrier board. This suggests documentation gaps or edge cases in standard flashing procedures.
⚡ Reducing ambiguity in flashing workflows for custom carriers accelerates OEM adoption of NVIDIA modules in Physical AI products.
- Unable to handle kernel NULL pointer dereference at virtual address 0000000000000004nvidia-forum:robotics-edge-computing11/9/2026other
User reports a kernel NULL pointer dereference at virtual address 0x0000000000000004. This indicates a kernel crash that can halt the device and prevent progress until debugged.
⚡ Kernel stability on Jetson platforms is a prerequisite for dependable Physical AI edge deployments.
- R3840 开机在uefi阶段卡住nvidia-forum:robotics-edge-computing11/9/2026other
User reports the device gets stuck during the UEFI stage on boot. This is a boot-blocking failure preventing the system from starting.
⚡ Boot-stage robustness and recovery paths are critical to operating Physical AI systems in the field without manual intervention.
- Jetson thor冷启动低概率失败问题nvidia-forum:robotics-edge-computing11/9/2026hardware-integrationhardware
User reports a low-probability cold boot failure issue on Jetson Thor. The intermittent nature suggests a hard-to-reproduce stability or hardware/firmware initialization problem.
⚡ Reducing intermittent cold-boot failures is essential for reliable always-on Physical AI robots and appliances.
- OAK-D CAMERAnvidia-forum:robotics-edge-computing11/9/2026sensors-perceptionsensors
User topic indicates questions or issues related to using an OAK-D camera. The lack of details suggests a request for compatibility, setup, or troubleshooting guidance.
⚡ Broader camera ecosystem compatibility on Jetson accelerates real-world Physical AI prototyping and deployment.
- What are the considerations when using the T4000 instead of the T5000 in the Thor development kit?nvidia-forum:robotics-edge-computing11/9/2026hardware-integrationjetson-thorthor-devkitgpu-selectiont4000t5000
A user asks for considerations when using a T4000 instead of a T5000 in the Thor development kit. The post body is empty, so constraints like performance, power, thermals, and compatibility are unspecified.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Expose optional native MPM sleeping and bounded GPU evaluationgithub:isaac-sim/IsaacLab11/9/2026otherisaac-labnewtonwarp
Request to expose Newton's proposed GPU particle sleeping and bounded evaluation pools through a thin Isaac Lab interface. The goal is to reduce per-step work/buffers for MPM granular scenes while preserving defaults and capture/replay correctness.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Add optional GPU particle sleeping and bounded working sets for implicit MPMgithub:newton-physics/newton11/9/2026integrationintegrationisaac-labwarp
Proposal to add optional GPU particle sleeping and bounded working sets for implicit MPM to avoid evaluating large numbers of quiet particles in granular beds. Notes that sleeping is an approximation and must preserve full-MPM behavior by default and guard cap
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Avoid padded sorting for unique-incidence FEM restrictionsgithub:NVIDIA/warp11/9/2026hardware-integrationhardwarenewton
Warp captured PIC/MPM path is inefficient because evaluation-point restrictions build/sort padded incidence data even though each point has exactly one incident cell. Capture bodies also allocate scratch arrays, requiring persistent backing to survive conditio
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Unable to Reproduce DROID-W Benchmark Performance from NVIDIA Workshopnvidia-forum:isaac11/9/2026other
User reports they cannot reproduce DROID-W benchmark performance from an NVIDIA workshop (Isaac forum). The post body is empty, so the environment and performance metrics cannot be assessed.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Unable to Reproduce DROID-W Benchmark Performance from NVIDIA Workshopnvidia-forum:isaac-ros11/9/2026other
User reports they cannot reproduce DROID-W benchmark performance from an NVIDIA workshop (Isaac ROS forum). The post body is empty, so root-cause signals like model version, ROS graph, and accelerator utilization are missing.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- Regarding NVIDIA's continued YOCTO supportnvidia-forum:robotics-edge-computing11/8/2026deploymentyoctojetsonlinuxsupport-policyembedded
User asks about NVIDIA's continued YOCTO support. The post body is empty, so timelines, boards, and support scope are not specified.
⚡ What this means strategically for NVIDIA's Physical AI platform.
- [REQ] Support 64-bit integer indices for BVH and Meshgithub:NVIDIA/warp11/8/2026otherwarp
Request to support 64-bit integer indices for `wp.Mesh` and BVH traversal to avoid int32 overflow with extremely large meshes or merged scenes. The ask includes BVH build and query functions handling int64/uint64 indices correctly.
⚡ Supporting very large scene geometry enables higher-fidelity synthetic environments and larger-scale simulation workloads on NVIDIA’s Physical AI stack.
- Expose the libmathdx version through the public APIgithub:NVIDIA/warp11/8/2026otherwarp
Request to expose the bundled libmathdx version via `wp.get_libmathdx_version()` returning a `(major, minor, patch)` tuple or `None`. This supports compatibility checks and code-path selection for applications using MathDx-backed tile operations.
⚡ Exposing dependency versions in public APIs reduces integration risk and improves debuggability for downstream Physical AI applications built on Warp.
- تطبيق تنزيل فيديوgithub:google-deepmind/mujoco11/8/2026other
Issue title and body appear unrelated to MuJoCo and provide no actionable feature/bug details. It looks like noise/spam rather than a valid request.
⚡ Maintaining high-signal issue trackers is necessary to keep developer feedback loops effective for Physical AI tooling ecosystems.
- Float texture sampling uses more registers on sm_120 since 1.19.0github:NVIDIA/warp11/8/2026renderingrenderinghardwaresensorsnewtonwarp
Warp texture sampling since 1.19.0 adds extra integer conversion paths even for float textures, increasing register pressure on sm_120. This slows texture-heavy kernels and shows up in register counts for compiled cubins.
⚡ Performance regressions on new GPU architectures can erode the advantage of NVIDIA’s compute stack for Physical AI workloads that rely on heavy rendering/sampling kernels.
- TM1637 7-Segment 4-Digit Displaynvidia-forum:robotics-edge-computing11/8/2026hardware-integrationtm16377-segment-displaygpioperipheralsrobotics-edgeintegration
User asks about using a TM1637 7-segment 4-digit display. This indicates a need for guidance or support for integrating this common peripheral with the edge robotics platform.
⚡ Clear pathways for interfacing common peripherals improves the completeness of edge robotics reference designs tied to NVIDIA platforms.
- Isaac Sim 6.2 container headless on GCP: renderer never advances (no frames) on G4 and T4 vWSnvidia-forum:simulation11/7/2026otherv6.2isaac-sim
Isaac Sim 6.2 headless in a container on GCP does not advance the renderer and produces no frames on G4 and T4 vWS. This blocks headless usage for those cloud GPU setups.
⚡ Cloud headless rendering reliability is a prerequisite for scalable synthetic data generation and remote training for Physical AI.
- Why the viewer `Inertia` shows `Convex Hull`, but the `Convex Hull` not working?github:google-deepmind/mujoco11/7/2026asset-pipelineusdrenderingdocsmujocowarp
In MuJoCo viewer, toggling 'Inertia' appears to toggle 'Convex Hull', while the 'Convex Hull' toggle appears to do nothing. The UI behavior is confusing and may indicate a viewer toggle bug or mislabeled control.
⚡ Viewer/debug tooling clarity matters because it shapes how developers diagnose dynamics and geometry issues across Physical AI simulators.
- [BUG] SolverImplicitMPM with separate_worlds=True: one world's result depends on the other worlds (shared stopping test, dense-grid bounds, construction-time material switches)github:newton-physics/newton11/7/2026training-infrarlrenderinghardwarenewtonwarp
With separate_worlds=True in SolverImplicitMPM, worlds remain coupled via a shared stopping test and other shared behaviors, so one world's settings can change another world's results. This breaks isolation guarantees needed for batch RL/evaluation since traje
⚡ True independence in batched/multi-world simulation is critical for scalable RL and evaluation pipelines central to Physical AI development.
- Issue of Isaac Sim on RTX Sparknvidia-forum:simulation11/6/2026renderingrenderingisaac-sim
A forum report indicates an Isaac Sim issue specifically on RTX Spark, but provides no details in the corpus. The lack of body content makes the failure mode and impact unclear.
⚡ Unclear compatibility or setup issues on emerging RTX hardware can slow rollout of Isaac Sim across new developer workstations and edge/dev kits.
- Sharing a result in case it helps others on R36.5.2, and asking which fix NVIDIA recommendsnvidia-forum:robotics-edge-computing11/6/2026other
A forum post references sharing results on R36.5.2 and asks which fix NVIDIA recommends, but the corpus contains no body details. The exact issue and remediation options are not described here.
⚡ Edge-platform reliability depends on clear, recommended fixes for specific Jetson/L4T releases so robotics deployments don’t fragment into ad-hoc patches.
- [Isaac Sim 6.2] Moving RTX LiDAR with many emitter states: FSD re-syncs the whole prim every frame (~1.5–1.9 ms per LiDAR per frame)github:isaac-sim/IsaacSim11/6/2026renderingv6.2renderinghardwaresensorsisaac-sim
In Isaac Sim 6.2, moving an RTX LiDAR prim with many emitter states triggers Fabric Scene Delegate to re-sync the whole prim every frame. The added main-thread cost (~1.5–1.9 ms per LiDAR per frame) scales with emitter state count and disappears when the LiDAR
⚡ Sensor simulation performance regressions on moving platforms threaten real-time robotics simulation and scalable synthetic data generation pipelines.
- [NEWS] SimReady Assets Reach 1 Million Downloads on Hugging Facenvidia-forum:robotics-edge-computing11/6/2026other
A forum news post reports SimReady Assets reaching 1 million downloads on Hugging Face, with no additional technical details in the corpus. It signals growing usage/interest in the asset library.
⚡ Asset distribution momentum is a strategic lever for accelerating ecosystem adoption, but needs continued pipeline and integration investment to translate downloads into usage.
- Sparse mju_cholFactorNumeric returns a wrong factor when it merges rows with different descendants (Newton solver, since 4.0.0)github:google-deepmind/mujoco11/6/2026training-infrarlmanipulationmujoconewton
MuJoCo's sparse `mju_cholFactorNumeric` can return an incorrect factor when merging rows with different descendants, causing `L'L != H`. This breaks the Newton solver step for models with sparse Jacobians (since 4.0.0), leading to wrong dynamics/optimization b
⚡ Core numerical linear algebra correctness is critical for stable, high-DOF robot simulation and optimization loops used in Physical AI research and productization.
- Skip the negative-index wrap for unsigned array subscriptsgithub:NVIDIA/warp11/6/2026othernewtonwarp
Request to skip Python-style negative-index wrapping when array subscripts are unsigned types (uint32/uint64) so indices stay unsigned through offset arithmetic. Current implementation applies an `if (i < 0) i += shape[d];` check on every dimension for every a
⚡ Low-level kernel codegen and indexing semantics impact Warp performance ceilings, which cascades into faster simulation and training workloads in Physical AI stacks.
- Jetson TX2: Recommended procedure for Wi‑Fi RF test mode (continuous TX) on the on-board BCM4354 Wi‑Fi (L4T R32.3.1)nvidia-forum:robotics-edge-computing11/6/2026hardware-integrationhardware
A forum thread asks for the recommended procedure to put Jetson TX2 on-board Wi‑Fi (BCM4354) into RF continuous transmit test mode on L4T R32.3.1. The corpus contains no further details beyond the title.
⚡ Clear, authoritative hardware test procedures reduce time-to-validation for edge robotics platforms used alongside Physical AI development.
- Jetson Orin NX 16GB no UART output, not appearing as USB in recovery mode after flashnvidia-forum:robotics-edge-computing11/6/2026hardware-integrationhardware
A Jetson Orin NX 16GB reportedly has no UART output and does not appear as USB in recovery mode after flashing. Without recovery visibility, the device may be effectively unusable until resolved.
⚡ Flash/recovery failures that brick access paths can halt edge-device bring-up, delaying robotics deployment and data collection critical to Physical AI.
- Orin Nano (L4T 36.5) production fuse set without KDK0: odmfuse.sh vs FSKP, fTPM impact, and post-burn verificationnvidia-forum:robotics-edge-computing11/6/2026hardware-integrationhardwaredeployment
A forum post concerns setting production fuses on Orin Nano (L4T 36.5) without KDK0 and compares odmfuse.sh vs FSKP, including questions about fTPM impact and verification after burning. The corpus provides only the title, so specific errors or guidance gaps a
⚡ Production security fuse flows must be unambiguous and verifiable to enable safe at-scale deployment of edge robotics platforms supporting Physical AI workloads.
- EoL for AGX Xavier productnvidia-forum:robotics-edge-computing11/6/2026other
User is asking about end-of-life status for the Jetson AGX Xavier product. This suggests concern about long-term availability, support horizon, and planning for upgrades or replacements.
⚡ Clear hardware lifecycle and migration paths protect developer trust and reduce platform churn for Physical AI deployments at the edge.
- UPHY1 used as C4nvidia-forum:robotics-edge-computing11/6/2026other
Question about configuring or using UPHY1 as C4. Indicates a hardware configuration or lane assignment uncertainty during platform bring-up.
⚡ Board-level lane mapping and PHY configuration questions reflect the need for clearer hardware integration guidance for edge platforms.
- [Jetson AGX Thor][JP7.1/R38.4.0] RT5640 AUD_MCLK shutdown glitch at end of playback — request for official BPMP/kernel patchnvidia-forum:robotics-edge-computing11/6/2026hardware-integrationhardwarefeature-request
Audio master clock (AUD_MCLK) appears to glitch/shut down at end of playback on Jetson AGX Thor JP7.1/R38.4.0. User is requesting an official BPMP/kernel patch to resolve it.
⚡ Physical AI edge deployments need stable, vendor-supported BSP fixes for peripheral timing/glitch issues to reduce integration risk on Jetson/Thor.
- Read `mjc:inertia` from MjcPhysics USD meshesgithub:newton-physics/newton11/6/2026asset-pipelineusdmujoconewton
Newton’s `parse_usd` does not read MuJoCo `mjc:inertia` from `MjcPhysics` USD meshes, even though MJCF `<mesh inertia>` is honored elsewhere. The same model can import with different mesh masses depending on whether it comes from MJCF or MjcPhysics USD.
⚡ Consistent cross-format physics import (MJCF vs USD schemas) is critical to trustworthy sim behavior and reproducible training results.
- Need to know at what speed camera i2c frequency driver on Nvidia Drive Thor SDK7.2.5 is runningnvidia-forum:robotics-edge-computing11/6/2026hardware-integrationhardwaresensors
User needs to determine the operating I2C frequency used by the camera driver on NVIDIA Drive Thor SDK 7.2.5. The issue is about identifying the runtime bus speed.
⚡ Camera bring-up on DRIVE/Thor depends on transparent low-level bus timing details to debug sensor reliability and performance.
- JP7.2 RT kernel: BPMP clock request never completes after wk2xxx sets 4800, SPI runtime PM and thermal block for ~600snvidia-forum:robotics-edge-computing11/6/2026other
On JP7.2 RT kernel, a BPMP clock request reportedly never completes after wk2xxx sets 4800, and SPI runtime PM plus thermal causes a long block (~600s). This can stall the system for minutes.
⚡ Real-time robotics stacks on Jetson require reliable clock/PM interactions; long stalls can invalidate hard real-time assumptions in deployed systems.
- Jetson orin nano 96724- 9295D 单路单独取图sensor出图，两路sensor同时起流，起流失败nvidia-forum:robotics-edge-computing11/6/2026hardware-integrationhardware
On Jetson Orin Nano, a sensor can stream when started alone, but when two sensors are started simultaneously the stream start fails. This blocks dual-camera configurations.
⚡ Multi-camera streaming reliability on Jetson is essential for perception stacks and must scale from single-sensor bring-up to multi-sensor operation.
- How can I set the DP output timing during Linux kernel boot on the Thor platform?nvidia-forum:robotics-edge-computing11/6/2026hardware-integrationhardware
User asks how to set DisplayPort output timing during Linux kernel boot on the Thor platform. They need control over DP timing early in boot.
⚡ Deterministic display init on embedded platforms reduces integration time for HMI/bring-up flows in robotics and automotive edge systems.
- [Bug Report] `collision_props` on `MjcfFileCfg` / `UrdfFileCfg` skip the mesh colliders the importers author inside USD instancesgithub:isaac-sim/IsaacLab11/6/2026crashes-stabilityv6.2crashusdrenderinghardwareisaac-simisaac-lab
In Isaac Sim 6.2, MJCF/URDF importers author mesh colliders inside USD instances, but `spawn_from_usd_file` skips instance proxies when applying schema properties like `collision_props`. As a result, collider properties are not applied to mesh colliders author
⚡ USD instancing-safe configuration application is key for scalable robot import workflows and consistent physics settings in Isaac Lab environments.
- Decision models (d1, JEV, Perplexity Flash, d3-mini, Kev) measured on Jetson – answer accuracy, latency, memory per requestnvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
Post references measuring multiple decision models on Jetson, focusing on answer accuracy, latency, and memory per request. It indicates a need for comparative deployment metrics on Jetson.
⚡ On-device model selection for Jetson benefits from transparent benchmarking to guide practical Physical AI deployment tradeoffs.
- Could you please list the compatible Wi-Fi models with Jetson Orin Nano/Jetpack 7.2 /39.2 24.04?nvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
User is asking for a list of compatible Wi‑Fi models for Jetson Orin Nano on JetPack 7.2 / 39.2 (Ubuntu 24.04). Lack of an explicit compatibility list forces trial-and-error and slows deployments.
⚡ Clear compatibility guidance reduces time-to-deploy for Jetson-based robotics stacks that feed into NVIDIA’s end-to-end Physical AI workflow.
- JETSON Thor PCIE Testnvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
Post requests information related to PCIe testing on Jetson Thor. Missing details in the report indicate a need for clearer test procedures and expected behavior around PCIe bring-up/validation.
⚡ Robust PCIe validation and test guidance is a prerequisite for reliable sensor/compute expansion in deployed Physical AI systems.
- Intermittent EMBED_RUNAWAY with AR0235 TOP2 metadata on Jetson AGX Thor (R39.2.1)nvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
Intermittent EMBED_RUNAWAY is reported when using AR0235 TOP2 metadata on Jetson AGX Thor with R39.2.1. The issue is intermittent, suggesting a timing/driver/metadata handling instability in the camera capture stack.
⚡ Camera pipeline stability on Jetson directly impacts real-world perception reliability, which is foundational for Physical AI deployments.
- Orin Nano (JetPack 7.2.1 / L4T R39.2.1) Race Conditionnvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
A race condition is reported on Orin Nano with JetPack 7.2.1 / L4T R39.2.1. The post provides no additional details, but it indicates intermittent concurrency-related instability on this platform/software combination.
⚡ Reliability of the Jetson software stack is a gating factor for deploying Physical AI systems at the edge without brittle workarounds.
- Guidance Required: Multi-Sensor Synchronization and Sensor Fusion on Jetson Orin NX (JetPack 7.2.1)nvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
User requests guidance for multi-sensor synchronization and sensor fusion on Jetson Orin NX using JetPack 7.2.1. The post indicates uncertainty about best practices or supported approaches for coordinating multiple sensors and fusing their data.
⚡ Developers need clearer, robust platform guidance for time synchronization and sensor fusion on Jetson to reliably deploy multi-sensor Physical AI stacks at the edge.
- Increasing PCIe XDMA C2H backpressure causes FPGA FIFO overflownvidia-forum:robotics-edge-computing11/5/2026other
Increasing PCIe XDMA C2H backpressure reportedly causes an FPGA FIFO overflow. This suggests flow-control/backpressure handling issues that can lead to data loss or instability in FPGA-to-host transfers.
⚡ PCIe data-path reliability issues between Jetson/host and FPGA can block real-time sensor/actuator pipelines needed for Physical AI deployment and hardware-in-the-loop validation.
- Organize and combine solver coupling examplesgithub:newton-physics/newton11/5/2026other
Request to organize and combine solver coupling examples in the newton-physics/newton repository. This implies current examples may be fragmented or harder to discover and reuse.
⚡ Better organization of solver coupling examples improves developer learning loops and increases adoption of advanced physics capabilities in Physical AI workflows.
- Recommended way to show product / module name on the first-boot OOBE page (Jetson Thor, R39.2)nvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
User asks the recommended way to show a product/module name on the first-boot OOBE page for Jetson Thor on R39.2. The corpus provides only the title.
⚡ Configurable first-boot/OOBE branding affects OEM adoption and smooth deployment of NVIDIA-based edge systems.
- The system hangs while running the fTPM stress test script on the Orin Nanonvidia-forum:robotics-edge-computing11/5/2026crashes-stabilitycrashhardware
Report that the system hangs while running an fTPM stress test script on Orin Nano. The corpus includes only the title, so root cause and scope are unknown.
⚡ System hangs during security module stress tests can halt qualification and deployment timelines for edge Physical AI devices.
- Jetson Thor T5000 Module - 10G Ethernet PHY Implementation Referencenvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
Request for a 10G Ethernet PHY implementation reference for the Jetson Thor T5000 module. The corpus contains only the title.
⚡ Availability of hardware reference guidance (10G PHY) reduces integration risk and accelerates time-to-market for Jetson Thor platforms.
- [ORIN DEVKITS_P3737] ETH RESET TIMINGnvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
Question about ETH reset timing on ORIN devkits (P3737). The corpus includes only the title without timing diagrams, constraints, or failure modes.
⚡ Ethernet reset timing uncertainties on Orin devkits can cause bring-up delays and reliability issues for deployed edge robotics systems.
- ORIN NANO 8G HDMI 显示电路 高温显示异常，串阻设计nvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
Report of abnormal HDMI display behavior at high temperature on ORIN NANO 8G, mentioning display circuitry and series resistor design. The corpus contains only the title in Chinese with no additional details.
⚡ Display instability at high temperature affects robustness expectations for edge deployments and requires clear design guidance for carrier boards.
- Orin Nano 8G 下电时序确认nvidia-forum:robotics-edge-computing11/5/2026hardware-integrationhardware
User is asking to confirm the power-down sequencing for Orin Nano 8G. Lack of a clear, authoritative sequence can block correct carrier-board design and system reliability validation.
⚡ Power sequencing clarity is foundational for reliable Jetson-based Physical AI deployments and partner hardware bring-up.
- Jetson TX2 latest EU RF Test reportsnvidia-forum:robotics-edge-computing11/5/2026docs-onboardingjetson-tx2rf-complianceeuregulatorydocumentationedge-deployment
User is requesting the latest EU RF test reports for Jetson TX2. Missing or hard-to-find compliance documents slow procurement and certification workflows.
⚡ Regulatory documentation availability affects Jetson product adoption and deployment readiness for Physical AI systems.
- Alpamayo 2 Edge (2B): expected release date for weights,?nvidia-forum:robotics-edge-computing11/5/2026docs-onboardingalpamayo-2-edgemodel-weightsrelease-timelineedge-aideveloper-support
User is asking for an expected release date for Alpamayo 2 Edge (2B) weights. Lack of timing information blocks planning for evaluation and integration.
⚡ Clear release timelines for model weights drive developer planning and adoption of NVIDIA-provided edge AI components.
- MIPI CSI-2 D-PHY Internal Eye Scan and RX Margin Validation on Jetson Thornvidia-forum:robotics-edge-computing11/5/2026sensors-perceptionjetson-thormipi-csi2d-phycamerasignal-integrityrx-margin
User is asking about MIPI CSI-2 D-PHY internal eye scan and RX margin validation on Jetson Thor. The request implies a need for test methodology, tooling, or platform support to validate camera signal integrity.
⚡ Sensor link validation tooling and guidance accelerates camera bring-up, a prerequisite for robust Physical AI perception stacks.
- How to Add Custom Kernel Modules to the Jetson Installer ISO (R39.2.1)nvidia-forum:robotics-edge-computing11/5/2026docs-onboardingjetsonl4t-r39-2-1kernel-modulesinstaller-isocustomizationdeployment
User asks how to add custom kernel modules to the Jetson Installer ISO for R39.2.1. The question suggests missing or unclear documentation/process for integrating out-of-tree modules into the installer media.
⚡ Custom kernel module workflows are a common edge deployment need and smooth paths reduce time-to-integration for Physical AI hardware stacks.
- float16 atan2 loses gradients after coordinate scalinggithub:NVIDIA/warp11/4/2026renderingrenderinghardwaredeploymentwarp
wp.atan2 can return zero gradients in float16 after scaling inputs, even though the forward result is correct. This silently breaks gradient-based optimization and causes SGD-style checks to stall.
⚡ If half-precision autodiff is unreliable, it undermines scalable optimization and learning workloads that depend on Warp for Physical AI training loops.
- Avoid redundant FP64 diagonal work in native tile Choleskygithub:NVIDIA/warp11/4/2026hardware-integrationhardwarewarp
wp.tile_cholesky redundantly recomputes diagonal FP64 work across threads in a block, making small-tile factorization slower than necessary. This is especially costly on GPUs with limited FP64 throughput like the L40.
⚡ Kernel-level inefficiencies in math primitives compound across large-scale simulation/optimization workloads, reducing platform efficiency on deployed GPUs.
- Add wp.tile_lu with cuSOLVERDx accelerationgithub:NVIDIA/warp11/4/2026hardware-integrationhardwareintegrationwarp
Request to add wp.tile_lu for general dense tiles inside Warp kernels, using cuSOLVERDx via libmathdx when available with a native fallback. The scope includes float32/float64 and partial pivoting with exposed pivot/permutation info.
⚡ Adding LU tile factorization with cuSOLVERDx enables more general in-kernel solves that benefit physics, optimization, and perception geometry workloads.
- Unaddressed issue in the Isaac Ros reponvidia-forum:robotics-edge-computing11/4/2026docs-onboardingrobotics-edge-computingisaac-rosrepo-maintenancesupport-triageonboarding
A forum post reports an unaddressed issue in the Isaac ROS repo. The lack of detail aside, it signals support/triage friction for users relying on Isaac ROS.
⚡ Unresolved repo issues and unclear ownership slow adoption and integration of the Isaac ROS stack into Physical AI applications.
- [BUG] wp.atomic_cas() on float arrays handles -0.0 and NaN differently on CPU and CUDAgithub:NVIDIA/warp11/4/2026renderingrenderinghardwarewarp
wp.atomic_cas on float arrays behaves differently on CPU vs CUDA because CUDA uses bitwise equality while CPU uses numerical equality. This causes mismatches for -0.0 and NaN, including CAS loops that never succeed on CPU for NaN.
⚡ Cross-device determinism and consistent atomics semantics are foundational for reproducible simulation and learning pipelines on NVIDIA GPUs.
- Jetson TX2 EFI boot via SPI possible?nvidia-forum:robotics-edge-computing11/4/2026hardware-integrationjetson-tx2efispi-bootbootloaderrobotics-edge
User asks whether Jetson TX2 can EFI boot via SPI, but provides no further details. This indicates a need for clearer documentation or support around TX2 boot options.
⚡ Clear boot/firmware guidance for Jetson devices reduces deployment risk for edge Physical AI systems.