Patch friction vs point friction

03 · Contact and friction

A box rests on a slope at 98% of its friction limit (tan θ = 0.49, μ = 0.5). Point friction gives each of the four contacts its own friction cone, so the load shifts between them and static friction creeps. Patch friction, the default, pools the contacts of a body pair into a region whose friction acts at up to two anchors. The anchors carry their tangential displacement across steps and pull it back, so a held object does not creep.

drag to orbit · scroll to zoom

0.00 s

Cost

Time per step, CUDA graph replay, median of 7 interleaved rounds on one NVIDIA RTX PRO 6000 Blackwell Server Edition; ratios are against the first row. Solver allocation is the GPU memory the solver allocates at construction, before warm-up and graph capture. Scene: the 8-box stack in each world, 8 steps of 1/240 s per frame.

Option	1 world	64 worlds	1024 worlds	solver allocation, 1024 worlds	result in this tutorial
patch friction (default)	1.00 ms	1.20 ms	1.65 ms	215 MB	slope creep 0.45 µm in 10 s
point friction	1.38 ms 1.38×	1.52 ms 1.27×	1.78 ms 1.08×	132 MB	slope creep 2.24 mm

What to look at

The point-friction box creeps about 2 mm in 10 s.
The patch-friction box stays within a micrometre.

Key parameter

friction_anchor_beta

Position gain of the friction anchors. The default None means 0.2; 0 selects point friction, two tangent rows per contact.

Try changing

Raise the slope to tan θ = 0.51: both boxes must slide.
Try friction_anchor_beta=0.05 and 0.5.

Concept

Friction patches and anchors

What it is

A patch groups the contacts of one body pair that share a normal, plane and friction coefficient. Its friction acts at up to two anchors that share the pooled normal load and remember how far they have been displaced; the friction rows push that displacement back with gain β. An anchor is released when its patch unloads or slides at the edge of the friction cone.

tangent target = β · displacement / h

In this scene

At 98% of the friction limit, point friction creeps 2.24 mm in 10 s; the patch holds the box to 0.45 µm.

Considerations

Cost: Here patches were cheaper than point friction: 1.0 vs 1.4 ms per step in one world, 1.6 vs 1.8 ms in 1024 worlds, since two anchors replace two rows per contact.
Interactions: Contact compliance and friction_mode other than "fixed_normal" switch to point friction. Call reset after teleporting a body to drop its anchors.
Use it when: Default. Point friction (β = 0) suits scenes that need per-contact friction forces.

The script

cd tutorials && python 03_patch_friction.py prints each variant's final metric; --record out.json also writes the replay. Every tutorial imports replay.py from the same folder.

03_patch_friction.py
replay.py

# SPDX-FileCopyrightText: Copyright (c) 2026 Skild AI and the Steel contributors
# SPDX-License-Identifier: Apache-2.0

"""Patch friction vs point friction: a box at 98% of its friction limit on a slope."""

import math

import newton
import numpy as np
import warp as wp
from newton.solvers import SolverSteel

from replay import Recorder

DT = 1.0 / 240.0
SUBSTEPS = 8  # 30 Hz frames
MU = 0.5
SLOPE = math.atan(0.49)  # tan(slope) = 0.98 * MU


def build():
    builder = newton.ModelBuilder()
    tilt = wp.quat_from_axis_angle(wp.vec3(0.0, 1.0, 0.0), -SLOPE)
    builder.add_shape_box(
        -1,
        xform=wp.transform((0.0, 0.0, 0.3), tilt),
        hx=0.8,
        hy=0.3,
        hz=0.05,
        cfg=newton.ModelBuilder.ShapeConfig(mu=MU),
    )
    normal = np.array(wp.quat_rotate(tilt, wp.vec3(0.0, 0.0, 1.0)))
    start = np.array([0.0, 0.0, 0.3]) + normal * 0.11
    box = builder.add_body(xform=wp.transform(start, tilt))
    builder.add_shape_box(
        box, hx=0.12, hy=0.08, hz=0.06, cfg=newton.ModelBuilder.ShapeConfig(mu=MU)
    )
    return builder.finalize(), start


def simulate(rec, label, **options):
    model, start = build()
    solver = SolverSteel(model, **options)
    pipeline = newton.CollisionPipeline(model)
    contacts = pipeline.contacts()
    state_0, state_1, control = model.state(), model.state(), model.control()

    run = rec.variant(label, model)
    for _ in range(rec.frames(10.0)):
        slip = np.linalg.norm(state_0.body_q.numpy()[0, :3] - start)
        run.frame(state_0, slip_mm=1000.0 * slip)
        for _ in range(SUBSTEPS):
            pipeline.collide(state_0, contacts)
            solver.step(state_0, state_1, control, contacts, DT)
            state_0, state_1 = state_1, state_0
    run.check(solver)


if __name__ == "__main__":
    rec = Recorder()
    simulate(rec, "point friction (friction_anchor_beta=0)", friction_anchor_beta=0.0)
    simulate(rec, "patch friction (default)")
    rec.save()

# SPDX-FileCopyrightText: Copyright (c) 2026 Skild AI and the Steel contributors
# SPDX-License-Identifier: Apache-2.0

"""Record Steel tutorial runs as replay JSON for the tutorials page, or print a summary.

Every tutorial imports this module. ``python NN_name.py`` runs the variants and prints
the final metric of each; ``--record out.json`` also writes body poses, shapes and metric
series that the page replays in 3D.
"""

import argparse
import base64
import json
import os
import subprocess
import sys
import time

import newton
import numpy as np
import warp as wp

from source_hash import source_sha256

_RENDERED_TYPES = {
    newton.GeoType.PLANE,
    newton.GeoType.SPHERE,
    newton.GeoType.CAPSULE,
    newton.GeoType.ELLIPSOID,
    newton.GeoType.CYLINDER,
    newton.GeoType.BOX,
    newton.GeoType.MESH,
    newton.GeoType.CONE,
    newton.GeoType.CONVEX_MESH,
}


class Recorder:
    """Collect the variants of one tutorial; ``fps`` is the replay frame rate."""

    def __init__(self, fps: int = 30):
        parser = argparse.ArgumentParser()
        parser.add_argument(
            "--record", metavar="OUT.json", help="write a replay file for the tutorials page"
        )
        parser.add_argument("--device", default=None, help="Warp device, for example cuda:0")
        self.args = parser.parse_args()
        if self.args.device:
            wp.set_device(self.args.device)
        self.fps = fps
        self.script = os.path.abspath(sys.argv[0])
        self.variants: list[Variant] = []

    def frames(self, seconds: float) -> int:
        """Number of replay frames in ``seconds``, including the first."""
        return int(round(seconds * self.fps)) + 1

    def variant(
        self,
        label: str,
        model: newton.Model,
        *,
        ghost: bool = False,
        slot: int | None = None,
        scene: str | None = None,
    ) -> "Variant":
        """Start one run; ``ghost`` overlays run ``slot``, ``scene`` groups runs for charts."""
        variant = Variant(label, model, self.fps, ghost, slot, scene)
        self.variants.append(variant)
        return variant

    def save(self) -> None:
        """Print a summary and, with ``--record``, write the replay file."""
        for variant in self.variants:
            last = variant.metrics[-1] if variant.metrics else {}
            flag = "  (rows dropped)" if variant.overflow else ""
            print(f"{variant.label:<48} {json.dumps(last)}  {variant.wall:.2f} s{flag}")
        if not self.args.record:
            return
        source_sha = source_sha256(self.script)
        doc = {
            "script": os.path.basename(self.script),
            "script_sha256": source_sha,
            "newton_commit": _newton_commit(),
            "newton_version": newton.__version__,
            "warp_version": wp.config.version,
            "gpu": wp.get_device().name,
            "fps": self.fps,
            "variants": [v.to_json() for v in self.variants],
        }
        with open(self.args.record, "w") as f:
            json.dump(doc, f, separators=(",", ":"))
        print(f"wrote {self.args.record}")


class Variant:
    """One recorded run: a pose per frame for every body, plus named metric values."""

    def __init__(self, label, model, fps, ghost, slot, scene):
        self.label, self.model, self.fps, self.ghost, self.slot = label, model, fps, ghost, slot
        self.scene = scene
        self.poses: list[np.ndarray] = []
        self.metrics: list[dict] = []
        self.awake: list[list[int]] = []
        self.particles: list[np.ndarray] = []
        self.overflow = False
        self.start = time.perf_counter()
        self.wall = 0.0

    def frame(self, state: newton.State, awake=None, particles=None, **metrics) -> None:
        """Record ``state.body_q``, optional awake flags, particle positions and metric values."""
        self.poses.append(state.body_q.numpy().astype(np.float32).copy())
        if particles is not None:
            self.particles.append(particles.numpy().astype(np.float32).copy())
        self.metrics.append({k: _round(v) for k, v in metrics.items()})
        if awake is not None:
            self.awake.append([int(x) for x in awake.numpy()])
        self.wall = time.perf_counter() - self.start

    def check(self, solver) -> None:
        """Note whether the solver dropped rows; call once after the run."""
        try:
            solver.check_constraint_capacity()
        except RuntimeError:
            self.overflow = True

    def to_json(self) -> dict:
        poses = (
            np.stack(self.poses)
            if self.poses
            else np.zeros((0, self.model.body_count, 7), np.float32)
        )
        return {
            "label": self.label,
            "fps": self.fps,
            "ghost": self.ghost,
            "slot": self.slot,
            "scene": self.scene,
            "bodies": int(self.model.body_count),
            "frames": int(poses.shape[0]),
            "poses": base64.b64encode(poses.astype(np.float32).tobytes()).decode(),
            "shapes": _shape_table(self.model),
            "metric": self.metrics,
            "awake": self.awake or None,
            "particles": self._particles_json(),
            "wall_s": round(self.wall, 3),
            "overflow": self.overflow,
        }

    def _particles_json(self):
        if not self.particles:
            return None
        frames = np.stack(self.particles)
        tris = self.model.tri_indices
        radius = self.model.particle_radius.numpy()
        return {
            "count": int(frames.shape[1]),
            "frames": base64.b64encode(frames.astype(np.float32).tobytes()).decode(),
            "radius": round(float(radius.mean()), 5) if radius.size else 0.0,
            "tris": [int(v) for v in tris.numpy().reshape(-1)]
            if tris is not None and tris.shape[0]
            else None,
        }


def _shape_table(model):
    types = model.shape_type.numpy()
    scales = model.shape_scale.numpy()
    xforms = model.shape_transform.numpy()
    bodies = model.shape_body.numpy()
    flags = model.shape_flags.numpy()
    shapes = []
    for i, t in enumerate(types):
        if int(t) not in _RENDERED_TYPES or not flags[i] & int(newton.ShapeFlags.VISIBLE):
            continue
        shape = {
            "b": int(bodies[i]),
            "t": int(t),
            "s": [round(float(v), 5) for v in scales[i]],
            "x": [round(float(v), 6) for v in xforms[i]],
        }
        if int(t) in (newton.GeoType.MESH, newton.GeoType.CONVEX_MESH):
            mesh = model.shape_source[i]
            shape["v"] = [round(float(v), 5) for v in np.asarray(mesh.vertices).reshape(-1)]
            shape["i"] = [int(v) for v in np.asarray(mesh.indices).reshape(-1)]
        shapes.append(shape)
    return shapes


def _round(value):
    value = float(value)
    return round(value, 5) if np.isfinite(value) else None


def _newton_commit():
    if os.environ.get("NEWTON_COMMIT"):
        return os.environ["NEWTON_COMMIT"]
    try:
        root = os.path.dirname(os.path.dirname(newton.__file__))
        return subprocess.check_output(["git", "-C", root, "rev-parse", "HEAD"], text=True).strip()
    except (OSError, subprocess.CalledProcessError):
        return None

Recorded with SolverSteel on Newton 1.9.0.dev0, Warp 1.19.0, NVIDIA RTX PRO 6000 Blackwell Server Edition. Replays read the state back every frame, so they are not a speed benchmark. Script · page text

Steel docs and tutorials are licensed under Apache-2.0. Third-party notices. "Skild", "Skild AI" and the Skild logo are trademarks of Skild AI and are not licensed under Apache-2.0.