Skip to content

feat(core): per-realm tick rate — fixed-step at a declared hz - #257

Open
NathanWalker wants to merge 3 commits into
pocket-stack:mainfrom
NathanWalker:feat/tick-rate
Open

feat(core): per-realm tick rate — fixed-step at a declared hz#257
NathanWalker wants to merge 3 commits into
pocket-stack:mainfrom
NathanWalker:feat/tick-rate

Conversation

@NathanWalker

@NathanWalker NathanWalker commented Aug 10, 2026

Copy link
Copy Markdown

Stacked on #256 (which stacks on #255) — review the top commit only; rebases as the stack merges. Delivers the ProMotion follow-up #256's description records.

What

The 1/60 s step becomes a per-realm constant declared before the first tick. Sessions stay fixed-step and deterministic at their declared rate; every default remains 60 and the 60 Hz path is bit-for-bit the original — goldens, tapes, and existing bundles are untouched.

  • Core: Ui::set_tick_rate(hz) (gated on frame == 0) threads dt through the spring integrators and ms→frame conversions; exact integer hz is kept alongside dt so frame counts stay byte-stable.
  • Surface/ABI: UiSurface::set_tick_rate, pocket_apple_set_tick_rate / pocket_apple_core_set_tick_rate (1..=240, ERR_BAD_STATE after the realm ticks), and PocketSurfaceView.tickRate pinning preferredFrameRateRange to the declared rate.
  • Guest virtual time bakes like glyphs do: tools/build.ts --hz defines __POCKET_TICK_HZ__; clock, kinetics, input, and deepzoom derive their per-tick constants from it (rebased as x ** (60/hz) with an explicit === 60 early-return so the default path keeps the original literals). A bundle built at one rate must run at that rate.
  • CLI: pocket ios build|stage|play --hz=60|120 stages tickHz through current.json; the shell pins the surface to it. CADisableMinimumFrameDurationOnPhone added to the shell plist (required for >60 on iPhone).
  • Hero's headline now reads the baked rate (JSX at ${TICKS_PER_SECOND} FPS.).

Validation

  • iPhone 16 Pro (ProMotion), 120 Hz bundles: hero's wall-clock FPS counter reads 120 on device; springs, sprite motion, and the guest↔host channel behave identically to 60 Hz — just at 120 steps/s.
  • Simulator honesty check: simulators report maximumFramesPerSecond = 60, so a 120 realm there renders at half speed by design — device-only evidence.
  • engine/core: 112 tests green (+2 new: the pre-first-tick gate, 120 Hz tween frame counts). cargo build -p pocket-apple clean, bunx tsc --noEmit clean, canonical bun tools/test.ts unit stages match the pre-change baseline, VALID_HZ at 60 is byte-identical to the old literal.
  • Known drift risk (pre-existing): hosts/web/engine.js hardcodes its own 60-anchored VALID_HZ with no import path to clock.ts — untouched here, noted for when the web host wants non-60 realms.

NathanWalker and others added 3 commits August 9, 2026 11:51
…xternal-guest mode

A new engine/apple workspace member packaging modern iOS as a PocketJS
host. The composition mirrors hosts/pocketbook: one pocket_mod::Guest
realm, one pocket_ui_surface::UiSurface, and pocketjs_core::raster driven
incrementally through a DamageTracker, exposed behind a small C ABI.

- engine/apple/src/lib.rs — guest-owning mode: create/load_pak/eval_bundle/
  frame/render/hit_test_bounds/destroy, plus an effect channel over the
  ui.svc* ops (set_effect_callback drains guest svcSend lines during frame;
  post_event queues lines for the guest's next svcPoll).
- engine/apple/src/core_host.rs — external-guest mode: pocket_apple_core_*
  owns only the core, pak feed, raster pipeline, and svc queues, for hosts
  whose JS engine lives elsewhere (demonstrated with the NativeScript
  runtime evaluating the guest bundle in its own context).
- engine/apple/apple/PocketSurfaceView.{h,m} — CADisplayLink capped at
  60 Hz, latched touch contacts (a down+up between two ticks still reaches
  the guest as one present frame then a release), aspect-fit inverse touch
  mapping, damage-gated compositing of the ARGB32 framebuffer.
- engine/apple/build-xcframework.sh — clang-linked dynamic framework per
  slice (device arm64 + simulator arm64), no Xcode project.
- pocket-ui-surface additionally mounts hitTestBounds (spec op 42), the
  touch-path hit authority the gesture layer prefers over the ink-claiming
  hitTest.
- apps/nsengine — reference guest: an effect driver over svcSend, a
  per-frame poll pump, a focusable pressable button, and a platform-reach
  probe that distinguishes a sidecar realm from an embedding-runtime host.

rquickjs uses its bindgen feature: no pregenerated bindings exist for
aarch64-apple-ios targets. Validation: the render_hero example drives the
ABI end to end — 180 frames of apps/hero/main.tsx at 480x272 density 2
render non-blank and byte-identical across two independent instances.
Build guests with --density matching the surface density (glyphs bake at
build time; density 4 supersamples cleanly on 3x screens).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…ll, play on the simulator

The CLI follow-up feat/apple-host's description names: a dev-profile target
id and a first-class run flow for the Apple host. `pocket play ios nsengine`
builds a guest from a resolved plan, stages it into a committed NativeScript
shell, and launches it on an arm64 iOS simulator.

tools/ios-profile.ts registers the transitional ios-dev profile (hostAbi 7,
platform ios, form embedded, fixed 480x272, raster density 1..4 as a registry
input, capabilities input.touch + text.glyphs.baked) following the
iphone2g-profile pattern: out of POCKET_TARGETS until device acceptance.
apps/nsengine gains a committed manifest — the root-template fallback demands
input.buttons, which this surface cannot advertise.

PocketSurfaceView now applies pocket_apple_set_identity("ios-dev", 7) at
init: plan-built bundles bake __POCKET_TARGET__/__POCKET_HOST_ABI__ and
assertNativeHostContract refuses hosts that publish nothing; the C ABI
existed but nothing called it. External-guest hosts (@nativescript/pocketjs)
mount the same pair on their ui namespace, and a source-text test guards the
agreement.

tools/ios.ts owns the flow — doctor/setup (symbian shape; Rust targets are
the only mutation, and only --rebuild-native needs them since the published
plugin ships a prebuilt PocketApple.xcframework), devices, native, build,
stage, and play (admissible-simulator pick, simctl boot, ns run ios
--device <udid> --no-hmr --justlaunch). --plugin-path/--runtime-tgz point the
shell at local builds for pre-publish validation and restore the committed
template afterwards. pocket play ios delegates here; bin.mjs gains the ios
passthrough.

hosts/apple/ns-shell is the committed shell: plan-driven (viewport, density,
and guest mode read from the staged plan + current.json), answers the ns.ping
service channel, and runs in place so repeat runs rebuild incrementally
(~47 s cold, ~16 s warm on Apple Silicon). Its tsconfig pins
@nativescript/core paths so the plugin's typings resolve from a file:
symlink.

Validated on the iOS 26.5 simulator in both guest modes: unprompted ns.ping
round trip renders pong 1, and --external-guest reads
UIDevice.currentDevice.systemVersion from guest code. Unit stage green
including tests/ios-profile.test.ts, the nsengine admission-matrix row, and
the CLI dispatch cases.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The 1/60 s step becomes a per-realm constant chosen before the first tick:
Ui::set_tick_rate threads dt through the spring integrators and ms-to-frame
conversions (exact integer hz kept alongside dt so frame counts stay
byte-stable), UiSurface and the pocket-apple C ABI expose it
(pocket_apple_set_tick_rate / pocket_apple_core_set_tick_rate, gated on the
first tick), and PocketSurfaceView pins its CADisplayLink to the declared
rate. Guest-side virtual time bakes the same way glyphs do: tools/build.ts
--hz defines __POCKET_TICK_HZ__, clock/kinetics/input/deepzoom derive their
per-tick constants from it (the 60 path stays bit-for-bit the original), and
pocket ios gains --hz=60|120 staged through current.json to the shell. Hero's
headline reads the baked rate. Defaults everywhere remain 60, so existing
goldens, tapes, and bundles are unchanged.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
@NathanWalker
NathanWalker marked this pull request as ready for review August 10, 2026 00:26
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant