Commit graph

3 commits

Author SHA1 Message Date
Antoine
b9d6319302
The page stays free while big networks play: ambient map at 30 fps, saves in the worker, less to carry back
The game plays exactly as before; only what holds the page changes.

* mapStill() tells three cases apart: still (drawn once a second), ambient (only the animation clock moved: drawn at
  most every 29 ms, so 30 fps on 60 and 120 Hz screens) and moving (camera, pointer, selection, network: every frame).
  The animations follow the clock, so they keep their speed.
* save() skips the storage write when the text is the one the slot already holds (SLOT.txt) and no write is on its
  way; leaving the page still writes. A big autosave that moved on is packed, turned into text, stamped and
  compressed in the route worker (rwSave(), rwGzip()); the page only writes what comes back. A lost worker hands its
  pending saves back to the page (rwSaveBack(), rwGzBack()).
* Less crosses back from the worker: R.efg stays there while R is its own last answer (RW.wR, efg 0), R.egDesc comes
  as 0 or as the entries that changed (rwDescOut()/rwDescIn()), what crossed each link travels as typed arrays decoded
  on demand (rwLOut()/rwLIn()), and the full ports (hotMap) come along with R. Every message is read in order
  (rwMsg()), so both sides always hold the same last answer.
* Idle frames build the diverse-paths map and the latency floors a millisecond at a time (diversityWarm(),
  latWarm()), before the first offer asks.
* Diagnostic info shows how fast days really come: speed asked, speed held, how long a day takes (15-day-pace.js).
* perf-bench: "The NOC intern's shift: opsDay()" and, in Chromium, "One game day in play: page time only", a day
  played with the real route worker counting only the time it holds the page.

Tests: frame-pacing, day-pace, save-size, persist, route-worker (efg as typed arrays, hotMap from the worker),
day-steps (a second page plays the worker: answers without R.egDesc, with only its changes, posts without R.efg,
renewal haggles answered through the worker), intern-dark-fiber-worker, webgl-zoom-sprites, speed.
2026-11-07 22:22:51 +01:00
Antoine
1e9b1d458b
GPU benchmark: isolate trials, time the card, say which map drew
- A trial now ends before the next starts: the clock stops, the frame
  callback is cancelled (and a queued one turned away), a day still in the
  route worker is waited for (or dropped), the clock's leftovers are cleared,
  and every trial starts from the same random numbers.
- The renderers take turns going first, scenario by scenario; ?reps=N runs
  each N times and reports medians.
- draw() is labelled as CPU submission time; the card's own time comes from
  EXT_disjoint_timer_query_webgl2 where available (read frames later, never
  waited for, disjoint results dropped); a run drawn by the other map is
  flagged.
- npm run perf --browser reads its sync pixel back from the surface the map
  was drawn on (the WebGL canvas waits for the card) and says which map each
  version drew with.
2026-11-04 22:37:13 +01:00
Frederic Jung
c76dd4d556
Benchmark in real Chromium too, and only call a row slower when the samples say so
npm run perf -- --browser times the same workloads in Chromium as a desktop
and as a phone (390x844 at 3x, CPU slowed 4x), one browser per version. Each
run collects garbage first and ends by forcing layout and the map canvas, so
deferred work counts. The jsdom DOM is about 30x slower than a browser's and
has no layout, so the interface rows need this.

Shared machines swing single runs by 30% on identical code, CPU time
included. A row now counts as slower or faster only when it moved 15% and a
Mann-Whitney U test gives p < 0.01 (15 runs a side by default). Identical
code: no flags in three full runs; a routing slowdown planted in a copy is
flagged in jsdom and on the phone.

The panel workload renders both selections every run instead of alternating,
which split its samples in two. CI drives the runner's Chrome through
playwright-core, no browser download.
2026-10-26 12:04:57 +01:00