Publish sanitized applied tick membership measurements with CI provenance

This commit is contained in:
Ludovic Marie
2026-10-04 01:52:49 +02:00
parent e091816ce0
commit f6b36854ed
2 changed files with 46 additions and 0 deletions
@@ -0,0 +1,41 @@
{
"schema": 1,
"platform": "Linux-6.17.0-1022-azure-x86_64-with-glibc2.39",
"architecture": "x86_64",
"minecraft": "26.3",
"distant_horizons": "3.3.4",
"results": [
{
"enabled": false,
"allocated_bytes": 245760208,
"cpu_ns": 423649902,
"wall_ns": 481491629,
"nodes_per_tick": 512,
"measured_ticks": 20000,
"verified_transitions": 6000,
"checksum": 10240000
},
{
"enabled": true,
"allocated_bytes": 208,
"cpu_ns": 185922780,
"wall_ns": 192151484,
"nodes_per_tick": 512,
"measured_ticks": 20000,
"verified_transitions": 6000,
"checksum": 10240000
}
],
"notes": [
"Actual publisher tick holder under Fabric, warmed with preallocated nodes and a reused action list.",
"Checks duplicate/null membership, enable/disable precedence, published iteration order and clear boundaries.",
"Only private membership checks change; action collections and traversal remain upstream.",
"CPU and allocation measure the calling thread in one isolated loop; no statistical speed claim.",
"Not a game FPS, visible terrain completion, whole-game RAM or Mac GPU benchmark."
],
"provenance": {
"source_commit": "e091816ce06686d64d0f7a181a42701aa67e5bf7",
"ci_run": "https://github.com/ludovic111/minecraft-optimization-lab/actions/runs/37162939262",
"artifact": "horizon-ubuntu-latest"
}
}
+5
View File
@@ -37,3 +37,8 @@ statistical speedup, game FPS gain, whole-game RAM reduction, completed terrain
or performance on an M2 GPU. Actual game profiles and repeated Mac comparisons
are required for those conclusions. Table capacity remains proportional to the
largest tick membership, as the upstream hash table already did.
The [sanitized Linux CI result](../benchmarks/tick-membership-2026-10-04/linux-applied-hooks.json)
includes its source commit and workflow provenance. It independently reproduced
the same allocation counts and transition checks. Its CPU timings are another
single warmed observation, not a statistical estimate or Mac result.