4fcce5bff9
Add /version.json build identity, deploy 16e7930 release markers,
DNS/Named Tunnel prep docs, portable browser benchmark, and replay
stability probe. Keep Quick Tunnel until permanent domain cutover.
Co-authored-by: Cursor <cursoragent@cursor.com>
34 lines
911 B
Markdown
34 lines
911 B
Markdown
# PERFORMANCE_AFTER_MAXIMUM_DEMO
|
||
|
||
## Production identity
|
||
|
||
| Field | Value |
|
||
| ----- | ----- |
|
||
| Release | `20260715-2215` |
|
||
| Bundle | `index-AagIOJbd.js` |
|
||
| Commit | `16e7930` (via `/version.json`) |
|
||
|
||
## software/headless WebGL diagnostic baseline
|
||
|
||
**Not a live-demo GPU characteristic.**
|
||
|
||
```text
|
||
Average FPS: ~8.9–9.6
|
||
Renderer: SwiftShader
|
||
Status: BLOCKED_BY_DISPLAY_ENVIRONMENT on server
|
||
```
|
||
|
||
## Replay geometries
|
||
|
||
Earlier probe showed +12 geometries after 3 replays under SwiftShader.
|
||
Re-run on production (`npm run perf:replay-stability`): **STABLE** (147 geos flat through 10 replays; +3 after GC pause only).
|
||
|
||
## Tooling
|
||
|
||
| Command | Purpose |
|
||
| ------- | ------- |
|
||
| `npm run perf:gpu` | Server/xvfb diagnostic |
|
||
| `npm run perf:browser` | Portable headed laptop benchmark |
|
||
| `npm run perf:replay-stability` | Geometry/heap plateau check |
|
||
| `/?perf=1` → Export benchmark | Local JSON download |
|