Files
hyprland-headless-stream-pl…/spikes/capture-speed/results/final-motion/py-screencopy.output-jpeg.run2.stdout
T
Hermenix 36efa316fd spike: the verdict, the table, the chart -- in-process pywayland wins on zwlr-screencopy
Four routes measured on one frozen contract at one session's container size, with
RAM and CPU as first-class results: in-process pywayland is ~1.5x faster (78 vs
117 ms), uses ~2.8x less of our own CPU, and nearly halves the compositor's
per-frame CPU (67 vs 116 ms). grim-rs buys nothing and costs 746 MB RSS.
ext-image-copy-capture measures identical to screencopy and appears in a
compositor crash stack, so the deprecated-but-working protocol is the pick.

Also fixes the client's unclean exit (pywayland teardown segfaulted after the
result was printed, 16 times, one batch coinciding with a compositor crash).
2026-10-11 17:08:13 +00:00

2 lines
814 B
Plaintext

{"route": "pywayland-in-process", "language": "python", "protocol": "zwlr-screencopy-v1", "output": "hh-chromium-gqkxz3", "geometry": [1280, 720], "mode": "jpeg", "warmup": 10, "frames": 200, "frame_ms": {"min": 39.218, "p50": 76.971, "mean": 71.88, "p95": 95.795, "max": 149.48}, "stage_ms": {"capture": 69.53, "encode": 2.305, "stitch": 0.0}, "fps_sustained": 13.91, "wall_ms": 14378.78, "bytes_p50": 105813, "persistent": true, "notes": "one connection, buffers reused, turbojpeg q80 in-process; 1 output(s); forcing a repaint per frame", "cpu_ms_per_frame": 1.782, "rss_peak_mb": 29.3, "rss_peak_kb_raw": 30000, "system_cpu_ms_per_frame": 143.05, "compositor_cpu_ms_per_frame": 67.55, "machine_cpu_pct": 199.0, "machine_cpu_pct_of_all_cores": 99.5, "frames_bytes_allocated_p50": 3686400, "resource_notes": ""}