T3 Code · Performance over time

Main thread involuntary context switches

All metrics
41 / 120
Main thread involuntary context switches (lower is better)
3.26/s
18 Nov+2.1% vs the median of up to 7 previous complete nightlies, within noiseFork perf.9 0.57 /s, −83.5% vs its upstream base
AgentsRange

When upstream merges a fix, the lines meet.

Upstream nightlyPerformance forkNoise bandRange of attempts green = better, red = worse, beyond the noise band

As agents grow.

Involuntary context switches of the main thread per second of the window.

Lower is better. Noise band 57.1% on Linux x64 (n = 6), so a change counts beyond 62.1%.

Unit
/s
Better
lower
Noise band
57.1%, threshold 62.1% (n = 6)
Recorded from
17 Oct 2026; before that the series is a gap, never a zero
Platforms
Linux x64
Lane
Every record, inside the agent ladder
Upstream builds
Every upstream build, Linux only
Source
proc-1hz.ndjson: main thread nvcs, window delta

Main thread involuntary context switches (lower is better) at 1 to 250 agents, Linux x64

against
Records, 50 agents
DateTagMedianMin to maxValidvs up to 7 previousVerdict
18 Nov 2026v0.1.11-nightly.20261118.49033.26 /s3.24 to 3.693+2.1%within noise
17 Nov 2026v0.1.11-nightly.20261117.48513.19 /s2.95 to 3.203+1.9%within noise
16 Nov 2026v0.1.10-nightly.20261116.48003.25 /s3.19 to 3.383+35.9%regressed beyond noise
15 Nov 2026v0.1.10-nightly.20261115.47493.73 /s3.29 to 3.853+126.0%regressed beyond noise
14 Nov 2026v0.1.10-nightly.20261114.47093.13 /s2.74 to 3.433+95.4%regressed beyond noise
13 Nov 2026v0.1.10-nightly.20261113.46563.30 /s2.78 to 3.693+112.3%regressed beyond noise
12 Nov 2026v0.1.10-nightly.20261112.46063.25 /s3.09 to 3.553+109.3%regressed beyond noise
11 Nov 2026v0.1.10-nightly.20261111.45441.42 /s1.40 to 1.533−8.6%within noise
10 Nov 2026v0.1.9-nightly.20261110.44911.44 /s1.27 to 1.473−7.5%within noise
9 Nov 2026v0.1.9-nightly.20261109.44431.54 /s1.43 to 1.553−0.8%within noise
8 Nov 2026v0.1.8-nightly.20261108.43801.54 /s1.33 to 1.543−1.3%within noise
7 Nov 2026v0.1.8-nightly.20261107.43401.54 /s1.27 to 1.653−1.1%within noise
6 Nov 2026v0.1.8-nightly.20261106.42881.45 /s1.35 to 1.593−7.1%within noise
5 Nov 2026v0.1.8-nightly.20261105.42341.56 /s1.45 to 1.933−5.7%within noise, thin reference
4 Nov 2026v0.1.8-nightly.20261104.41921.65 /s1.58 to 1.723+8.8%within noise, thin reference
3 Nov 2026v0.1.7-nightly.20261103.41381.52 /s1.47 to 1.593no reference yet
2 Nov 2026v0.1.7-nightly.20261102.40931.65 /s1.50 to 1.773no reference yet
1 Nov 2026v0.1.7-nightly.20261101.40481.29 /s1.20 to 1.553no reference yet
31 Oct 2026v0.1.7-nightly.20261031.40041.58 /s1.35 to 1.593no reference yet
30 Oct 2026v0.1.6-nightly.20261030.39401.46 /s1.28 to 1.533no reference yet
29 Oct 2026v0.1.6-nightly.20261029.38921.49 /s1.32 to 1.673no reference yet
28 Oct 2026v0.1.4-nightly.20261028.38411.31 /s1.24 to 1.433no reference yet
27 Oct 2026v0.1.4-nightly.20261027.37961.53 /s1.47 to 1.773no reference yet
26 Oct 2026v0.1.3-nightly.20261026.37401.50 /s1.50 to 1.633no reference yet
25 Oct 2026v0.1.3-nightly.20261025.36821.20 /s1.17 to 1.393no reference yet
24 Oct 2026v0.1.3-nightly.20261024.36431.19 /s1.13 to 1.493no reference yet
23 Oct 2026v0.1.3-nightly.20261023.35921.55 /s1.33 to 1.743no reference yet
22 Oct 2026v0.1.2-nightly.20261022.35411.48 /s1.19 to 1.503no reference yet
21 Oct 2026v0.1.2-nightly.20261021.34991.35 /s1.33 to 1.473no reference yet
20 Oct 2026v0.1.1-nightly.20261020.34381.51 /s1.37 to 1.523no reference yet
19 Oct 2026v0.1.1-nightly.20261019.33731.30 /s1.15 to 1.453no noise band yet, no verdict
18 Oct 2026v0.1.1-nightly.20261018.33351.33 /s1.12 to 1.503no noise band yet, no verdict
17 Oct 2026v0.1.1-nightly.20261017.32721.25 /s1.25 to 1.293no noise band yet, no verdict
Ladder as a table
AgentsNightly 18 NovFork perf.9
10.17 /s0.054 /s
503.26 /s0.57 /s
1003.77 /s0.90 /s
1504.86 /s1.20 /s
2004.39 /s1.47 /s
2504.10 /s1.81 /s
Download this series