Where each lane stands
The most recent run of each lane, what it tested, and, when the verdict is a divergence, which scenarios disagreed and what they reported.
Builds
Every Stratum build the suite has tested: pinned stable releases from the parity lane, and pre-releases from the daily indev scout (non-gating, kept out of the trend charts). Verdict is the build's most recent run; the CI link opens it.
| Build | Channel | Latest verdict | Runs | Scenarios v · s | Last tested | CI |
|---|
Run verdicts
One tile per recorded run on the pinned lane: ✓ parity holds, ✗ outcome divergence. Click opens the CI run.
Total suite duration
Wall time of the full suite per flavor, across runs. Seconds.
Per-scenario duration
Pick a scenario to follow its duration on both flavors across runs. Seconds.
Tick cost
Server work time per tick under a fixed straw-dummy load, ms/tick. Informational: vanilla and Stratum run on different CI machines, so read each trend on its own, not the gap. Near band measures the core tick loop (comparable across flavors); far band shows Stratum's distance throttle (reads near zero by design).
Latest run
| Scenario | Vanilla | s | Stratum | s | Δ s |
|---|
Δ duration is a performance signal only for scenarios whose flow is identical on both flavors. Probe scenarios proving an absence on Stratum use fixed observation windows and expectedly run longer there.