Skip to content

v0.18.5

Choose a tag to compare

@dkorunic dkorunic released this 15 Nov 16:57
· 16 commits to master since this release

Changelog

  • 4d113ca chore(gitignore): track .claude, ignore only settings.local.json
  • 5b5a210 docs(claude): note M3 Pro/Max relabel and derived-offset keys
  • bac4ecb docs(claude): note M5 Air probe findings and base M5 coverage
  • c414a0c docs(claude): note TG Pro core-mapping findings and open M5 leads
  • 1935ad2 docs(claude): note base M3 layout, per-SKU row splits and m3-air-run gaps
  • 548b0f8 docs(guess): state that guess places tiers, not individual cores
  • c91db52 docs(sensors): record why M2 offset classes are not P vs E
  • 9b3908f feat(guess): read the SMC ft sensor index to drop non-core groups
  • 3086ee2 feat(guess): sample the M5 sd* die-temperature estimates
  • 1a4b4f1 feat(sensors): import M3 Pro dump with ft index, relabel M3 Pro P sensors
  • b5b1e0b feat(sensors): import M4 Pro 14-core dump (Mac16,8, iSMC#28)
  • deb1437 feat(sensors): import MacBook Air 13 M5 dump and guess run (Mac17,3)
  • fde9bf2 feat(sensors): import base M3 dump (Mac15,13), split M3 rows per SKU
  • e464067 fix(guess): group M1-M4 triplets by constant A/B probe offset
  • b18d8a3 fix(guess): list cluster-maximum groups as cluster sensors, not cores
  • 90302fe fix(guess): list core groups that copy another group's key as cluster sensors
  • 5be5a87 fix(guess): run the E phase first, assign Te* groups by name as fallback
  • 53861b8 fix(sensors): M3 Pro E groups, M3 Max P groups, base M3 E probes
  • 88cfedf test(guess): add MacBook Air 15 M3 (Mac15,13) saved run as replay fixture
  • f046d5a test(guess): replace M4 Pro replay run with v0.18.4 capture