the project lead, 5 October 2026, 22:45 BST: "make sure we have ember tuning every single card for efficiency out of the box, the
more data = the better the tune, make an awesome system." Built on lever 3 (docs/plans/miner-eff.md), lever 2's signed
tuning section (docs/design/miner-tuning.md), the AMD telemetry helper (423936b, its --tune/--set-gmax/--set-plimit/
--reset contract) and the Power control switch (057f0ec). Design, data flow, tiers and the privacy line:
docs/plans/ember-tune.md.
- src/ember.rs (new): two knobs per card (power limit %, core clock cap MHz; memory clock never touched), the full plan
(power ladder 100..50%, then the clock ladder 90..60% at the chosen power), the confirm plan (the fleet prior and one
neighbour), the baseline plan (measure only), the marks (faulted, hot, memory_clock_dropped, unapplied, no_readings),
the choice (best MH/W within 1% of the top rate, then rate, then draw), the fleet record (a hash of the install id,
no address), the prior lookup and the kill switch (tuning.ember), the state machine on a fake clock. 9 unit tests.
- engine.rs: tick_sweep schedules every NVIDIA, AMD and Apple card (120 s steady, 600 s to the boundary, no job hold,
no pause, weekly, again after a driver major or program-class change, never under the manifest kill switch); the
probe (nvidia-smi clocks.max.gr + driver_version and the direct/helper mode; igneum-gpu-telemetry --tune for AMD);
tune_apply (nvidia-smi -pl / -lgc 0,<MHz> / -rgc directly or through the helper; the AMD helper per request);
Cmd::TuneProbe, Cmd::TuneSet; faults from rejected and mismatched hashes mark the step; the TUNE lines and the TUNE
{json} record, uploaded with the log; the Tuned line on the card state. The NVIDIA helper starts only with Power
control on: the --sweep job never counts as permission (no prompt on a PC with nobody there).
- sweep.rs: the helper protocol gains lgc/rgc (clock cap and reset) and resets the clocks after 20 idle minutes.
- state.rs, config.rs: the tune fields (clock cap, driver, class, source, the Tuned line); the nvidia-smi telemetry
query carries clocks.gr and clocks.mem; the AMD sample line's plimit_pct and gmax_mhz are parsed.
- ui: "Tuned: X MH/s at Y W (Z MH/W)" with the point, the source and when; measure-only cards say why; the Ember Tune
switch; tune-line.test.mjs.
- relay/lib/ember.mjs + relay/test/ember.test.mjs: the aggregation per (card model | driver major | program class):
median point, MH/W, spread, samples, machines; five samples converge, an outlier does not move the median, baselines
make no prior, de-duplication, the manifest merge keeps lever 2's cards. api/console.mjs fn=tuning and
tools/console.mjs tuning; tools/tuning.mjs --priors [--write tuning.json] [--site] [--tuning-off].
- site: the fleet priors table on /miners (site/miner-priors.json), the lever text.
- relay/playbooks/ember-tune-pc1.ps1: the PC 1 run (second engine with --sweep from a scratch copy of the install).
Measured tonight: see the bench log entry that follows the PC 1 run. The 9070 XT left PC 1's bus at 20:40 UTC and the
5090 needs the administrator prompt the project lead cannot answer asleep, so tonight's PC 1 run is the baseline plan on the 5090
through the whole pipeline; the two-knob tune on both cards is owed.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Relay (X23, X27): the intake key is its own tier (upload and file drops only, RELAY_INTAKE_COMPAT=0 closes it);
a run task needs an Ed25519 signature by the Mac run key over {to, nonce, body sha256, flags} (RELAY_RUN_PUB,
401 without) and an HMAC tag with the target's machine secret that the agent verifies before anything runs;
results and registration are bound to the machine the secret proves (403 on a forged from).
X24: every client and Mac tool sends x-relay-token as a header to /api/relay?fn=; the path token stays for the
phone page only. X25: the agent arms the logon task only for a restart a task asked for and disarms on start
and exit. X26: 30-day retention with blob deletion, feed capped at 100, the dl base as RELAY_DL_BASE held by the
agent, never in a body. X28: GET inbox never acks (POST inbox does), RELAY-REBOOT on its own line and only with a
reboot flag, 120/min and 10 failed auths/min per IP, no username or folder on register, WSL sudo scoped to
apt-get and dpkg with SETENV, no password on a command line. X29: the intake key reaches curl through -K in
upload.sh and both upload-log.bat; tools/ci/curl-header-check.sh fails the class. G14: TZ=UTC in ship-app.mjs
and publish-jobs.sh; tools/ci/commit-tz-check.sh fails the class; history-rewrite.md names the .old-2026-10-05
files as the values in the history. The handler moved to relay/lib/handler.mjs with injected sql and blobs
(relay/lib/blob.mjs holds @vercel/blob) so relay/test/handler.test.mjs drives it without a database:
47 tests across 6 suites, all green.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
5 October 2026: an RX 9070 XT went into PC 1 through a Sonnet USB4 box while the app ran and nothing noticed; the
app detected cards once at start. Now src/hotplug.rs compares every enumeration with the list (key, else vendor +
name when unique; a card whose tool did not answer is never called removed): a new usable card starts a worker,
enabled by default like a card at start, with "New card: <name>, mining" on the strip and in the log; a card
Windows lists with a problem code (Win32_VideoController Status / ConfigManagerErrorCode) is shown as "<name>: not
usable (Code 43)" with the reboot-or-reinstall hint and no worker; a card that disappears has its worker stopped
(quit, 8 s) and its row says removed for five minutes, then hides; an unchanged list touches nothing. The Windows
host sends "detect" on WM_DEVICECHANGE; the engine polls every 60 s (300 s on macOS, no GPU hot-plug there).
detect.rs: the Ryzen iGPU is "gfx1036" to the OpenCL worker, so the APU gfx codes count as integrated, plus the
adapter row's Intel processor string and a dedicated memory under 1 GB; integrated defaults to off with "integrated
GPU, off by default (2 to 3 MH/s for 30 W)" on the row, and the user's choice is kept across re-detections and
restarts (settings, found by key or by vendor + name when the index moved).
Console: the engine logs "cards: <name> [<kind>, <state>] | ..." at start, on every change and every 10 minutes;
relay/lib/parse.mjs reads it and the hot-plug events, the machines API and tools/console.mjs machines show them.
Tests: hotplug.rs (added, removed, moved, errored, recovered, revived, unchanged, twins, user override kept,
the console line), detect.rs (PC 1's adapter lines, the Mac, kind classification, the unusable row),
notices.test.mjs (card notices), relay parse.test.mjs (cards line). cargo test -p igneum-app: 91 passed.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
relay/api/console.mjs reads the log intake (miner_logs) and console_items in Neon, the OTA manifest, the
CI json and the jobs file from the downloads host (DL_TOKEN in the project env, never in the client), and
igneum.network/api/live; 10 s cache per answer. tools/console.mjs: post --kind log|build|note, log, machines,
chain, jobs, builds, results, sync-bench, sync-dl, sync-hetzner, sync, url. The two Mac-side build scripts
post build events. Screenshots at 375 px and desktop in docs/design/console/.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
