igneum-network/igneum
1
0
Fork 0
Code Issues Pull requests Releases Activity

Ember run 6 closed: the 4070 table and the 9070 XT abort in the plan and the bench log, the 4070 row on the fleet page, the playbook prefers ember-kit-7

Browse source 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
igneum-labs 2026-10-06 16:44:36 +00:00
parent 57de3112b8
commit 887bff36aa
5 changed files with 53 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

8
docs/bench-log.md
View file

@ -2254,3 +2254,11 @@ The clock ladder: 2781 MHz 298.8 W 0.428; 2472 MHz 262.0 W 0.488; 2163 MHz 239.9
(the floor, not the optimum: the next cut's ladder goes to 45%). RTX 4070: caps 100 to 60% all 106.0 W 28.71 MH/s
(0.271); 50% 99.1 W 28.70 (0.290); clock 2794 MHz at 50% 99.1 W (0.290); 2484 MHz 81.1 W 28.73 (0.354); further rows
and the 9070 XT ladder below once the run closes.
Run 6 closed 16:39:37Z, exit 0, 2317 s, 19 rows. RTX 4070 chosen 1863 MHz at 50% = 28.78 MH/s at 75.6 W (0.381) against
28.72 at 106.0 W (0.271): 30 W saved for no rate lost; its clock ladder at 50%: 2794 MHz 99.1 W 0.290; 2484 MHz 81.1 W
0.354; 2173 MHz 77.7 W 0.370; 1863 MHz 75.6 W 0.381 (the floor). RX 9070 XT: aborted at step 1, "card reports 0 W,
acknowledged true" = the applied rule demanded watts from a card that reports offsets (fixed bd7fcf4); the draw itself
was read on every tick (amd_watts_source=engine_telemetry, 363 samples). The helper registered on the scratch copy
(fixed 200362a: task_exe, the reregister verb). PC 1 mined through the installed app again by 16:44Z: 170.6 MH/s over
the three cards, 0 faults.

28
docs/plans/ember-tune.md
View file

@ -256,9 +256,31 @@ The per-user number: 84 W saved for 0.15% of rate, 37% more hashes per watt. The
tolerance as the guard; a chosen point on the floor carries `floor=1 note=floor, not optimum` in the chosen line and
the record. Ember 2's climb starts from the chosen point.
RTX 4070 (same driver and class): caps 100 to 60% all 106 W at 28.71 MH/s (0.271); 50% (100 W) 99.1 W 28.70 (0.290);
clock 2794 at 50% 99.1 W (0.290); 2484 MHz 81.1 W 28.73 MH/s (0.354); the rest of the ladder and the 9070 XT rows in
the bench log entry for the run.
RTX 4070 (same driver and class), 10 rows, every one ok, 55 C at most:
| Step | Clock cap | Power | Draw | Rate | MH/W |
|---|---|---|---|---|---|
| 0 (before) | unlocked | 100% 200 W | 106.0 W | 28.72 | 0.271 |
| 1 to 4 | unlocked | 90, 80, 70, 60% | 106.0 W | 28.71 | 0.271 (the cap never binds) |
| 5 | unlocked | 50% 100 W | 99.1 W | 28.70 | 0.290 |
| 6 | 2794 MHz | 50% | 99.1 W | 28.70 | 0.290 |
| 7 | 2484 MHz | 50% | 81.1 W | 28.73 | 0.354 |
| 8 | 2173 MHz | 50% | 77.7 W | 28.74 | 0.370 |
| 9 (chosen) | 1863 MHz | 50% | 75.6 W | 28.78 | 0.381 |
Per user: 30 W saved for no rate lost (+0.2%), 41% more hashes per watt; the floor again.
RX 9070 XT: no row. "TUNE aborted: the setting (0 MHz, 100% = 100 W) did not take within 30 s (card reports 0 W,
acknowledged true)", twice. The cause was `Run::applied`: a power step demanded the limit read back in watts, and
ADLX reports an offset, never watts, so no AMD power step could ever confirm. Fixed (bd7fcf4): a card with no limit
readback is applied on the acknowledgement. The watts themselves were read on every tick (`amd_watts_source=
engine_telemetry`, 363 nonzero engine samples, 456 nonzero app-state samples). The ladder reruns with kit-7. The
"100 W" in that line is the percent scale printed as watts (owed: a unit word for AMD).
After the run (nvidia-smi, the engine gone, the installed miners not yet back): 5090 1845 MHz locked at 575 W;
4070 at the 100 W limit. The installed app mined again within a minute: 5090 122.9 MH/s, 4070 28.8, 9070 XT 18.9,
170.6 MH/s, 0 faults (console, 16:44Z). The 9070 XT's ADLX state read "gmax 0 plimit 0 factory 0" (before: factory
1): zero offsets, the factory flag cleared by the engine's set of 0; the 3.0 coordinator's reset job puts it back.
## 8a. Next-cut notes (for the 0.3.12 shipper)

4
relay/playbooks/ember-tune-pc1.ps1
View file

@ -50,8 +50,10 @@ $ember = $null
# a kit is used when the installed app predates Ember Tune (0.3.13), or when a run carries fixes the installed app
# lacks (ember-kit-6: the cmd quote fix and the self-registering elevated engine of 6 October 2026); a kit engine
# registers the Power Helper on the INSTALLED exe's path (powertask::task_exe), never its own
$k7 = Join-Path $appDir 'jobs\ember-kit-7\igneum-app-ember.exe'
$k6 = Join-Path $appDir 'jobs\ember-kit-6\igneum-app-ember.exe'
if (Test-Path $k6) { $ember = $k6 }
if (Test-Path $k7) { $ember = $k7 }
elseif (Test-Path $k6) { $ember = $k6 }
elseif (-not $installedHasEmber) { foreach ($cand in @((Join-Path $appDir 'jobs\ember-kit-2\igneum-app-ember.exe'), (Join-Path $appDir 'jobs\ember-kit-1\igneum-app-ember.exe'))) { if (Test-Path $cand) { $ember = $cand; break } } }
if ($ember) {
Copy-Item -LiteralPath $ember -Destination (Join-Path $bin 'igneum-app.exe') -Force

16
site/miner-priors.json
View file

@ -20,6 +20,22 @@
"untuned_mh_per_w": 0.411,
"sentence": "84 W saved for 0.15% of rate: 127.71 MH/s at 226.8 W (0.563 MH/W) against 127.9 MH/s at 311 W untuned (0.411 MH/W), 37% more hashes per watt.",
"note": "The chosen clock is the ladder's floor (60% of the maximum), not the optimum; the next cut's ladder goes to 45% and the climb walks on from here."
},
{
"card": "NVIDIA GeForce RTX 4070",
"driver_major": "617",
"class": "l128w0",
"date": "2026-10-06",
"job": "ember-tune-pc1-6",
"point": "1863 MHz core cap, power limit 50% (100 W)",
"mh_s": 28.78,
"watts": 75.6,
"mh_per_w": 0.381,
"untuned_mh_s": 28.72,
"untuned_watts": 106.0,
"untuned_mh_per_w": 0.271,
"sentence": "30 W saved for no rate lost: 28.78 MH/s at 75.6 W (0.381 MH/W) against 28.72 MH/s at 106 W untuned (0.271 MH/W), 41% more hashes per watt.",
"note": "The chosen clock is the ladder's floor (60% of the maximum), not the optimum; the next cut's ladder goes to 45%."
}
]
}

2
site/miners.html
View file

@ -190,7 +190,7 @@ th{font-family:var(--f-mono);font-size:12px;letter-spacing:.12em;text-transform:
<p>Ember Tune runs on every card the app mines with: the power limit and the core clock are stepped on the live program and the card keeps the point with the best MH per watt within 1% of its top rate. Every finished tune is reported back without anything that identifies the owner, and the fleet's median point per card model, driver major and program class comes back down inside the signed update manifest as the starting point for the next card of that model. A model needs 5 reports before its prior is used.</p>
<p>No tune reports yet. The first rows appear once five machines with the same card model have reported.</p>
<h3 id="team">Measured by the team</h3>
<ul><li><b>NVIDIA GeForce RTX 5090</b> (driver 617, class l128w0, 2026-10-06, job <code>ember-tune-pc1-6</code>): 1854 MHz core cap, power limit 100% (575 W). 84 W saved for 0.15% of rate: 127.71 MH/s at 226.8 W (0.563 MH/W) against 127.9 MH/s at 311 W untuned (0.411 MH/W), 37% more hashes per watt. The chosen clock is the ladder's floor (60% of the maximum), not the optimum; the next cut's ladder goes to 45% and the climb walks on from here.</li></ul>
<ul><li><b>NVIDIA GeForce RTX 5090</b> (driver 617, class l128w0, 2026-10-06, job <code>ember-tune-pc1-6</code>): 1854 MHz core cap, power limit 100% (575 W). 84 W saved for 0.15% of rate: 127.71 MH/s at 226.8 W (0.563 MH/W) against 127.9 MH/s at 311 W untuned (0.411 MH/W), 37% more hashes per watt. The chosen clock is the ladder's floor (60% of the maximum), not the optimum; the next cut's ladder goes to 45% and the climb walks on from here.</li><li><b>NVIDIA GeForce RTX 4070</b> (driver 617, class l128w0, 2026-10-06, job <code>ember-tune-pc1-6</code>): 1863 MHz core cap, power limit 50% (100 W). 30 W saved for no rate lost: 28.78 MH/s at 75.6 W (0.381 MH/W) against 28.72 MH/s at 106 W untuned (0.271 MH/W), 41% more hashes per watt. The chosen clock is the ladder's floor (60% of the maximum), not the optimum; the next cut's ladder goes to 45%.</li></ul>
<p>Rows: 0. Source file: <code>site/miner-priors.json</code> in the repository, written from the fleet records by <code>tools/tuning.mjs --priors --site</code>.</p></article>
</div>
<p class="gen">Generated from the repository at build time. Times are UTC. Machine names are model names.</p>