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

Income per tier: the three tariffs are USD 0.05, 0.10 and 0.25 a kWh (cheap industrial, US retail, UK retail; the coordinator's answer, 7 October 2026), the table regenerated, the cost per mined IGN shown to three significant figures

Browse source 
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
This commit is contained in:
igneum-labs 2026-10-07 08:54:44 +00:00
parent 6ec3308291
commit 704832ead8
5 changed files with 19 additions and 19 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

24
docs/analysis/income-tiers.md
View file

@ -55,22 +55,22 @@ Three network sizes, because income is a share of the network and nobody knows t
## Electricity a day, and the electricity cost of one mined IGN
Electricity a day = measured watts x 24 h x the tariff. The cost of one IGN divides that by the day-90 figure at 10 GH/s; at another network size it scales with the network (ten times the network, ten times the cost per coin). The three tariffs are the ones the mission named; USD 0.005 a kWh sits under every retail tariff on record and is a floor, not a tariff anyone pays (approximate, from memory of published tariffs).
Electricity a day = measured watts x 24 h x the tariff. The cost of one IGN divides that by the day-90 figure at 10 GH/s; at another network size it scales with the network (ten times the network, ten times the cost per coin). The three tariffs: USD 0.05 a kWh is cheap industrial power, USD 0.10 is a US retail rate, USD 0.25 is a UK retail rate at today's level (each approximate; tariffs move, the watt figures do not).
| Tier | Card | USD/day at 0.005/kWh | USD/day at 0.02/kWh | USD/day at 0.1/kWh | USD per IGN at 0.005/kWh | USD per IGN at 0.02/kWh | USD per IGN at 0.1/kWh |
| Tier | Card | USD/day at 0.05/kWh | USD/day at 0.1/kWh | USD/day at 0.25/kWh | USD per IGN at 0.05/kWh | USD per IGN at 0.1/kWh | USD per IGN at 0.25/kWh |
|---|---|---|---|---|---|---|---|
| 8 GB | NVIDIA RTX 4060 Ti 8 GB | 0.0087 | 0.035 | 0.174 | 0.0000 | 0.0000 | 0.0000 |
| 8 GB | NVIDIA RTX 4060 Ti 8 GB | 0.087 | 0.174 | 0.436 | 0.000007 | 0.000014 | 0.000035 |
| 8 GB | NVIDIA RTX 4060 | owed | owed | owed | owed | owed | owed |
| 12 GB | NVIDIA RTX 3060 | 0.012 | 0.050 | 0.249 | 0.0000 | 0.0000 | 0.0000 |
| 12 GB | NVIDIA RTX 4070 | 0.011 | 0.044 | 0.219 | 0.0000 | 0.0000 | 0.0000 |
| 12 GB | NVIDIA RTX 5070 | 0.016 | 0.066 | 0.329 | 0.0000 | 0.0000 | 0.0000 |
| 16 GB | NVIDIA RTX 4060 Ti 16 GB | 0.0087 | 0.035 | 0.174 | 0.0000 | 0.0000 | 0.0000 |
| 16 GB | AMD RX 9070 XT | 0.024 | 0.096 | 0.478 | 0.0000 | 0.0000 | 0.0000 |
| 24 GB | NVIDIA RTX 3090 | 0.027 | 0.110 | 0.549 | 0.0000 | 0.0000 | 0.0000 |
| 24 GB | NVIDIA RTX 4090 | 0.022 | 0.088 | 0.439 | 0.0000 | 0.0000 | 0.0000 |
| 32 GB | NVIDIA RTX 5090 | 0.031 | 0.124 | 0.620 | 0.0000 | 0.0000 | 0.0000 |
| 12 GB | NVIDIA RTX 3060 | 0.124 | 0.249 | 0.622 | 0.00000802 | 0.000016 | 0.0000401 |
| 12 GB | NVIDIA RTX 4070 | 0.109 | 0.219 | 0.547 | 0.00000671 | 0.0000134 | 0.0000335 |
| 12 GB | NVIDIA RTX 5070 | 0.164 | 0.329 | 0.822 | 0.00000602 | 0.000012 | 0.0000301 |
| 16 GB | NVIDIA RTX 4060 Ti 16 GB | 0.087 | 0.174 | 0.434 | 0.00000756 | 0.0000151 | 0.0000378 |
| 16 GB | AMD RX 9070 XT | 0.239 | 0.478 | 1.19 | 0.0000197 | 0.0000394 | 0.0000984 |
| 24 GB | NVIDIA RTX 3090 | 0.275 | 0.549 | 1.37 | 0.0000111 | 0.0000223 | 0.0000557 |
| 24 GB | NVIDIA RTX 4090 | 0.220 | 0.439 | 1.10 | 0.00000645 | 0.0000129 | 0.0000322 |
| 32 GB | NVIDIA RTX 5090 | 0.310 | 0.620 | 1.55 | 0.00000482 | 0.00000964 | 0.0000241 |
| Apple | Apple M5 Max (40 GPU cores) | owed | owed | owed | owed | owed | owed |
| Rig | 6 x NVIDIA RTX 4070 (a six-card rig of the 12 GB card, the rig's own power is six times the card's plus the host (the host is not measured here)) | 0.066 | 0.262 | 1.31 | 0.0000 | 0.0000 | 0.0000 |
| Rig | 6 x NVIDIA RTX 4070 (a six-card rig of the 12 GB card, the rig's own power is six times the card's plus the host (the host is not measured here)) | 0.656 | 1.31 | 3.28 | 0.00000671 | 0.0000134 | 0.0000335 |
## What it means per tier, and what is being done

2
docs/plans/launch-pack.md
View file

@ -215,7 +215,7 @@ Evidence: docs/analysis/mission/future.md section 8; testnet-go.md LG-5.
| 5 | The proving customer's signature | One customer paying for proofs at a published rate, or a signed letter of intent with a volume (Taiko is the named first customer; the brief is ledger X13's) | LG-11, mainnet |
| 6 | The prize | Escrow USD 50,000 with the entity; confirm the registered address on the staged text; give the publish word; pick the disclosure mailbox name | LG-13 |
| 7 | Two mailboxes | `developer@igneum.network` (or the project lead's name for it) and `security@igneum.network` | items 3, 4 and 6 |
| 8 | The three tariffs | The mission's three prices are USD 0.005, 0.02 and 0.10. `docs/plans/funding.md` section 4 uses the same three numbers as prices per IGN ("inputs, not expectations"); the instruction to this lane reads them as electricity tariffs per kWh, and the table is built that way (the lowest sits under any retail tariff on record). If the per-IGN reading was meant, say so: the generator takes a second axis in an hour, and the regulatory sentence 3 (no price language) is the reason this lane did not add it unasked | LG-1's final form |
| 8 | The three tariffs | DECIDED (the coordinator for the project lead, 7 October 2026, 11:xx UK): electricity tariffs per kWh, never a coin price; the triple is USD 0.05 (cheap industrial), 0.10 (US retail) and 0.25 (UK retail at today's rate), and the table was regenerated on it. Nothing left here | none |
| 9 | Permission to post the first outside block | reinvent.md 4.1 G-C2: the "first outside block" post needs the owner's permission | LG-7's post, not its check |
## 6. Not done here, and why

2
docs/plans/testnet-go.md
View file

@ -122,7 +122,7 @@ Added by the launch-pack lane (branch `launch-pack`) from `docs/analysis/mission
| Gate | What must be true | Check | When | State at 10:1x UK, 7 October 2026 | Who |
|---|---|---|---|---|---|
| LG-1 Per-tier income at three tariffs | `docs/analysis/income-tiers.md` is published before the testnet: one 8, 12, 16, 24 or 32 GB card, a rig and a pool user, IGN a day at three network sizes from the measured bench rows and `EmissionSchedule::TESTNET_1`, electricity a day and per mined IGN at USD 0.005, 0.02 and 0.10 a kWh, each tier with what it means and what is being done (the consequences rule) | `node tools/launch/income-tiers.mjs --check` (the page equals its inputs) and `node --test tools/launch/income-tiers.test.mjs` (the schedule arithmetic: 6,912 IGN a day for 100 MH/s on 100 GH/s at the launch rate, day 1 at 10 percent, one month step at 2^(-1/24)); the page is in the public export and passes `tools/ci/identity-check.sh` | T | DONE on the branch from the 6 October rented-card rows (ten-field class). OWED: re-generate at the testnet cut from class v4 rates; the RTX 4060 and Apple wall power; an Intel row | launch-pack lane, then whoever cuts 0.4.0 |
| LG-1 Per-tier income at three tariffs | `docs/analysis/income-tiers.md` is published before the testnet: one 8, 12, 16, 24 or 32 GB card, a rig and a pool user, IGN a day at three network sizes from the measured bench rows and `EmissionSchedule::TESTNET_1`, electricity a day and per mined IGN at USD 0.05, 0.10 and 0.25 a kWh (cheap industrial, US retail, UK retail; decided 7 October 2026), each tier with what it means and what is being done (the consequences rule) | `node tools/launch/income-tiers.mjs --check` (the page equals its inputs) and `node --test tools/launch/income-tiers.test.mjs` (the schedule arithmetic: 6,912 IGN a day for 100 MH/s on 100 GH/s at the launch rate, day 1 at 10 percent, one month step at 2^(-1/24)); the page is in the public export and passes `tools/ci/identity-check.sh` | T | DONE on the branch from the 6 October rented-card rows (ten-field class). OWED: re-generate at the testnet cut from class v4 rates; the RTX 4060 and Apple wall power; an Intel row | launch-pack lane, then whoever cuts 0.4.0 |
| LG-2 Hash-origin report on Devnet 2 for seven days before the go | The daily report (keys with a block, keys above dust, attested pools and shared payouts, the project fleet's share, the ten largest keys, the network step) posts from the Devnet 2 observer's tables on seven consecutive days before step 10 | `node tools/observer/hash-origin.mjs --prefix dn2_ --dry` prints a report; seven consecutive `hash-origin:<date>` keys in the poster's state for that prefix; `node --test tools/observer/hash-origin.test.mjs` (a known-finished day and a known-failed day) | T | The job exists and ran read-only on the live devnet tables today (113 keys, 0.76 GH/s). OWED: a Devnet 2 observer writing `dn2_live_*` (fleet lane), the fleet key file from the fleet registry, the systemd timer on the box (`docs/plans/launch-pack.md` section 3) | launch-pack lane (job), fleet lane (observer, key file), build-server lane (timer) |
| LG-3 Signed and notarised installers | The 0.4.0 DMG is signed with a Developer ID certificate of Igneum Labs LTD, notarised and stapled, and opens on a fresh macOS machine with no Privacy and Security visit; the Windows installer and every exe inside it carry an EV Authenticode signature of Igneum Labs LTD with a timestamp | `spctl --assess --type execute -vv <app>` prints `accepted` and `source=Notarized Developer ID`; `osslsigncode verify <installer>` (or `signtool verify /pa /v`) prints the signer `Igneum Labs LTD` and a timestamp; the shipper's `verify` step writes both into the manifest (`signed_by`, `notarized`) and the download page shows the signer and the sha256 (`docs/plans/launch-pack.md` section 2.4) | T | NOT DONE: certificates approved 7 October; the entity's documents and the two enrolments are the owner's (section 2, items for the owner); the signing step is a specification for the shipper (`docs/plans/launch-pack.md` 2.4) | the owner (enrolments), the shipper (the step) |
| LG-4 First-share time measured per release | Every release's Devnet 2 gate measures download to first accepted share on a fresh Windows 11 VM and a fresh macOS VM, as three timed steps (download and install, sync, dataset build), and the numbers are published on /evidence; the mission's bar is under 10 minutes in 9 of 10 fresh Windows installs | the gate log carries `FIRST-SHARE <platform> download=<s> sync=<s> dataset=<s> share=<s>` for both platforms (`docs/plans/launch-pack.md` 2.5 is the specification of the script that writes it, tools/fleet/first-share-time.sh, OWED: until it exists this row is OPEN); the /evidence row equals the log line | T | OWED: specification written, script not built; no fresh-machine time has been measured end to end (`docs/analysis/mission/reinvent.md` 3.1) | fleet lane with the shipper |

2
tools/launch/income-tiers.json
View file

@ -2,7 +2,7 @@
"_about": "Inputs of docs/analysis/income-tiers.md (tools/launch/income-tiers.mjs). One row per measured card: the best measured mining rate of the Igneum miner alone on that card, with the wall power the same run read. Every row cites its measurement; a row with power null has its electricity column OWED, never estimated. No row is from memory. Tiers follow the consequences rule (CLAUDE.md, 5 October 2026): one 8 GB card, one 12 GB card, one 16 GB card, one 24 or 32 GB card, a rig, a pool user. Re-generate at every miner cut: the rates move with the program class.",
"measured_on": "6 October 2026, rented single-card Linux boxes (docs/analysis/prover-tiers-real-cards.md, bench log 'Prover tiers on real cards: the rented fleet, 6 October 2026'): the 0.3.12 miner on the ten-field override, the miner alone on the card, nvidia-smi power at the same second. The RX 9070 XT row is the telemetry run of 5 October 2026 on a Windows rig (bench log 'Power, heat, fans and clocks, measured'); the Apple row is the Metal bench of 4 October 2026 (site/miner-bench.json, class v4). The class v4 rates on the live devnet run lower than the ten-field rates on the same card (RTX 5090: 124.2 MH/s class v4 against 98.5 on the rented box under a different driver and container, docs/bench-log.md); the table is re-generated at the testnet cut from that class.",
"network_ghs": [1, 10, 100],
"kwh_prices_usd": [0.005, 0.02, 0.10],
"kwh_prices_usd": [0.05, 0.10, 0.25],
"rows": [
{ "tier": "8 GB", "card": "NVIDIA RTX 4060 Ti 8 GB", "vram_gb": 8, "mh_s": 19.07, "watts": 72.6, "vendor": "nvidia", "source": "docs/analysis/prover-tiers-real-cards.md, RTX 4060 Ti 8 GB row" },
{ "tier": "8 GB", "card": "NVIDIA RTX 4060", "vram_gb": 8, "mh_s": 17.07, "watts": null, "vendor": "nvidia", "source": "docs/analysis/prover-tiers-real-cards.md, RTX 4060 row (the sampler read 0.0 W on that host: power OWED)" },

8
tools/launch/income-tiers.mjs
View file

@ -9,8 +9,8 @@
// key signs its checkpoints (decision 2: an unsigned key's tenth of the producer share goes to the proving pool).
//
// No price of the coin appears here. The dollar columns are electricity: a cost, from a measured watt figure and a
// tariff the reader picks. The three tariffs are the three the mission named (USD 0.005, 0.02 and 0.10 a kWh); the
// lowest is under any retail tariff on record and is kept because the mission named it.
// tariff the reader picks. The three tariffs (the coordinator, 7 October 2026, 11:xx UK): USD 0.05 (cheap industrial
// power), 0.10 (US retail) and 0.25 a kWh (UK retail at today's rate), all approximate as tariffs go.
//
// node tools/launch/income-tiers.mjs writes docs/analysis/income-tiers.md
// node tools/launch/income-tiers.mjs --check exit 1 when the file on disk is not what the inputs produce
@ -74,7 +74,7 @@ export const MOMENTS = [
];
const fmt = (x, d = 0) => Number(x).toLocaleString('en-GB', { minimumFractionDigits: d, maximumFractionDigits: d });
const fmtMoney = x => (x < 0.01 ? x.toFixed(4) : x < 1 ? x.toFixed(3) : x.toFixed(2));
const fmtMoney = x => (x < 0.01 ? Number(x.toPrecision(3)).toString() : x < 1 ? x.toFixed(3) : x.toFixed(2));
export function compute(inputs) {
const refSeconds = 90 * 86_400; // the ramp is over: the figure a miner keeps
@ -129,7 +129,7 @@ export function render(inputs, computed = compute(inputs)) {
L.push('');
L.push('## Electricity a day, and the electricity cost of one mined IGN');
L.push('');
L.push(`Electricity a day = measured watts x 24 h x the tariff. The cost of one IGN divides that by the day-90 figure at ${computed.rows[0].ref_network_ghs} GH/s; at another network size it scales with the network (ten times the network, ten times the cost per coin). The three tariffs are the ones the mission named; USD 0.005 a kWh sits under every retail tariff on record and is a floor, not a tariff anyone pays (approximate, from memory of published tariffs).`);
L.push(`Electricity a day = measured watts x 24 h x the tariff. The cost of one IGN divides that by the day-90 figure at ${computed.rows[0].ref_network_ghs} GH/s; at another network size it scales with the network (ten times the network, ten times the cost per coin). The three tariffs: USD 0.05 a kWh is cheap industrial power, USD 0.10 is a US retail rate, USD 0.25 is a UK retail rate at today's level (each approximate; tariffs move, the watt figures do not).`);
L.push('');
L.push(`| Tier | Card | ${P.map(p => `USD/day at ${p}/kWh`).join(' | ')} | ${P.map(p => `USD per IGN at ${p}/kWh`).join(' | ')} |`);
L.push(`|---|---|${P.map(() => '---').join('|')}|${P.map(() => '---').join('|')}|`);