Commit graph

6 commits

Author SHA1 Message Date
igneum-labs
6039ae8d9e Counter ASIC 3.0 item 1: the partial-store chip and the time-memory curve in chip-model-v3.md section 5
f = 0, 0.25, 0.5, 0.75, 1 on GDDR7 (the 5090's board), one HBM3 stack and eight HBM3 stacks, scored in energy per
hash, reads in flight per watt, rate per chip and dollars per MH/s against the RTX 5090 at 136.1 MH/s and 326 W.
The curve is monotone toward f = 1; the f = 1 chip reads 5.1x (GDDR7) to 9.2x (HBM3) per joule in the model and
2.1x to 4.8x by the Ethash precedent: over 2x. The mixer and item 2 do not touch it; the levers named are the
5090's watts under a power cap (a PC 2 job, owed) and program work in the latency shadow. Inputs cited with URLs
read 6 October 2026; the mixer counted from memhard.rs at 128 hoisted / 144 unhoisted ops per application.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-06 07:32:00 +00:00
igneum-labs
9597374c4d verifier regression of e08909f fixed (derive_items out of line, one instance per cache size with the line mask a constant: v2 0.609 ms per unit against readwidth's 0.607, was 1.33); x8 into class v3 under the delegated rule (V3_CLASS = MX8; mx8-genesis and mx8-devnet-epoch0 re-exported through the seam, the devnet one with the era inside; the mx4 packs kept as the x4 record, generator 2); tests/packs.rs and tests/mixer.rs on the x8 class; mixer-x4.md 6.2a, 6.4a, 6.5 decision, 6.6 the regression; chip-model-v3.md headline x8 (0.92x with the factor); bench-log addendum with the PC 1 build rows
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-05 22:10:55 +00:00
igneum-labs
829687b1ab mixer x4: the measure session (v2 / x4 / x8 verifier 1.33 / 1.94 / 2.79 ms per unit on a loaded core, 1.45x and 2.1x; the 256 MiB fill 172 to 175 ms; the Metal 1 GiB build flat at 21 ms, latency-bound), the verification-throughput consequences (C19), what is unverified and what is owed; the bench-log entry; the PC 1 two-card playbook; the measured verifier row in the chip model
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-05 22:02:28 +00:00
igneum-labs
d2d5f5d6f2 mixer-x4.md: the x8 packs' Metal and Apple OpenCL rows, the daily-build table per tier (5090, 9070 XT, M5 Max, gfx1036 per prepare, a scaled 8 GB-class row) and the x4 / x8 rule; chip-model-v3.md: the mixer row alone as the headline (layer 5 measured, not adopted, with the 5090 and 9070 XT g beside the Mac's), x8 rows at year 0 and year 4
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-05 22:02:23 +00:00
igneum-labs
86ef4eab1e chip-model-v3.md: the hot table in the added form priced with the Mac's g (0.93 / 0.87 at 32 / 64 MiB, the 5090's g pending the PC rows): 1.98x and 2.12x at the equal integer budget, 1.60x and 1.67x with the SRAM deducted; the margin section says which convention keeps the claim
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-05 22:02:23 +00:00
igneum-labs
e08909f138 mixer x4 and the cache growth rule (Counter ASIC 2.0, class v3 construction): LoadClass mixer_mult and growth, LoadClass::MX4 (v2 loads, no width roll), memhard::Shape in MixParams, m mixer applications per round with keys round_key(r m + j), Cache::fill_log2, the option C schedule (growth_doublings, cache_log2_words, dataset_log2_words, days_since_genesis) with its test table, day-sized Epoch entries, the three emitters (m loop only for m > 1, v2 text unchanged), program.h and program.json fields, packfile.h mixerMult, packbench and OpenCL host prints, --class mx4 and --days on the CLI; docs/plans/mixer-x4.md design and spec text, docs/analysis/chip-model-v3.md, the 5090 and 9070 XT dataset-build playbooks (measurements and vectors to follow)
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
2026-10-05 20:54:10 +00:00