grenade

grenade commented on issue blackbeard/miner#27

2026-09-14 16:44:11 +00:00

quanpool-miner 6.2.4 measured on beast: 1234 MH/s per 5090, what it does differently, what was tried

Round 2 (2026-09-14): the loop kernel's internal round, itemised

With mining_loop in (#29), every candidate is now screened by the static size of its internal-round loop body on sm_120…

grenade pushed to perf/int-round-chain at blackbeard/miner

2026-09-14 16:44:04 +00:00

4572a5e567 engine-cuda: attribute the loop kernel's internal-round excess; drop two no-op knobs

338b4a7d32 engine-cuda: LAIR_MULADD screening variants (0 stays; 1-4 measured worse)

Compare 2 commits »

grenade created branch perf/int-round-chain in blackbeard/miner

2026-09-14 16:44:03 +00:00

grenade released quantus-miner v4.0.2-lair.2 at blackbeard/miner

2026-09-14 16:41:19 +00:00

grenade pushed tag v4.0.2-lair.2 to blackbeard/miner

2026-09-14 16:39:37 +00:00

grenade deleted branch docs/readme-lair2 from blackbeard/miner

2026-09-14 16:39:28 +00:00

grenade pushed to main at blackbeard/miner

2026-09-14 16:39:26 +00:00

4a6ed42d6b Merge pull request 'readme: 4090 at 727 MH/s on the loop kernel, how to select it' (#30) from docs/readme-lair2 into main

bbbaf815ac readme: 4090 at 727 MH/s on the loop kernel, how to select it

Compare 2 commits »

grenade merged pull request blackbeard/miner#30

2026-09-14 16:39:25 +00:00

readme: 4090 at 727 MH/s on the loop kernel, how to select it

grenade created pull request blackbeard/miner#30

2026-09-14 16:30:56 +00:00

readme: 4090 at 727 MH/s on the loop kernel, how to select it

grenade created branch docs/readme-lair2 in blackbeard/miner

2026-09-14 16:30:48 +00:00

grenade pushed to docs/readme-lair2 at blackbeard/miner

2026-09-14 16:30:48 +00:00

bbbaf815ac readme: 4090 at 727 MH/s on the loop kernel, how to select it

grenade pushed to main at blackbeard/observer

2026-09-14 14:21:49 +00:00

efaf48711c feat(ui): a caution mark where the tagline's bullet was

grenade pushed to main at blackbeard/observer

2026-09-14 14:02:15 +00:00

b04a45f54a feat(ui): a face on the about page

grenade pushed to main at blackbeard/observer

2026-09-14 13:48:27 +00:00

6ed9a6c35a feat: an about route, and somewhere to say thanks

grenade pushed to main at blackbeard/miner

2026-09-14 13:44:59 +00:00

ad77d99dfa Merge pull request 'engine-cuda: loop-structured mining_loop kernel, opt-in via MINER_CUDA_KERNEL=loop' (#29) from perf/loop-kernel into main

12bd36aee2 deploy: per-host CUDA kernel choice; benjy runs the loop kernel

816b1efb61 engine-cuda: loop-structured mining_loop kernel behind MINER_CUDA_KERNEL

Compare 3 commits »

grenade deleted branch perf/loop-kernel from blackbeard/miner

2026-09-14 13:44:58 +00:00

grenade merged pull request blackbeard/miner#29

2026-09-14 13:44:57 +00:00

engine-cuda: loop-structured mining_loop kernel, opt-in via MINER_CUDA_KERNEL=loop

grenade commented on pull request blackbeard/miner#29

2026-09-14 13:36:18 +00:00

engine-cuda: loop-structured mining_loop kernel, opt-in via MINER_CUDA_KERNEL=loop

Second commit (12bd36a) adds the per-host switch to the deploy matrix: cuda_kernel is rendered into miner.env as MINER_CUDA_KERNEL, kernel stays the id validate asserts. benjy runs loop /…

grenade pushed to perf/loop-kernel at blackbeard/miner

2026-09-14 13:36:09 +00:00

12bd36aee2 deploy: per-host CUDA kernel choice; benjy runs the loop kernel

grenade commented on pull request blackbeard/miner#29

2026-09-14 13:34:30 +00:00

engine-cuda: loop-structured mining_loop kernel, opt-in via MINER_CUDA_KERNEL=loop

benjy and quadbrat, same binary, three interleaved rounds, parity 200/200 on the loop kernel on each: