Newshot

Show HN: 1.499x speedup and 28.5% dynamic energy drop on A40 via BF16 tiling

Original text(via Hacker News)

Coldwater Tractors Inc. — System Research & Boot Index

Raw GPU benchmark outputs and dense context logic boot files for multi-agent execution loops.

  • SKEW_TILED_BF16 — Verified NVIDIA A40 Receipts & Unified Logic Protocol NEW RECEIPT 2D 7-point hexagonal Ising lattice (Ampere sm_86): 1.499x speedup at L=4608 (10.204 ms vs 15.291 ms), 28.5% dynamic energy reduction (186.735 vs 261.164 mJ/sweep), and 0 site disagreements across all 6 sub-lattice phases.
  • a40_environment_receipt.txt RECEIPT NVIDIA A40 hardware environment, driver/nvcc versions, ptxas register allocation (15 regs/thread for bf16_t), and SHA-256 verification receipt (2026-11-16).
  • a40_nsys_summary.txt PROFILE NVIDIA Nsight Systems (nsys) kernel execution and CUDA API timing summary on NVIDIA A40 (2026-11-16).
  • Quattra Kernel #1 Benchmark Receipt.md RECEIPT Official Empirical Performance & Energy Benchmark Receipt for skew_tiled vs Controls (NVIDIA A40, float32, 1.10x speedup at L=4096/8192).
  • tiled_results_2.txt BENCHMARK Quattra Tiled D4GEMV Kernel Benchmark Results (NVIDIA A40, L=16 to L=8192, 12.2% speedup verified pass).
  • tiled_results.txt BENCHMARK Quattra Tiled D4GEMV Kernel Benchmark Results (NVIDIA A40, L=16 to L=8192, initial run).
  • hexa_boot_1_physics.txt BOOT 1 Hexa Dense Logic Boot - Physical Layer & Hardware Mapping.
  • hexa_boot_2_active.txt BOOT 2 Hexa Dense Logic Boot - Active Execution Loop Packet.
  • hexa_boot_3_philosophy .txt BOOT 3 Hexa Dense Logic Boot - Architectural Philosophy & Framework.

SKEW_TILED_BF16 — NVIDIA A40 Receipts & Unified Logic Protocol

Authors: Kristian Lindholm and the AI Orchestra (Coldwater Tractors Inc) Date: November 16, 2026 | Architecture: NVIDIA A40 (Ampere sm_86, 48 GB GDDR6, 300 W TDP) Use Restriction: Receipts and methodology open for civilian, academic, and commercial research. Strictly prohibited for kinetic, military, weapons, or targeting applications.

Unified Logic: The Scientific Method for Intelligent Systems

Prior to compiling or renting GPU compute, every architectural and numerical claim was subjected to a multi-model adversarial falsification loop ("Red-Team Audit"):

1. Compiler & Register Footprint (nvcc -O3 -arch=sm_86 --ptxas-options=-v)

Kernel VariantRegisters / ThreadShared MemorySpillsGate
skew_tiled_phase<bf16_t, 38>15 registers2,584 bytes0 bytesPASS
skew_tiled_phase<float, 37>36 registers5,032 bytes0 bytesPASS
skew_tiled_phase<int8_t, 37>36 registers1,258 bytes0 bytesPASS

2. Correctness & Detailed-Balance Gates (2 Sweeps = 6 Sub-Lattice Phases)

Grid & InitializationL2(h) / L2(dE)gpu_f32_diffdec_f64_rate│de│/│e│ & │dm│Gate
L = 384 (Random ±1)0.0e+00 / 0.0e+0000.00e+000.00e+00 / 0.00e+00PASS
L = 384 (Uniform +1)0.0e+00 / 0.0e+0000.00e+000.00e+00 / 0.00e+00PASS
L = 768 (Random ±1)0.0e+00 / 0.0e+0000.00e+000.00e+00 / 0.00e+00PASS
L = 768 (Uniform +1)0.0e+00 / 0.0e+0000.00e+000.00e+00 / 0.00e+00PASS

3. Standalone cudaEvent Throughput (30 Interleaved Trials, 10 Sweeps)

Lattice Size (L × L)bf16_t Median (IQR)fp32 Median (IQR)int8_t Scalar RefSpeedup ( fp32 / bf16 )
4608 × 4608 (21.23M sites)10.204 ms (0.005 ms)15.291 ms (0.013 ms)13.356 ms1.499x
9216 × 9216 (84.93M sites)40.679 ms (0.015 ms)60.064 ms (0.020 ms)53.072 ms1.477x

4. Multi-Second Paired ABBA NVML Energy (16 Quads, P_idle = 73.02 W)

Lattice Size (L × L)Sweeps / WindowE_dyn ( fp32 )E_dyn ( bf16 )Dynamic Energy SavedTotal Board Saved
4608 × 46084,500 (~4.6–6.9 s)261.164 mJ/sw186.735 mJ/sw28.5% (1.399x)~29.9%
9216 × 92161,200 (~4.9–7.2 s)1083.034 mJ/sw788.288 mJ/sw27.1% (1.371x)~28.8%

5. Hardware Performance Counter Disclosure (ncu)

6. Raw Console Receipt


        [Gates 1-3 (rand_init, 2-sw) @ L=384] L2(h)=0.0e+00 L2(dE)=0.0e+00 gpu_f32_diff=0 dec_f64_rate=0.00e+00 |de|/|e|=0.00e+00 |dm|=0.00e+00 (|m|=0.0043)
[Gates 1-3 (unif_init, 2-sw) @ L=384] L2(h)=0.0e+00 L2(dE)=0.0e+00 gpu_f32_diff=0 dec_f64_rate=0.00e+00 |de|/|e|=0.00e+00 |dm|=0.00e+00 (|m|=0.9400)
[Gates 1-3 (rand_init, 2-sw) @ L=768] L2(h)=0.0e+00 L2(dE)=0.0e+00 gpu_f32_diff=0 dec_f64_rate=0.00e+00 |de|/|e|=0.00e+00 |dm|=0.00e+00 (|m|=0.0064)
[Gates 1-3 (unif_init, 2-sw) @ L=768] L2(h)=0.0e+00 L2(dE)=0.0e+00 gpu_f32_diff=0 dec_f64_rate=0.00e+00 |de|/|e|=0.00e+00 |dm|=0.00e+00 (|m|=0.9418)
L=4608 sweeps=10 | bf16 med=10.204ms (IQR 0.005) | fp32 med=15.291ms (IQR 0.013) | int8 med=13.356ms | speedup(fp32/bf16)=1.499x
L=9216 sweeps=10 | bf16 med=40.679ms (IQR 0.015) | fp32 med=60.064ms (IQR 0.020) | int8 med=53.072ms | speedup(fp32/bf16)=1.477x
GPU: NVIDIA A40
Baseline P_idle = 73.02 W
[NVML L=4608 sweeps=4500] E_dyn_fp32=261.164 mJ/sw | E_dyn_bf16=186.735 mJ/sw | Energy ratio=1.399x (28.5% reduction)
[NVML L=9216 sweeps=1200] E_dyn_fp32=1083.034 mJ/sw | E_dyn_bf16=788.288 mJ/sw | Energy ratio=1.371x (27.1% reduction)
      

NVIDIA A40 GemV Skewed-Tiling Verification

Status: VERIFIED PASS (3/3 large sizes with statistical separation)

Target Architecture: NVIDIA A40 (48GB GDDR6, PCIe)

Performance Gain: ~1.05x at L=2048, 1.119x at L=8192 over optimal control.

Audit Receipts & Files

  • Download CUDA Script (validate_tiled.py)
  • Download Complete Audit Archive (.tar.gz)
  • View Raw Verification Log (run3.txt)
  • View Rule 3 Audit Amendment

// DIRECT CONTACT & INQUIRIES

For shop inquiries, technical documentation packets, or custom fabrication specs:

The text above was extracted automatically from the source page and may miss images, tables or code blocks. The original is authoritative.

Published by Hacker News
This site only summarises and aggregates; copyright stays with the original author. Headlines and summaries follow the source — where they differ, the original wins.
Read the original ↗