====================================================================
Device: NVIDIA GeForce RTX 4050 Laptop GPU (10de:28e1, AD107M, 20 SMs)
Driver: NVIDIA 615.78.08 (open kernel module / GSP-RM)
Kernel: Linux 7.2.10-1-cachyos (Clang 23.1.3 built)
Host: ASUS TUF Gaming A16 FA607NUG
====================================================================

STOCK BASELINE (Half-Rate):
--------------------------------------------------------------------
  [1] FP16 -> FP32 Accumulate :  26.48 TF/s  (Best Time: 209.02 ms)
  [2] TF32 -> FP32 Accumulate :  13.21 TF/s  (Best Time: 209.56 ms)
  Telemetry: SM Clock: 2655 MHz | Temp: 45 C

UNLOCKED (Full-Rate, 20,000 iterations, 3 reps):
--------------------------------------------------------------------
  [1] FP16 -> FP32 Accumulate :  52.97 TF/s  (Best Time: 104.51 ms)  [2.00x speedup]
  [2] TF32 -> FP32 Accumulate :  26.42 TF/s  (Best Time: 104.78 ms)  [2.00x speedup]
  Telemetry: SM Clock: 2655 MHz | Temp: 46 C
====================================================================
