AVX2 FMA throughput with independent vector accumulators. Higher scores indicate stronger SIMD execution for media, physics, and data workloads.
Vector Math AVX21T
110+51%
AVX2 FMA throughput with independent vector accumulators. Higher scores indicate stronger SIMD execution for media, physics, and data workloads.
Vector Math AVX28T
570+50%
AVX2 FMA throughput with independent vector accumulators. Higher scores indicate stronger SIMD execution for media, physics, and data workloads.
Vector Math AVX28T
570+50%
Compiles and executes programs exercising many different CPU pipelines. Higher scores indicate stronger JIT throughput, branch recovery, and general CPU performance.
JIT VM1T
50+52%
Compiles and executes programs exercising many different CPU pipelines. Higher scores indicate stronger JIT throughput, branch recovery, and general CPU performance.
JIT VM8T
200+54%
Compiles and executes programs exercising many different CPU pipelines. Higher scores indicate stronger JIT throughput, branch recovery, and general CPU performance.
JIT VM8T
190+47%
Measures average memory access speed via random pointer chasing in a 64 MB buffer. Higher scores indicate lower access latency. A higher score is better.