Matrix multiplication. RGBA16F textures + transposed B benchmark

WebGL2 HALF FLOAT matrix multiplication C = A x B (SGEMM) benchmark. Fixed random A, B 4096x4096 arrays are used for calculations, therefore You can set new N = M = K value (up to N=4096, multiple of 4). FLOPS = 2 N3/ time.

calculating
N= it=

GeForce RTX 2070, N = 4096     1600 GFLOPS (RGBA16F)     835 GFLOPS (RGBA32F).

I re-run all programs many times (press "Run" button) and try to choose "highest representative" values.


GEMM tests on RTX 2070     updated 17 July 2019