Matrix multiplication. RGBA32F textures + transposed B benchmark

WebGL2 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. Random CGPU and CGPU - CCPU are shown.

calculating
N=

GeForce GT 710     ~23.6 GFLOPS (5800 ms, N = 4096).

I re-run all programs many times (press "Run" button) and try to choose "representative" values. Results differ very much on different GPUs.


WebGL2-compute vs. WebGL?     updated 27 Apr 2019