Two fragment shaders following the same Mandelbrot zoom path. The left canvas uses float32 arithmetic, while the right canvas uses the fp64arithmetic shader module from @luma.gl/shadertools.

On Apple WebGPU/Metal, fp64arithmetic automatically selects an integer-controlled double-single implementation. Other GPUs use the classic floating-point error transforms by default.

The opt-in WebGPU benchmark below the canvases compares automatic selection, both explicit double-single paths, and native float32 for add, multiply, divide, and square-root compute workloads. It reports timing and numerical error together; the results are diagnostic and do not impose performance thresholds.