3 ms·
This benchmark, lut and divide&conquer lut, seems to be really slow compared to paul's new itoa. His biggest trick is to preallocate the target buffer size, and
by rurban 9y ago
This benchmark, lut and divide&conquer lut, seems to be really slow compared to paul's new itoa.
His biggest trick is to preallocate the target buffer size, and to avoid all this heavy branching.