Indeed, but these days I'm beginning to prefer:

Finding the product of two NxN matrices representing element-to-element broadcast mapping between two vector registers.

VR[vd] = VR[vs] * VR[vt] for the `rs & 0xF` bitmask subset of an SIMD machine command, == zero.

Supposedly there is a way to use a direct multiplication formula to solve for RSP opcodes like VMULz, VMACz etc. but I haven't looked into it yet.

.........

^^ that's just a spambot btw you damn fool