Search
15:27:02
shka__
oh, i missunderstood in such case
15:31:10
beach
It seems SBCL can do the thing with 1 multiplication and 1 addition in around 3.4ns.
15:31:40
beach
That's 3 times as fast as my back-of-envelope calculation from this morning.
15:33:36
beach
Another addition is required for mixing the oscillators of course.
15:39:44
beach
So with a single core, max performance is around 20μs/5ns = 4000 simultaneous oscillators of this type.
15:47:17
beach
The technique with complex numbers is slightly faster: 3ns ???
15:47:45
beach
Nah, probably a cache problem.
15:48:05
beach
Now the first technique is less than 2ns.
15:48:29
beach
Still, both are very encouraging numbers.