I want to make sure that I'm understanding "parallel slack".

In assignment 1, when we were computing the mandelbrot image, each row is independent. If we spawned a thread for each row, then for an image of 1200 rows being processed on a machine with 4 cores, 2 threads per core with available hyper-threading, would the best case parallel slack be 1200/8?

If instead each thread processed 2 rows, would the parallel slack drop to 600/8?

@kostun I think you're correct in calculating the parallel slacks, but 1200/8 is not the "best case". The slide says ~8 is a good ratio, so we'd better have 8*8 = 64 parts of independent work. Thus we can divide 1200 rows into 64 parts, each with 1200/64 rows.