I’ve finished my Bifurcations program (including the spit-polish, minus one feature). It’s all very nice, I think.
I’ve also finished my Mandelbrot program (again), this time it it’s multithreaded! I really want to bench it against my last iteration. It’s about 10x faster than the old one – instant calculation, unless you zoom to a part with a lot of black (and therefore a lot of iterations).
The default way to run it uses two threads, but I’ve compiled it with 35 threads and it’ll still run (without as much of a speed boost, of course.) At a certain point extra threads add nothing – in general anything over the number of processor cores, although I haven’t tested explicity (which is why I want to bench everything.)
Mr. Andersen is using my work as I finish it for his chaos class, I’m rather proud of that =). It was, after all, what its itended use was. He showed them the bifurcation program today, I heard, and he’s shown my preview video to the class as well. It’s the video I made (to E-30-something) so that the people overseeing all distinction projects know that I’m doing something. It’s purely a preview, my final (for my presentation) will hopefully go to around E-100.
