Hacker News Reader

Comment by bjourne | Hacker News Reader

I'll pass - Hacker News comments are not dissertations. The C code ran faster on my machine. YMMV.

Ran faster on your machine than what? Mojo was only runnable locally starting 2 days ago, so it doesn't sound like you compared it with Mojo.

I don't really believe you ran either the Mojo or the Julia code. There's no way your single-threaded C code outperformed multi-threaded simd optimized Julia or Mojo. It's flat out impossible.

The only other explanation is if you ran the non-simd Julia version under a single thread.

I ran this Julia code which one of the thread participants claimed to be the "best": https://discourse.julialang.org/t/julia-mojo-mandelbrot-benc... Run your own benchmarks if you don't believe me. IDGAF.

Just to make sure, did you start Julia with multithreading enabled? `julia --threads=auto` should do it (or `julia -tauto` if you prefer).

Without it, Julia starts single threaded, which means the code does all this work to enable multithreading and then doesn't get to benefit from it.

I did. Running with threads improves performance by 50%, but is still nowhere near C performance. My machine only has two cores so threading doesn't help much.

That's interesting. It makes sense that a two core machine doesn't benefit too much from multithreading, but "nowhere near C performance" is pretty surprising. I'll try out both the programs around this weekend on my own fairly anaemic machine, and see how they fair for me. Thanks for responding!

Cool. If Julia runs much faster for you than for me I'd be interested in hearing it. I was honestly surprised the performance was so bad so perhaps I did something wrong.

If you want speed, why not use, say, GLSL, instead of C? The code is far simpler, and runs much faster than C.

Reply on news.ycombinator.com