5mo ago

"this program generates around 31GiB of FizzBuzz per second"

codegolf.stackexchange.com

High throughput Fizz Buzz

I've spent months working on this program. I've long thought that "how fast can you make a FizzBuzz" would be a really interesting question for learning about high-performance programming, and when I subsequently saw this question posted on CGCC, I pretty much had to try.

This is so beautiful 🥹

edit: outjerked yet again https://codegolf.stackexchange.com/a/269772

TechSploits @reddthat.com

Red @reddthat.com

2y ago

High throughput Fizz Buzz

codegolf.stackexchange.com /questions/215216/high-throughput-fizz-buzz/236630

12 comments

Why did you post that implementation and not David Frank's with 1.7 Terrabit/s ?
- I saw it before that one, and by that time i already posted. I think deleting and reposting is a little weird so I guess I'll have to make due X)
  I'll mention it in the post body though, thanks lol
  
  Didn't realise that it's yours! Kudos :)
- Can we get a tldr for the approach used?
  
  hand-optimized SIMD-instructions probably
I already have a master's thesis. This was harder.
Lmao
- God I miss PCJ 😭
This isn’t even really C, it’s a bunch of assembly instructions jammed into C
Not that it’s not impressive… but still.
- Correction: This is not C. It may seem so because of the #include's and #define's but this is just standard gnu assembler syntax. You can notice that the included files are all in asm/ and the macros are just valid assembly.
  It's built with GCC instead of as because gcc links the standard library (which include those asm/ files) for you. Though you can build it normally with as, too.
  Heavy, iirc. I haven't done this stuff in a while so please correct me if i'm wrong.
  
  Oh dang ok. TIL
  I saw gcc and assumed c or c++ but it didn’t look like cop

12 comments