Cyclone the Next Level CPU Solver

Hey folks! Really pumped to show you something I’ve been brewing over the last few months Cyclone, touted as the fastest CPU puzzle solver out there! It’s meant for tackling Satoshi puzzles using a CPU. Why? Gotta chase those rewards! Here’s how Cyclone compares to the competition:

Nice one. But gotta say, your tool is kinda slow since it leans on that JLP-based secp256k1 math. I know a few users here who are already smashing your speeds. Private clients from some mining pools are also likely quicker than this tool. So, just be cautious with those ‘world’s fastest’ claims, ok?

Appreciate the feedback! Just finished an update for Cyclone. The AVX512 version is out now, and it’s about 15-17% faster than the AVX2 version on a Ryzen 9. Next, I’m looking at adapting it for CUDA since I’ve got some ideas to boost hash160 performance against the keyhunt CUDA version.

Great to see those CPU improvements!

Hey! Nothing really beats this: https://github.com/JayDDee/cpuminer-opt/tree/master/algo/ripemd (ripemd) https://github.com/JayDDee/cpuminer-opt/tree/master/algo/sha (sha) It supports 4-way, 8-way, avx2/avx512.... Been using CPU miner parts in Vanity Search for a year. Don’t try to let AI handle this or your server will crash.

Puzzle: 67, 730fc235 Possibilities: 13 Seed: 2113081982 KHex: 730fc235... Searching range: 730fc235000000000:7

67 found super quickly, way faster than the 66 before.

If you wanna see notable speed ups, implementing the whole SECP256K1 curve from scratch using AVX2 would be essential. Could mean 2x to 5x faster. Plus, assembly could help even more. Just optimizing hashing alone won't cut it.

I’m getting around ~190 Mkey/s directly searching Hash160 targets in pure random mode. https://github.com/NoMachine1/Cyclone

Hey everyone! Nomachine, you’re doing an awesome job! What CPU did you run to get those speeds?

AMD Ryzen Threadripper PRO 7985WX. But it’s not just puzzles, I'm doing a lot more with it.

Yo, can you drop all files like *.cpp, *.h, etc? Would love to have them for testing...

The 7985WX is a bit faster than the 7945HX (around 145% faster I think). Just about done with a distributed Cyclone version. It’s got a client-server setup. If you've got multiple PCs or servers, this setup splits the range for computing and sends the results back. Expect a release on March 4-5! Also planning a GPU version.

You can use files from Cyclone’s github, the script will work with it.

Hi there! Thanks for sharing the code. I’m having trouble compiling it on Ubuntu 24.04. Getting errors when I try to 'make'. Anyone got suggestions?

Sounds like a mishap during copy/pasting! You probably missed the <array> header. Here’s the corrected version again: https://github.com/NoMachine1/Cyclone I’m hitting about ~380M keys/s here.

@nomachine: Huge thanks, that really did the trick!

Prefix and auto_random setup: https://github.com/B0dre/Cyclone_Prefix

Hey @b0dre, Could you modify the code for a Windows 64 project? Right now, it’s just for Linux. Thanks a bunch!

I don't have a Windows setup for testing, only macOS and Linux. Try using WSL (Windows Subsystem for Linux) or check this out: https://github.com/NoMachine1/Cyclone which runs faster than mine.