Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: SHAllenge – Compete to get the lowest hash

shallenge.quirino.net

Post reply on HN
Print view View original

121–130 of 136 posts

Re: Show HN: SHAllenge – Compete to get the lowest hash

#121

Post by jhobag »

mindcandy wrote: post #62
moffkalast wrote: post #59

Earlier quoted context omitted.

You could stake fiat in a bank account I guess?

That's pretty much Tether. You could probably bootstrap a POS coin by staking Tether. There would be a lot of details in getting it to eventually grow independent of Tether. I don't claim to know how that could happen. Especially since Tether is not a highly-regarded/trusted token in the community.

lol what this nonsense. tether's regarded as the safest bc their biz model is providing entities usdt for usd without offering interest on the cash.

now why would any entities want to deposit usd into a private bank and receive usdt without any yield on their cash? you think you understand tether so well maybe think about it for a few minutes

Re: Show HN: SHAllenge – Compete to get the lowest hash

#122

Post by kadajett2 »

Just got to 11 zeros with some beginner rust code! Gonna keep it running for a while to see if I can get higher than 24, but this was a ton of fun! And it was my first time doing parallel rust code!

I couldn't get wgsl to work properly, as it would need to have a sha256 implementation, and that's a bit heavy for a buzzed coding session. So, I used the simd commands, and that got me up to around 7mh/s on the m1, and a whole lot more on my desktop.

I will share the code once I get all the other optimizations I want in!

Re: Show HN: SHAllenge – Compete to get the lowest hash

#124

Post by kadajett2 »

anacrolix wrote: post #89

I'm sitting on just under 40MH/s on an M1 Pro with a multithreaded Rust implementation I whipped up. I doubt I can go much faster without using the GPU.

Man, I can't get higher than 8mh/s on my m1 Pro with multithreaded rust. I already got 11 zeros, but I am now more interested in beating at least 20mh/s.

Time to comb through this code :D

Re: Show HN: SHAllenge – Compete to get the lowest hash

#125

Post by neonz80 »

kadajett2 wrote: post #122

Just got to 11 zeros with some beginner rust code! Gonna keep it running for a while to see if I can get higher than 24, but this was a ton of fun! And it was my first time doing parallel rust code! I couldn't get wgsl to work properly, as it would need to have a sha256 implementation, and that's a bit heavy for a buzzed coding session. So, I used the simd commands, and that got me up to around 7mh/s on the m1, and a…

You can probably get a 2x speedup by shortening the string so that it fits in one block.

Re: Show HN: SHAllenge – Compete to get the lowest hash

#126

Post by kadajett2 »

neonz80 wrote: post #125
kadajett2 wrote: post #122

Just got to 11 zeros with some beginner rust code! Gonna keep it running for a while to see if I can get higher than 24, but this was a ton of fun! And it was my first time doing parallel rust code! I couldn't get wgsl to work properly, as it would need to have a sha256 implementation, and that's a bit heavy for a buzzed coding session. So, I used the simd commands, and that got me up to around 7mh/s on the m1, and a…

You can probably get a 2x speedup by shortening the string so that it fits in one block.

Ah, nice! I will shorten it after I finish the simd addition! I didn't even think of that. I was just overcomplicating my thread partitions to avoid collisions.

Re: Show HN: SHAllenge – Compete to get the lowest hash

#127

Post by kadajett2 »

neonz80 wrote: post #125
kadajett2 wrote: post #122

Just got to 11 zeros with some beginner rust code! Gonna keep it running for a while to see if I can get higher than 24, but this was a ton of fun! And it was my first time doing parallel rust code! I couldn't get wgsl to work properly, as it would need to have a sha256 implementation, and that's a bit heavy for a buzzed coding session. So, I used the simd commands, and that got me up to around 7mh/s on the m1, and a…

You can probably get a 2x speedup by shortening the string so that it fits in one block.

Using simd properly(I think) and lowering the nonce text, I went from that 7MH/s to 10MH/s

I have a bunch of other ideas to get me to my goal of 20, but I am happy to let this run for a bit! Thanks for the suggestion!

Re: Show HN: SHAllenge – Compete to get the lowest hash

#128

Post by anacrolix »

grishka wrote: post #109
boyter wrote: post #87

Earlier quoted context omitted.

I have always wanted to learn about GPU programming. Now might be the time.

I couldn't resist, so here's the GPU hasher: https://gist.github.com/grishka/c1be1c035f39564debfd4b195bdf... It gets around 232.5 MH/s on the M1 Max. There's also an optimization that exploits the fact that SHA256 is incremental, so the state for the common prefix could be pre-computed and then copied and only updated with the variable part. This yields around 30% performance boost.

I'm able to get 250MH/s with this. M1 Pro

Re: Show HN: SHAllenge – Compete to get the lowest hash

#129

Post by anacrolix »

kadajett2 wrote: post #124
anacrolix wrote: post #89

I'm sitting on just under 40MH/s on an M1 Pro with a multithreaded Rust implementation I whipped up. I doubt I can go much faster without using the GPU.

Man, I can't get higher than 8mh/s on my m1 Pro with multithreaded rust. I already got 11 zeros, but I am now more interested in beating at least 20mh/s. Time to comb through this code :D

yeah I wonder what you could be doing differently. feel free to link it and I'll point out anything that could be the issue.

I just started on a metal implementation and it's at 320MH/s

Post reply on HN