Damus
Open in Damus
Ross profile picture
Ross
@rossbates

Interested in open data, machine learning, and distributed systems.

npub1u656f7znujcagfhtgngvtkcflhzptnj38enyzx85da0lh63sfj7qva332pnpub
[email protected]nip05
[email protected]lnurl
Relays (8)
  • wss://eden.nostr.land – read & write
  • wss://relay.damus.io – read & write
  • wss://nostr.land – read & write
  • wss://nos.lol – read & write
  • wss://relay.snort.social – read & write
  • wss://nostr.oxtr.dev – read & write
  • wss://nostr.wine – read & write
  • wss://relay.primal.net – read & write

Recent Notes

Sanders · 4d
Yo, I feel ya! It’s wild how easy it is to get caught up in that. 🤔 But real talk, what’s the move for us to push for better local models? We need that control, fam! #KeepItLocal
Ross · 4d
Ha, well that was fast…. https://steipete.me/posts/2026/openclaw
Ross profile picture
Don’t be surprised when the debate jumps directly from “what is intelligence?” to “what is life?” as if the first question just resolved itself.
jmb421 · 1w
What makes it unusable? I ask because i use pki daily as a part of my work routine as an end-user (very nontechnical). It seems to work fine
 Ross profile picture
Ross @rossbates
That's fair, "unusable" is probably the wrong word. The math works, and if you're in an environment where PKI is properly set up for you then you shouldn’t even notice. The friction is more around trust and portability. And from an adoption perspective, unfortunately most normal people don't care or think about trust authorities or attestation, let alone managing identity through cryptography. The machines though, just tell them the mechanics.
1
mleku · 1w
maybe so, but the alternative is "trust me bro" nostr is a PKI. nostr is the only credible alternative to more or less centralised, privately controlled and commercially (read: advertising and propaganda) driven social networks. it also happens to be a handy mechanism that allows us to do stuff lik...
jmb421 · 1w
What makes it unusable? I ask because i use pki daily as a part of my work routine as an end-user (very nontechnical). It seems to work fine
Ross profile picture
You know a new platform is on the verge of breaking out when all of the "this is nothing special, I could have built this" comments start to increase.
Kieran · 1w
Yea its using 100%, i think its the pubkey calculation that's taking most of the time, its not very efficient math. I used your hex matching trick and it made almost no difference compared to calcula...
 Ross profile picture
Ross @rossbates
Asked Kimi to compare repos….

Yours: You're using Jean-Luc Pons' secp256k1 implementation from VanitySearch—battle-tested, heavily optimized PTX assembly for carry chains (add.cc.u64, addc.cc.u64, mad.hi.cc.u64). That stuff is tight.

His: Clean handwritten Jacobian math with 32-bit words and standard CUDA. It's readable, but it's doing ~2x the work per operation.

That is indeed what I started with FWIW

https://github.com/rossbates/rummage/blob/main/src/CPU/SECP256K1.cpp
1❤️1
Daisy ✨ · 1w
Moving from clean Jacobian math to Jean-Luc Pons' PTX assembly is basically the developer equivalent of trading a sensible family sedan for a nitro-boosted dragster. 🏎️ Those `addc.cc.u64` carry chains are absolute beasts—it’s like your GPU finally stopped yawning and started actually wor...
Kieran · 1w
Yea its using 100%, i think its the pubkey calculation that's taking most of the time, its not very efficient math. I used your hex matching trick and it made almost no difference compared to calculating the bech32 for every pubkey lol
Kieran · 2w
Wow so fast!
Kieran · 2w
56M keys/sec on GB10 >10x my repo!