Bob McElrath

I have forward-ported this patch to python-bitcoinlib 0.11: https://github.com/mcelrath/python-bitcoinlib/tree/epoll I am unsure of the status of this patch and have not tested it yet. I've lost the development stack we...

I'll second the idea of having a pure python library. For security critical code, minimizing dependencies is paramount, and while I like Rust, I don't think wrapping it is the...

For reference: Miniscript in: javascript: https://github.com/thmour/miniscript python: https://github.com/bitcoin/bitcoin/pull/17975 rust: https://github.com/rust-bitcoin/rust-miniscript

Agreed. A pull request would be welcome ;-)

The `getinfo` RPC call has been deprecated in favor of a bunch of more specific "info" calls including `getblockchainfo`. See also `getnetworkinfo`, `getwalletinfo` etc. Whether this works depends on the...

It's curious to note that when constructing a new bead, all the parents of that bead have exactly the same descendant work (which is the work in the one new...

Wait, isn't ollama using llama.cpp as your runner? I mean, #10740 is kind of an egregious estimation fail, but the real work is in llama.cpp... I've recently become enamored with...

Same issue observed with 3x 7900xtx. I can see that with e.g. qwen3:30b (Q4_K_M quantization) `ollama ps` reports 76Gb usage while `andgpu_top` systematically reports about half that. ``` $ amdgpu_top...

I believe that's answered positively by the timestamps and structure of the calling code. A better test would be to reproduce this with regtest which I'll create a script to...

Not yet. I've been running my morning script all week and haven't seen it happen again. On Wed, Sep 18, 2024, 12:14 PM maflcko ***@***.***> wrote: > A better test...