Riccardo Casatta

same for me

utxo set consensus encoding should be in a separate crate so that libraries could use only that (also BlockExtra should probably be there)

muhash https://lists.linuxfoundation.org/pipermail/bitcoin-dev/2017-May/014337.html python impl https://github.com/fjahr/bitcoin/blob/36ec9801a4edb9663ef9ce9ad298233766b903e8/test/functional/test_framework/muhash.py https://github.com/bitcoin/bitcoin/pull/18000

~~Actually, I checked the main executable and there was no change in size~~ Fault on my side, sorry. The main executable reduce size of ~100k over ~13M

> Anyway: on the examples I can compile I see almost 50% improvement?! Seems a bit too good to be true so please double check Wow I see the 50%...

Great, have a look at the examples if you didn't already, and reach me on discord if you have questions

> Oh, I see that you're using block file parsing to get the block data. Most bitcoin related projects do, and I never got to try benchmark the difference. In...

> My favorite trick... lol, here I did exactly the opposite, measuring the busy time as the time the thread spends non-blocked on `recv`. > I'm really surprised. That would...

Finally, I am very happy with the last 2 improvements: decreased memory needed a lot (see readme for stat table). In particular the most obvious but shamely missing, is skipping...

> > is skipping the provably unspendable scripts > > Out of curiosity, how much does it save? It's about 3GB, but in practice it avoids the last doubling of...