Lion - dapplion

CC: @noot @wemeetagain are you avail to participate in the spec-ing process?

Oh damn! :rocket: How can check if my host supports SIMD?

> > Oh damn! rocket How can check if my host supports SIMD? > > cpuid gives you this. A C++ call is here https://github.com/potuz/mammon/blob/main/ssz/hasher.cpp#L43-L55 Thank you! ``` $ cpuid...

That's huge then! Would love to see this in Lodestar. I did some comparisons with Lighthouse on our hashing throughput and somehow Lodestar is x5 slower when bench-marking hashing a...

I've tested using TypedArrays to convert numbers and it's actually slower than the current implementation in as-sha256 ``` utils ✓ hashObjectToByteArray 50023 times 407.5760 ops/s 2.453530 ms/op x1.406 4073 runs...

> @dapplion just to confim, the `endianness` in ssz serialization is `little`? Yes it's little endian

@g11tech Thanks! Keep us updated

For NodeJS where performance is important we should just use Buffer utils which will probably be the faster and more memory efficient. > Given how efficient protobuf creates a string...

@tuyennhv For memory efficiency there's this library that flattens strings. Check it out it's magic https://github.com/davidmarkclements/flatstr

> @tuyennhv For memory efficiency there's this library that flattens strings. Check it out it's magic https://github.com/davidmarkclements/flatstr Note this was recommended by Ben (the libuv mantainer)