Why Nostr? What is Njump?
2024-07-04 15:21:26
in reply to

mleku on Nostr: it reminds me, zeena has talked about using the cgo to import libsecp256k1 into Go - ...

it reminds me, has talked about using the cgo to import libsecp256k1 into Go - and i think the speedup is like 10x or so, but just keep in mind that ecah additional bit is a square or more of each previous bit, so where i said 6 months, if you used a C based vanity generator maybe you'd get it in 2 months

there is no guarantee in these things, it is based on random numbers and hash functions and clockwork arithmetic so your mileage literally varies, the solutions can come out anywhere, at any time, it is possible that you will start the miner and the next day it has the secret key

no, it's got nothing to do with the speed of 5 to 8 bit scaling of bech32, that is the cheapest part of the operation, it is linear based on the bit length in terms of CPU ticks

no, it is literally the cost of deriving a public key from a secret key using secp256k1 elliptic curve signature algorithm, in this case, for BIP340 pubkeys but they are the same as standard ecdsa pubkeys except for the funny thing about inverting the key value if it comes out odd, to eliminate the last bit and squeeze it into 32 bytes 256 bit instead of 257 as the standard ecdsa public key

that last operation is literally a check and invert operation, it is ridiculously cheap, like 2 or 3 cpu cycles at best, compared to the hundreds that are used to do the scalar multiplication operation that gives you the pubkey at all... it's probably thousands, in fact, and probably varies at each step based on how much the value overflows through over and over again into your fixed length (256 bit) finite field
Author Public Key
npub1fjqqy4a93z5zsjwsfxqhc2764kvykfdyttvldkkkdera8dr78vhsmmleku