Hash bitcoin brain wallet address from passphase?

[philsmd]

how do you want to customize it? are you trying to generate the password candidates in a very specific way ? because for "just" sha256 you do not need to customize anything

[kangduwang]

oh. we are suddently speaking about brain wallets? The OP didn't mention that and the other post was modified to include "brain wallet". Yeah, that's a different story.

---

Nevermind, the title says it too. I missed that. sorry

---

Therefore, if you are sure that it is just a sha256 hash, it could be feasible with a lot of hardware. What is your character set?
The number of combinations 53981544960 seems quite strange for 6 characters because that is something like 61^6 = 51520374361 but not exactly

it use all upper, lower & numeric characters, no special chars. Oh i calculate wrong, i used 64 instead of 61

[kangduwang]

how do you want to customize it? are you trying to generate the password candidates in a very specific way ? because for "just" sha256 you do not need to customize anything

My idea is add a new hash which convert the passphase into address & compare value. The brute force method still work the same way

[undeath]

You can use the sha256 kernel and modify that to add ecdsa public key derivation, then ripemd160. But well, that's what needs to be done.

[philsmd]

yeah I agree, the code seems to be this one:
https://github.com/pointbiz/bitaddress.o...js#L27-L32

and this means that you need to do the elliptic curve operations.

[undeath]

instead of using hashcat you are probably better off using an existing implementation: https://github.com/ryancdotorg/brainflayer

[kangduwang]

yeah I agree, the code seems to be this one:
https://github.com/pointbiz/bitaddress.o...js#L27-L32

and this means that you need to do the elliptic curve operations.

wait, do hashcat support javascript?

[kangduwang]

instead of using hashcat you are probably better off using an existing implementation: https://github.com/ryancdotorg/brainflayer

Thanks a lot, i'll give it a try

[philsmd]

before jumping into this I would recommend doing the math because 64^6 (as far as I understood you corrected your first number above) is quite a huge number.
You would need to have a huge amount of bitcoin in that wallet and a lot of hardware to throw at it (don't forget about the energy cost etc too) to justify this investment/time/cracking.
but it is of course much better compared to hashing algorithms like scrypt etc, therefore it could be feasible with "just" sha256

---

I think with GPU support it would be much more efficient and a mask like -1 ?d?l?u ?1?1?1?1?1?1 would be done in just a couple of minutes with a modern GPU for -m 1400 (without the elliptic curve operations)... but I agree with undeath that adding elliptic curve code is probably not worth it either and you should do the speed test with already available crackers to get a feeling how long it could take (but should be doable)

[undeath]

You would need to have a huge amount of bitcoin in that wallet and a lot of hardware to throw at it (don't forget about the energy cost etc too) to justify this investment/time/cracking.

It's slow and definitely nowhere near the performance you could achieve on GPU, but my old Intel i7-2600 can do 62k keys per seconds with brainflayer, per core. It would probably take 2-3 days to crack the brain wallet.