Topic: Pretty Addy Giveaway - part 2 (Read 49078 times)

Yeah, I think it might be. I tested my theory above by using the private key 0000....0001 on bitaddress, and then using the associated public key on VanitySearch. Out of 10 keys I generated, only 2 of them correctly combined on bitaddress (although obviously all combined correctly on VanitySearch). So obviously it is not related to modulo n, since it would be impossible to have 8 different keys which are all greater than n when added to the private key 0000....0001.

Edit: Poked around the code and found the culprit: https://github.com/JeanLucPons/VanitySearch/blob/c8d48ce5f03f5357c0e87cbdb3e1e93cd50af88b/main.cpp#L314

So when combining partial keys, VanitySearch simply checks all 6 combinations of symmetry and endomorphism and then returns whichever key matches the address in the text file. bitaddress.org obviously doesn't do this and just does a straight addition.

Here is how you would use bitaddress to find the right key:

Vk will be the partial private key generated by VanitySearch, while Bk will be the partial private key generated by the user on bitaddress. L1 and L2 will refer to the values below, as specified here: https://github.com/JeanLucPons/VanitySearch/blob/c8d48ce5f03f5357c0e87cbdb3e1e93cd50af88b/main.cpp#L255


The six combinations that VanitySearch checks are as follows:


You can actually work all this out using a combination of bitaddress's add and multiply functions, as well as knowing how to negate a private key modulo n (which is just n minus your privkey), but it doesn't seem like there would be any easy way to explain this to a user who doesn't really understand what they are doing without huge potential for error. The only other option would be to disable symmetry and endomorphism when searching for a key in VanitySearch.

Hmm, very interesting - nice find! I do recall my first attempt at splitkey segwit vanity working when I tested it myself, but it didn't work when I tried to do it for someone else so I figured I just somehow messed up in my own testing. That might explain why I was able to do it the first time.

I do see a GitHub issue where someone asked if symmetry and endomorphism optimizations could be turned off when doing split-key generation for a third party, so I wonder if that could be related. Unfortunately I don't have the time to look through the code myself right now 

"Invalid input" means exactly that - one of the two keys you have copied in is invalid. I've double checked and both the keys I gave you above are valid, which means either you are not copying/pasting them accurately, or you did not accurately save your original private key.

Double check everything again, and if it still fails, then generate a new key pair and I will generate some new addresses for you.

Have you tried to reproduce this example?

@LoyceV @o_e_l_e_o, I appreciate the both of you for your efforts, thank you so much, but unfortunately, the operation did not work, and I can assure you guys that I followed the instructions as it is on the first page, but after inputting the first private key which i save and also inputting the partial private key that was provided here, I got the error in the screenshot below

Probably, there's something am doing wrong, I am not sure honestly, but I am very sure I followed the instructions as its written in page 1 of this thread.

Here you go:

PubAddress: 15starM7dZoz3bzD4utKJ6RRxXGv2mCRXq
PartialPriv: 5K9effT7wMAjcDRcLfsp9XX2QwStGUudKMU3G3skmHkNLD3qB3x

PubAddress: 1FiveSt1NBDYRd9Ce4LBceHCnjSXza4H9x
PartialPriv: 5JeHJcfpqPWeMCB3JUrEgdyr7ch4JAvYwoRHcu5Ef7XrBu9zGty

---

So, I've just been poking about with this interaction between VanitySearch and bitaddress and discovered a few things.

Firstly, it works sometimes, but not other times. My initial thought was that since both pieces of software say they are adding the private keys together to reach the final private key, that it has something to do with how they deal with resulting private keys which are beyond the upper limit of the curve order. I thought that if the resulting private key is below the upper limit then they both return the same private key, but if it is above the upper limit then something is different with how they wrapping it back around and so you end up with different private keys and addresses. Both VanitySearch and bitaddress say they are simply adding the keys together modulo n, however, so this doesn't appear to the case. *shrug*

Secondly, it is not unique to segwit addresses, and the same behavior can be seen with legacy address.

However, you can still make it work by just generating a bunch of addresses with the relevant prefix and then testing them yourself in bitaddress one by one using the two public keys rather than the two private keys. You will eventually find one that works.

There's a problem: oclvanitygen broke after updating graphics drivers, and I can't compile it anymore. Somehow it's linked to the wrong libraries by now.

I tried WhyFhy's site 1SPLiTKEY.com, but didn't receive an email. @WhyFhy: can you check this?

Hi LoyceV,
Thank you for such a wonderful giveaway, i have come to have my own slice of the cake .
please find my details below.

Public key :   
04532E32E45E8F5EB098D740B2CC926F53BEE40566ECF393CF5D87085FBD91971AE00F226517013 A453B19945E3EFBF0234B18A4573D257A6889067346C890E801

Prefix: 15star (if this is possible, then i will go with it since "1Fivestar" which would have been my preferred prefix is more than 6 characters which is the limit , but incase for some reasons, this doesnt or can not work, then check the option below).

Second Prefix(incase first prefix does not work): 1FiveSt

i will prefer case sensitive.
Thanks much for all the effort. 

I got you this:
Address: 1FutureZDmv2GbmJEkugaGvCvExbpqAaEj
PrivkeyPart: 5J7g65w4fytWk69evhZH7WbSj5ErUj8M9QG2Hr3cDGkbrHzwgog

I can do it for a fee, but keep in mind it can't be much longer: every additional character makes it 58 times harder to find. So 7 characters can take a while and 8 might not be possible.

Pardon my lapses, here is the new Prefix.


Edited;

Is there a fee if I want it to be more than 6 characters? (I will save up for this in-case)

Lol
Please read the part on Limitations

Greetings Loyce!
I hope my Prefix is eligible.

Public Key:
Prefix:

The best I can do for 7 characters is case insensitive:
Address: 1BTCGoSUfdnhdj7rMuqupJEvbHo8UW9aQE
PrivkeyPart: 5KZGcbD6W5j9eo42U7725zyD9zV8dCGNrEMQE8n8ZW2U3bPurrY

Hello Loyce,

I hope you are well!

Would you mind getting me another GOSU address? 

Public Key: 048B7F75DB3A24342206BAC18DAB1E176D2E6C222B588D613BA3B89BB7CB523F10BCBAE8D9F57C8 5060B122FFAC2023D4B5979783328F1EBF87B6120038952240C

Prefix: 1btcgosu

Thank you very much in advance!

2 month Bump 
Free Pretty Addy still available, just follow the Instructions and post here.

Thank you LoyceV! Glad you got your setup working again. I experimented with vanitygen and vanitysearch myself in the past few days and luckily got my own 8-digits/character key working after I ran into a few issues.
bc1qsam00000rymzeeyfwh7f3vcmc5y79fsqam5rc2

Should've told you that you don't have to search for my prefix, sorry!

I will keep a look on this thread and maybe hop in another time when your system breaks

I got it to work again, but 6 characters case-insensitive is too much for this giveaway (it takes a day for 50% chance on my old system). I got you this:
Address: 1saMuELQnb5egBTck8WfFahtesjLSbgtX (Balance: )
PrivkeyPart: 5JktvqcBdMK8odxgXY6akT2k8Vw2rJe7B3W6ogjZGrsdEijQQRi

No problem, good luck fixing your setup

Maybe it's time for me to fire up the vanitygen myself

Nope, sorry My most recent upgrade broke things:
Getting it to work (years ago) was a lot more work than I expected, so I need some time to fix it again. Without GPU, it takes far too long to generate.

Maybe someone else can jump in for now?

Heyy, can you try "1SAMUEL..." please ? Wish I could go for 1sam00, too bad there are no zeros

public:
private:

thank you LoyceV!:)