Vanitygen: Vanity bitcoin address generator/miner [v0.22]

[stevegee58]

You poor, poor Mac people.  If you want to play with the big boys you're gonna have to break out of the walled garden and get some real power tools. 

[1713860874]

1713860874

[1713860874]

1713860874

[1713860874]

1713860874

[1713860874]

1713860874

[axus]

I compiled it on Windows, and I had the same problem.  Use the -o keyfile.txt option to write to file instead of standard out.

He's not the only one with the issue.

I think I had that same issue trying to gpu-mine with self-compiled code on linux.  I just gave up and use Windows.

[oldschool]

I posted a thread here:

https://bitcointalk.org/index.php?topic=108451.0

Asking how to decrypt the protkey back to a privkey when using the -e flag and specifying a password.

Output like this:

Pattern: 1wtf
Address: 1WtFgLCoA56syuVLbvzCNzooTrKEpqyA8
Protkey: PsRUGbymw545hfYUAg5v8uNuVcxwtXWg646JkZ7nSm1dDRvHk5rTd8HrQv452ctz6fP2

and -e password used was 'wtf'

I tried several wallets, none of them recognized the protkey, is there a specific wallet or a command line tool for decrypting the protkey back to a privkey?

[flatfly]

Is there anyway to force generation of the exact same sequence of addresses twice?
(I actually hope not.)

I tried playing with seed files but perhaps I misunderstood what those are for.

[oldschool]

Is there anyway to force generation of the exact same sequence of addresses twice?
(I actually hope not.)

I tried playing with seed files but perhaps I misunderstood what those are for.

Seed files would be to seed the random number generator.  If you ran the same seed file at the exact same computer time on the exact same computer it could be possible to produce the same results or at least greatly reduce the entropy needed to find the results.  The seed file is any file that will change the outcome of the random selection.

There is a lot more involved with the internal random number generator but this is the short explanation.  Basically the seed file gives another source of randomized input.

[Kazimir]

Vanitygen is great, but I'm having an issue with the OCL version: it works, but if I abort it (by pressing Ctrl+C or terminating the process or any other way), my PC completely freezes and I have to turn it off & on.
The non-OCL version (either 32 or 64 bit) do not show this behavior.

I'm running Windows 7 x64 and a GTX560 with latest drivers.

[tgbtc]

Can't get the ocl version to work...

On nvidia:

Code:

vg_ocl_context_callback error: CL_MEM_OBJECT_ALLOCATION_FAILURE error executing CL_COMMAND_NDRANGE_KERNEL on GeForce GTX 285 (Device 0).

clWaitForEvents(NDRange,0): CL_MEM_OBJECT_ALLOCATION_FAILURE
Device: GeForce GTX 285
Vendor: NVIDIA Corporation (10de)
Driver: 285.62
Profile: FULL_PROFILE
Version: OpenCL 1.0 CUDA
Max compute units: 30
Max workgroup size: 512
Global memory: 1073741824
Max allocation: 268435456

on 7970:
(seems it is finding matches, but disregarding due to mismatch?

Code:

CPU hash: 6707a76e848f5e9368c2d9d9ec6d60880df44891
GPU hash: 020b0bda39a9bbb2eeddd194a8be20934835dff6
Found delta: 94027 Start delta: 1

Numbers look good though on the ati:

Code:

Difficulty: 873388193410
[27.27 Mkey/s][total 4630511616][Prob 0.5%][50% in 6.1h]

(for reference, on an i7 @4.2ghz, vanitygen64 does about 1Mkey/s case insensitive)

[Lethos]

Amazing program.
Still playing with it, did a few which only took a few minutes.

Trying longer ones, but it seems to be be measured in many years by probability of success so I think it only find it if extremely lucky.
So I don't think I'll be seeing 1BattLeLords (First name of a game I'm developing) any time soon.
None the less very cool, my little laptop even happily manages 5 Mkeys/s which surprised me.

Yeah I got 1Lethos

[stevegee58]

Bitcoin addresses are Base 58 so every time you add a digit to the vanity part it multiplies the number of search permutations by 58.

[Lethos]

Bitcoin addresses are Base 58 so every time you add a digit to the vanity part it multiplies the number of search permutations by 58.

I noticed I know I was just being over optimistic in thinking I could get a long one.

[TTBit]

Where can I get keyconv for windows? I can't find it.

sorry, nevermind. Didn't see it in .21. Got it from .20.

[salfter]

Can't get the ocl version to work...

on 7970:
(seems it is finding matches, but disregarding due to mismatch?

Code:

CPU hash: 6707a76e848f5e9368c2d9d9ec6d60880df44891
GPU hash: 020b0bda39a9bbb2eeddd194a8be20934835dff6
Found delta: 94027 Start delta: 1

I had the same thing happen with my 7750.  IIRC, I got it working by passing the -S (safe mode) option to oclvanitygen.

[Dabs]

Mr Vanitygen author, can you make a version that spits out compressed private keys, or the keys that begin with letters instead of 5?

[flatfly]

Which gets me wondering, does Vanitygen search the entire keyspace? (*)
If not, what proportion of the entire keyspace does it search?

(*) What about addresses associated with compressed keys?

[BurtW]

Which gets me wondering, does Vanitygen search the entire keyspace? (*)
If not, what proportion of the entire keyspace does it search?

(*) What about addresses associated with compressed keys?

What I understand it does:

1) Generate a random private key (or increment from the previous random private key)
2) Calculate the corresponding public key from the private key
3) Calculate the corresponding bitcoin address of the public key
4) Compare, if there is a match output else go back to step 1)

Does that help to answer your question or not?

[deepceleron]

Which gets me wondering, does Vanitygen search the entire keyspace? (*)
If not, what proportion of the entire keyspace does it search?

(*) What about addresses associated with compressed keys?

The starting point of a phrase search is random, it gets a private key from the openssl rand ecdsa library and starts searching from there, simply incrementing the key. After searching many addresses for a phrase match, it will again get another random private key. The keys searched and returned with a found vanity phrase in the corresponding Bitcoin address can be from anywhere in the range of valid keys, but certainly it cannot "search the entire keyspace", as that would take somewhere just north of the age of the universe.

[flatfly]

Which gets me wondering, does Vanitygen search the entire keyspace? (*)
If not, what proportion of the entire keyspace does it search?

(*) What about addresses associated with compressed keys?

The starting point of a phrase search is random, it gets a private key from the openssl rand ecdsa library and starts searching from there, simply incrementing the key. After searching many addresses for a phrase match, it will again get another random private key. The keys searched and returned with a found vanity phrase in the corresponding Bitcoin address can be from anywhere in the range of valid keys, but certainly it cannot "search the entire keyspace", as that would take somewhere just north of the age of the universe.

Hmm perhaps my question wasn't very clear. I'm aware of what you said, what I meant to ask is, does it search within the full (2^256) range of possible private keys or a subset of it. I would have checked the source myself but my C/C++ is a little too rusty...

[stevegee58]

Hmm perhaps my question wasn't very clear. I'm aware of what you said, what I meant to ask is, does it search within the full (2^256) range of possible private keys or a subset of it. I would have checked the source myself but my C/C++ is a little too rusty...

It picks random starting points within the 256 bit private key space and increments from there.  So yes it considers the entire 256 bit space.

[flatfly]

Hmm perhaps my question wasn't very clear. I'm aware of what you said, what I meant to ask is, does it search within the full (2^256) range of possible private keys or a subset of it. I would have checked the source myself but my C/C++ is a little too rusty...

It picks random starting points within the 256 bit private key space and increments from there.  So yes it considers the entire 256 bit space.

Sorry to insist, but are you sure? How do compressed private keys (introduced in bitcoind/bitcoin-qt 0.6.0) fit in there? From what I've been able to read, vanitygen doesn't support them and they are actually separate from noncompressed keys (ie, can't be expressed as normal uncompressed base58 keys) - so what proportion of the 2^256 search range do they represent? Or am I totally misunderstanding compressed keys?

[stevegee58]

Well then I guess I have no idea.  I'm a C++ guy in my day job but I haven't looked into the fine details of the BTC algos.

OP hasn't posted for over a month so I'm not sure he's around anymore.