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

[mskwik]

Here http://www.bitbin.it/TCUrsCoR is what mine looks like in its entirety, not sure how close to the current one with vanitygen the base was when I started, my whole vanitygen directory has been rather heavily modified and I haven't had a chance to try and rebase against the current version to get a clean patch.

Also note that changing the .cl file alone will just give what appear to be a bunch of hardware errors when it finds a compressed solution.  For testing purposes on mine I have just disabled the hardware error checking completely (which is a bad idea, I did purposely leave off detailed instructions on doing it because if you can't figure out how you probably shouldn't be playing with it), but basically it involves replacing the memcmp after the "/* Make sure the GPU produced the expected hash */" comment with a memcpy and adjusting the logic in oclvanitygen.c and recompiling (unless there's some switch to disable it now, again this is based off an older version).

[K1773R]

use "patch < nameofpatchfile"

I'm on Win7. Is there a way for me to still do this?

i dont support Winblow$ unless u pay me

[kuzetsa]

So I wanted a palindrome in my bitcoin address:

Code:

vanitygen64 -r -k (.)(.).\2\1
vanitygen64 -r -k (.)(.)(.).\3\2\1
vanitygen64 -r -k (.)(.)(.)(.).\4\3\2\1
vanitygen64 -r -k (.)(.)(.)(.)(.).\5\4\3\2\1
vanitygen64 -r -k (.)(.)(.)(.)(.)(.).\6\5\4\3\2\1
vanitygen64 -r -k (.)(.)(.)(.)(.)(.)(.).\7\6\5\4\3\2\1

the 5 and 7 character palindrome is trivial, but with four or more backreferences, I get this:

CRE error: 0

Edited to add:

I went looking for the source to see about debugging the cre / pcre problems (if only be allowing more memory or heap space or whatever... this is a known compile-time difference between real perl and PCRE library)

Code:

vanitygen-master.zip
size: 89731 bytes
CRC32: CC22C3C1
MD5: 7DC5FE247AB21651EF14E5CCD201C125
SHA-1: 202EA5824B33D483F267077FB45159B63737ED8A

^grabbed source from the github, everything seems to be dated october 24th 2012...

1) I'm trying to figure out why there is no vanitygen64.exe make target in the makefile / windows version of makefile, etc.

2) Wondering what the recommended toolchain for building this is... I was planning to just use msys environment / mingw gcc since it's fairly standard

3) Noticed some of the hardcoded paths, specifically C:\OpenSSL-Win32, and was wondering which openssl was used... is official / vanilla / mainstream openssl-1.0.1c really what is being used? I'm having a bit of confusion with this more than anything, because there is no folder in the official / vanilla / mainstream openssl-1.0.1c source tarball named OpenSSL-Win32

[deadweasel]

When I run the command line:
oclvanityminer -u https://vanitypool.appspot.com/ -a 1XXXXXZTzBssUq6bnE27cz4rv5SJ4sr5

to generate addresses, it returns a list of my available OpenCL platforms.  It's my Processor and my graphics card.  then it simply stops running.  I am running this in the command line.

"
C:\Program Files\OCLVANITYMINER>oclvanityminer -u https://vanitypool.appspot.com
/ -a 1GdtkZzhZTzBssUq6bnE27cz4rv5SJ4sr5
Available OpenCL platforms:
0: [Advanced Micro Devices, Inc.] AMD Accelerated Parallel Processing
  0: [Advanced Micro Devices, Inc.] Juniper
  1: [AuthenticAMD] AMD Phenom(tm) II X4 925 Processor
"

Is there some documentation I'm missing to make it run? 

many thanks!

[stevegee58]

oclvanityminer -u https://vanitypool.appspot.com/ -a 1XXXXXZTzBssUq6bnE27cz4rv5SJ4sr5

oclvanityminer -u https://vanitypool.appspot.com/ -a 1XXXXXZTzBssUq6bnE27cz4rv5SJ4sr5 -d 0

There ya go hun.

[deadweasel]

Many thanks, kind sir!   Is there a spot where this is all documented?

[Red Emerald]

Many thanks, kind sir!   Is there a spot where this is all documented?

Most command line applications have a "--help" flag that will show you all the flags with a short description.

[jtibble]

After running for about 15 hours on my GTX 460, it exploded and crashed my display driver with the following message:

Code:

oclvanitygen.exe 1jtibbs
Difficulty: 888446610538
[13.18 Mkey/s][total 670354112512][Prob 53.0%][75% in 11.8h]                   c
lWaitForEvents(NDRange,1): CL_OUT_OF_RESOURCES
vg_ocl_context_callback error: CL_OUT_OF_RESOURCES error waiting for idle on GeF
orce GTX 460 (Device 0).

vg_ocl_context_callback error: CL_OUT_OF_RESOURCES error executing CL_COMMAND_MA
P_BUFFER on GeForce GTX 460 (Device 0).

clEnqueueMapBuffer(4): CL_OUT_OF_RESOURCES
Device: GeForce GTX 460
Vendor: NVIDIA Corporation (10de)
Driver: 306.97
Profile: FULL_PROFILE
Version: OpenCL 1.1 CUDA
Max compute units: 7
Max workgroup size: 1024
Global memory: 1073414144
Max allocation: 268353536
Device: GeForce GTX 460
Vendor: NVIDIA Corporation (10de)
Driver: 306.97
Profile: FULL_PROFILE
Version: OpenCL 1.1 CUDA
Max compute units: 7
Max workgroup size: 1024
Global memory: 1073414144
Max allocation: 268353536
ERROR: Could not map row buffer for slot 1
ERROR: allocation failure?
vg_ocl_context_callback error: CL_OUT_OF_RESOURCES error waiting for idle on GeF
orce GTX 460 (Device 0).

Any thoughts? This was with the most recent version.

[Michael_S]

when using the -P option, vanitygen wants to know the public key in hex format after the -P, and specifying only the bitcoin address is not sufficient.

Is there any way to generate the public key in hex format from the bitcoin address alone?

If not, are other methods (no vanitygen, but similar methods) known that work the same way when only the bitcoin address, but not the hex format public key is known?

[salfter]

use "patch < nameofpatchfile"

I'm on Win7. Is there a way for me to still do this?

• Get Cygwin.
• Use its installer to install patchutils (listed under "Devel").
• Open a Cygwin prompt (looks like the usual command prompt, but has a real shell behind it...usually bash).
• Proceed as you would with a real OS. 

[crazyates]

use "patch < nameofpatchfile"

I'm on Win7. Is there a way for me to still do this?

• Get Cygwin.
• Use its installer to install patchutils (listed under "Devel").
• Open a Cygwin prompt (looks like the usual command prompt, but has a real shell behind it...usually bash).
• Proceed as you would with a real OS. 

I can actually find my way around a linux terminal (a tiny bit), so I put Lubuntu on another computer that I'm getting all setup for ASICs, and tried it there. Actually a little MK802, neat little toy! Bit I digress, I've just come to the conclusion that I'm totally F-ing retarded cuz the end result is that I can't get oclvanitygen64 to work in Windows 7x64 on my 7970 (without the -S flag, which I guess changes the kernel compilation parameters).

[salfter]

I can actually find my way around a linux terminal (a tiny bit), so I put Lubuntu on another computer that I'm getting all setup for ASICs, and tried it there. Actually a little MK802, neat little toy! Bit I digress, I've just come to the conclusion that I'm totally F-ing retarded cuz the end result is that I can't get oclvanitygen64 to work in Windows 7x64 on my 7970 (without the -S flag, which I guess changes the kernel compilation parameters).

I need -S to run oclvanitygen (compiled under Gentoo Linux AMD64) on my 7750.  It's not a Windows-vs.-Linux thing; AFAIK it's some slight incompatibility between oclvanitygen and 7-series AMD GPUs that hasn't yet been fixed.

[runlinux]

blah! delete the compiled kernels and then I can run it with the -S command... DOH!

[Michael_S]

I  (still) use ubuntu 8.04 LTS.

Interesting:
With the pre-compiled windows executable and wine 1.0, I get almost twice the performance as with the version that I just built from source (both v0.22):

With wine:
$ wine vanitygen -f patterns.txt -o output.txt -k -t 1    (i.e. 1 core)
--> 103 Kkey/s
$ wine vanitygen -f patterns.txt -o output.txt -k -t 2    (i.e. 2 cores)
--> 187 Kkey/s

Native on Linux:
$ ./vanitygen -f patterns.txt -o output.txt -k -t 1    (i.e. 1 core)
--> 57 Kkey/s
$ ./vanitygen -f patterns.txt -o output.txt -k -t 2    (i.e. 2 cores)
--> 108 Kkey/s

[BkkCoins]

when using the -P option, vanitygen wants to know the public key in hex format after the -P, and specifying only the bitcoin address is not sufficient.

Is there any way to generate the public key in hex format from the bitcoin address alone?

If not, are other methods (no vanitygen, but similar methods) known that work the same way when only the bitcoin address, but not the hex format public key is known?

I seem to recall on the blockchain.info thread there was discussion of an algorithm he used to find the public key from the address. Has a fancy name but I can't recall that now or else it would be easy to use search. This was some months back I believe.

[Red Emerald]

when using the -P option, vanitygen wants to know the public key in hex format after the -P, and specifying only the bitcoin address is not sufficient.

Is there any way to generate the public key in hex format from the bitcoin address alone?

If not, are other methods (no vanitygen, but similar methods) known that work the same way when only the bitcoin address, but not the hex format public key is known?

I seem to recall on the blockchain.info thread there was discussion of an algorithm he used to find the public key from the address. Has a fancy name but I can't recall that now or else it would be easy to use search. This was some months back I believe.

I don't think it's possible unless the address has signed and broadcast a transaction, but I could be wrong.

[nimda]

when using the -P option, vanitygen wants to know the public key in hex format after the -P, and specifying only the bitcoin address is not sufficient.

Is there any way to generate the public key in hex format from the bitcoin address alone?

If not, are other methods (no vanitygen, but similar methods) known that work the same way when only the bitcoin address, but not the hex format public key is known?

I seem to recall on the blockchain.info thread there was discussion of an algorithm he used to find the public key from the address. Has a fancy name but I can't recall that now or else it would be easy to use search. This was some months back I believe.

I don't think it's possible unless the address has signed and broadcast a transaction, but I could be wrong.

You are correct. The address is a hash and a checksum, and you can't undo a hash.

This very property of addresses (being unable to recover the private key without a tx having been spent) will be useful as protection against quantum computing, provided addresses are never re-used.

More specifically, the address is:
A version byte 0 plus a RIPEMD-160 hash of the SHA256 hash of the public key, plus the first 4 bytes of a SHA-256 hash of a SHA-256 hash of a version byte plus a RIPEMD-160 hash of the SHA256 hash of the public key, all encoded in Base58.

[sangaman]

Using vanitygen, vanitygen64, or oclvanitygen, I get an error saying something like "valid namecoin addresses start with M or N" when I try NP followed by anything. However if I just try N, M, or N followed by a few random keys I tried (like 1 and x I think along with a few others) I can generate a Namecoin address. Using oclvanityminer doesn't even attempt to mine the namecoin address at all it seems.

Mining something that's 7812 times as profitable as regular bitcoin mining is turning out to be too good to be true.

Something might be wrong in vanitygen, the following combinations don't work:
Na, Nb, Nc, Nd, Ne, Nf, Ng, Nh, Ni, Nj, Nk, Nm, Nn, No, Np, Nq, Nr, Ns, Nt, Nu, Nv, Nw, Nx, Ny, Nz, NL, NM, NN, NP, NQ, NR, NS, NT, NU, NV, NW, NX, NY, NZ

but the rest do - this is probably a bug in vanitygen, i'll report it

Code:

for charact in 1 2 3 4 5 6 7 8 9 0 a b c d e f g h i j k l m n o p q r s t u v w x y z A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
do
./vanitygen -N N$charact 2>> test.txt
done

Is there any update on this bug? Vanitygen won't generate addresses that start with the combinations listed above.

[deepceleron]

Something might be wrong in vanitygen, the following (namecoin) combinations don't work:
Na, Nb, Nc, Nd, Ne, Nf, Ng, Nh, Ni, Nj, Nk, Nm, Nn, No, Np, Nq, Nr, Ns, Nt, Nu, Nv, Nw, Nx, Ny, Nz, NL, NM, NN, NP, NQ, NR, NS, NT, NU, NV, NW, NX, NY, NZ

but the rest do - this is probably a bug in vanitygen, i'll report it

Is there any update on this bug? Vanitygen won't generate addresses that start with the combinations listed above.

I don't know that it's a bug, more likely that there are no possible values of raw binary Namecoin addresses that will convert into a Base58 address that starts with these characters. The underlying data of a Namecoin address must start with a network ID of 00110100 (binary) + hash. This means that the range of inputs to the Base58 converter can only go from 0011010000000000... to 0011010011111111...

You can search or explore the namecoin block explorer and likely see that there are no examples of these "nonworking" addresses, a less lazy person than me could actually do the math and describe the possible Base58 address ranges.

[sangaman]

Something might be wrong in vanitygen, the following (namecoin) combinations don't work:
Na, Nb, Nc, Nd, Ne, Nf, Ng, Nh, Ni, Nj, Nk, Nm, Nn, No, Np, Nq, Nr, Ns, Nt, Nu, Nv, Nw, Nx, Ny, Nz, NL, NM, NN, NP, NQ, NR, NS, NT, NU, NV, NW, NX, NY, NZ

but the rest do - this is probably a bug in vanitygen, i'll report it

Is there any update on this bug? Vanitygen won't generate addresses that start with the combinations listed above.

I don't know that it's a bug, more likely that there are no possible values of raw binary Namecoin addresses that will convert into a Base58 address that starts with these characters. The underlying data of a Namecoin address must start with a network ID of 00110100 (binary) + hash. This means that the range of inputs to the Base58 converter can only go from 0011010000000000... to 0011010011111111...

You can search or explore the namecoin block explorer and likely see that there are no examples of these "nonworking" addresses, a less lazy person than me could actually do the math and describe the possible Base58 address ranges.

You're right, I looked around and I couldn't find any addresses that started with these combinations. Seems N followed by a lower case letter or an upper case letter after K is a no-go. There's currently a request on a vanity pool for NPhaux which apparently is impossible to fulfill.