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

[Mylon]

Ok. win32 second try, fully rebuilt:

https://www.dropbox.com/s/ikm9307wsx0ydqt/vanitygen_catalyst.zip

To build from source code:

download https://github.com/wyuzhe/vanitygen/archive/master.zip, setup dependencies, nmake -f Makefile.win32/win64 etc.

Once you'we built all the binaries (8 in total), you can wrap the result with AMD APP SDK 2.8 compiler which seems to produce correct code in case you're getting compute errors.

Beware that to get correct version of the compiler you need to download the whole SDK (200MB).

Test it against beta catalyst driver on system. In case you get a crash/your card is frozen, try removing some or all the dlls of compiler wrapper in 32/ or 64/ directory.

Wrapped version, when running with -v, should report:

Code:

Version: OpenCL 1.2 AMD-APP (1016.4)

Newer compiler versions produce incorrect results (block mining indeed unaffected - vanitygen kernel is nontrivial compared to simple sha256 one).

Detailed reports welcome as always.

Working here on 7970 with catalyst 13.3beta on Windows 7 Ultimate 64bit SP1 (precompiled dropbox version)

Directory of C:\Users\Mylon\Downloads\vanitygen_catalyst\64

04/10/2013  01:49    <DIR>          .
04/10/2013  01:49    <DIR>          ..
12/28/2012  11:14        32,635,904 amdocl64.dll
12/28/2012  11:14         1,187,342 amdocl_as.exe
04/10/2013  01:49                 0 amdocl_as.exe.local
12/28/2012  11:14         1,061,902 amdocl_ld.exe
04/10/2013  01:49                 0 amdocl_ld.exe.local
07/01/2012  00:58           214,312 cacert.pem
04/09/2013  16:43            36,865 calc_addrs.cl
04/10/2013  00:29           995,328 keyconv64.exe
04/10/2013  01:49                 0 keyconv64.exe.local
04/10/2013  00:29           976,896 oclvanitygen64.exe
04/10/2013  01:49                 0 oclvanitygen64.exe.local
04/10/2013  00:29         1,769,984 oclvanityminer64.exe
04/10/2013  01:49                 0 oclvanityminer64.exe.local
12/28/2012  11:14            54,784 OpenCL.dll
04/10/2013  00:29           936,960 vanitygen64.exe
04/10/2013  01:49                 0 vanitygen64.exe.local
              16 File(s)     39,870,277 bytes
               2 Dir(s)  17,782,820,864 bytes free

C:\Users\Mylon\Downloads\vanitygen_catalyst\64>oclvanitygen64.exe -p 0 -d 0 1MyL
oN13
Difficulty: 873388193410
Compiling kernel, can take minutes...done!
[27.54 Mkey/s][total 11777605632][Prob 1.3%][50% in 6.0h]                      ^
C
C:\Users\Mylon\Downloads\vanitygen_catalyst\64>

Awesome work!

I stand corrected:

C:\Users\Mylon\Downloads\vanitygen_catalyst\64>oclvanitygen64.exe -vv -d 0:0 1My

Prefix difficulty:                 1330 1My
Difficulty: 1330
Device: Tahiti
Vendor: Advanced Micro Devices, Inc. (1002)
Driver: 1016.4 (VM)
Profile: FULL_PROFILE
Version: OpenCL 1.2 AMD-APP (1016.4)
Max compute units: 32
Max workgroup size: 256
Global memory: -2147483648
Max allocation: 536870912
OpenCL compiler flags: -DDEEP_PREPROC_UNROLL -DVERY_EXPENSIVE_BRANCHES -DDEEP_VL
IW -DAMD_BFI_INT
Loading kernel binary 70ab729a347a3f14074a7eac8b174115.oclbin
Grid size: 2048x2048
Modular inverse: 8192 threads, 512 ops each
Using OpenCL prefix matcher
Match idx: 1
CPU hash: 0f461d94297c69d32942b44245f456225dc97af4
GPU hash: e61b21b8073b27f23a38458ff4f01d9bfd1bb205
Found delta: 4188886 Start delta: 1
[3.88 Mkey/s][total 4194304]                                                   M
atch idx: 1
CPU hash: 0fb3260103484534174b2e65766413393ba24fe9
GPU hash: e60029440a5d5b5b832ab279457988d417e42b91
Found delta: 4190179 Start delta: 1
[5.95 Mkey/s][total 8388608]                                                   M
atch idx: 1
CPU hash: 1554b1c2216a5afa3438a2b79547d6b8b0d88050
GPU hash: e616a47da08faef667df684d28f527ad0d52fd8a
Found delta: 4192106 Start delta: 1
Match idx: 1
CPU hash: bf5ebdc438ba383879fb9b2656e92ec8997746b5
GPU hash: e62a0c60aa38637c0d53030380757e71ac82874b
Found delta: 4186133 Start delta: 1
[8.20 Mkey/s][total 16777216]                                                  M
atch idx: 1
CPU hash: 36c42fa2bca7d1307c06d3028d753a55ec7fed2d
GPU hash: e60aeba133f1f91a6f584b21c72490f66ef8736a
Found delta: 4193992 Start delta: 1
Match idx: 1
CPU hash: 14e646c76751f43676c9e3381804925f11ed065f
GPU hash: e60fca90fb12280ba702356e2e799cae1a0cd70c
Found delta: 4190563 Start delta: 1
[9.35 Mkey/s][total 25165824]                                                  M
atch idx: 1
CPU hash: c81308f4eef2067d3baeda2148aeceb711423171
GPU hash: e609b57dedc601143bd9496e73fedadd96ea4e0d
Found delta: 4194075 Start delta: 1
Match idx: 1
CPU hash: e0f5af3d5899ff72b5e1d0523f125edea778ce6e
GPU hash: e629d0e76c0ced6ae684bba54e6724775002733a
Found delta: 4187109 Start delta: 1
Match idx: 1
CPU hash: 4788907ab9415ed7238ed66d793ab90ae109da17
GPU hash: e60b037285dfe9ab024bf28eb9457a95dbe0cafb
Found delta: 4192807 Start delta: 1
[10.30 Mkey/s][total 37748736]                                                 ^
C
C:\Users\Mylon\Downloads\vanitygen_catalyst\64>

Looks like I'll have to recompile from source to see what it does.

[1506313795]

1506313795

[1506313795]

1506313795

[1506313795]

1506313795

[1506313795]

1506313795

[1506313795]

1506313795

[Kontakt]

Anyone found a solution for the hanging?

[Mylon]

Anyone found a solution for the hanging?

The updated version for Yuzhu seems to do it for me on windows 7, however getting mismatch on pre-compiled version, and having a lot of trouble getting it complied on MinGW... (got no real linux box close at hand)

[Kontakt]

I was using the precompiled rewrap that Yuzhu posted. It's that program that hangs.

[daemondazz]

I downloaded the version at https://www.dropbox.com/s/ikm9307wsx0ydqt/vanitygen_catalyst.zip yesterday and it works on my machine, haven't tried building from source.

[Mylon]

I downloaded the version at https://www.dropbox.com/s/ikm9307wsx0ydqt/vanitygen_catalyst.zip yesterday and it works on my machine, haven't tried building from source.

You actually got keys? it works on mine, but gives bad output. (hence me needing to rebuild it from scratch)

To quickly see it, just set the search for the key really easy like 3 chars.

[Kontakt]

I was wondering what you guys thought of the idea of making a legitimate pool for vanity address mining.
I was looking into the code for it, and it seems pretty workable.
There just needs to be a way to do proof of work with vanityminer. It's the same theory as shares with bitcoin though, so not too complex.

[Mylon]

I was wondering what you guys thought of the idea of making a legitimate pool for vanity address mining.
I was looking into the code for it, and it seems pretty workable.
There just needs to be a way to do proof of work with vanityminer. It's the same theory as shares with bitcoin though, so not too complex.

They are already doing vanity mining, the main thing this new update is for is 7 series amd gpu's. And it looks great... I just need it to get producing keys... (managed to compile the linux version... now I just have to get that running somehow on windows...)

[Kontakt]

I was wondering what you guys thought of the idea of making a legitimate pool for vanity address mining.
I was looking into the code for it, and it seems pretty workable.
There just needs to be a way to do proof of work with vanityminer. It's the same theory as shares with bitcoin though, so not too complex.

They are already doing vanity mining, the main thing this new update is for is 7 series amd gpu's. And it looks great... I just need it to get producing keys... (managed to compile the linux version... now I just have to get that running somehow on windows...)

I meant a legit pool with shared income and better allocation of resources.

[daemondazz]

You actually got keys? it works on mine, but gives bad output. (hence me needing to rebuild it from scratch)

Yep, the key I got out of it seems to be legit, for address 1dazzrAbMqNu6cUwh2dtYckNygG7jKs8S

[Remember remember the 5th of November]

I was wondering what you guys thought of the idea of making a legitimate pool for vanity address mining.
I was looking into the code for it, and it seems pretty workable.
There just needs to be a way to do proof of work with vanityminer. It's the same theory as shares with bitcoin though, so not too complex.

If you are talking about actually mining PRE-EXISTING addresses in their full form, not just small patterns, sure, yes. But if you are talking about just patterns like 1MyAddy, then there is already a pool, it's http://vanitypool.appspot.com

[Mylon]

You actually got keys? it works on mine, but gives bad output. (hence me needing to rebuild it from scratch)

Yep, the key I got out of it seems to be legit, for address 1dazzrAbMqNu6cUwh2dtYckNygG7jKs8S

vanitygen (cpu only) works for me, oclvanitygen (gpu) gives crap output 2 Mkey on cpu vs 28 Mkey on gpu... so wish I could use my gpu.

[Kontakt]

I was wondering what you guys thought of the idea of making a legitimate pool for vanity address mining.
I was looking into the code for it, and it seems pretty workable.
There just needs to be a way to do proof of work with vanityminer. It's the same theory as shares with bitcoin though, so not too complex.

If you are talking about actually mining PRE-EXISTING addresses in their full form, not just small patterns, sure, yes. But if you are talking about just patterns like 1MyAddy, then there is already a pool, it's http://vanitypool.appspot.com

That's not a pool, that's a joblist board. It's exactly like solo mining bitcoin. For people with low hash, it's useless. Vanity address mining is significantly more profitable that mining, in theory, but for me any of the addresses listed will take ~30 days for a 50% cdf. I can't mine for 30+ days with no return.

The fundamental thing that I'm saying vanitygen is missing is pooled rewards, with pooled mining. I was suggesting a pool that pays that same way a bitcoin pool does, with percentile shares per address. So everyone could be working on the same address simultaneously, and when someone gets it,the rewards are distributed to the miners based on work put in, rather than the current way of a bunch of people going solo, and one person getting the entire reward.

[fizzisist]

I was wondering what you guys thought of the idea of making a legitimate pool for vanity address mining.
I was looking into the code for it, and it seems pretty workable.
There just needs to be a way to do proof of work with vanityminer. It's the same theory as shares with bitcoin though, so not too complex.

If you are talking about actually mining PRE-EXISTING addresses in their full form, not just small patterns, sure, yes. But if you are talking about just patterns like 1MyAddy, then there is already a pool, it's http://vanitypool.appspot.com

That's not a pool, that's a joblist board. It's exactly like solo mining bitcoin. For people with low hash, it's useless. Vanity address mining is significantly more profitable that mining, in theory, but for me any of the addresses listed will take ~30 days for a 50% cdf. I can't mine for 30+ days with no return.

The fundamental thing that I'm saying vanitygen is missing is pooled rewards, with pooled mining. I was suggesting a pool that pays that same way a bitcoin pool does, with percentile shares per address. So everyone could be working on the same address simultaneously, and when someone gets it,the rewards are distributed to the miners based on work put in, rather than the current way of a bunch of people going solo, and one person getting the entire reward.

How would you structure a proof of work? My only idea is to accept partial matches to the pattern as proof if work, and pay earn a share percentage for that proportional to how "close" it is to the desired pattern. This seems complicated, but probably possible. I need to spend some time thinking about how the math would work out.

I've actually been working on building my own site for this recently, but have started by building it simply as a job list as well, to get that working first. I'm calling it a "market place" instead of a pool, since I think that name is more apt. I'm happy to hear any other ideas on how to implement pooled mining, while still minimizing risk for the pool operator. I might start a new thread for the discussion.

[nimda]

I was wondering what you guys thought of the idea of making a legitimate pool for vanity address mining.
I was looking into the code for it, and it seems pretty workable.
There just needs to be a way to do proof of work with vanityminer. It's the same theory as shares with bitcoin though, so not too complex.

If you are talking about actually mining PRE-EXISTING addresses in their full form, not just small patterns, sure, yes. But if you are talking about just patterns like 1MyAddy, then there is already a pool, it's http://vanitypool.appspot.com

That's not a pool, that's a joblist board. It's exactly like solo mining bitcoin. For people with low hash, it's useless. Vanity address mining is significantly more profitable that mining, in theory, but for me any of the addresses listed will take ~30 days for a 50% cdf. I can't mine for 30+ days with no return.

The fundamental thing that I'm saying vanitygen is missing is pooled rewards, with pooled mining. I was suggesting a pool that pays that same way a bitcoin pool does, with percentile shares per address. So everyone could be working on the same address simultaneously, and when someone gets it,the rewards are distributed to the miners based on work put in, rather than the current way of a bunch of people going solo, and one person getting the entire reward.

How would you structure a proof of work?

Mining vanity addresses is almost exactly what Bitcoin mining is. For Bitcoin, you're looking for a vanity hash; it must start with a certain number of zeroes. Addresses are, in essence, hashes. The proof of work, a partial match, is not really much different.

[jacky4566]

Hey guys. Im getting an error with Oclvanityminer64
This is the error:

Code:

Get work request failed: Peer certificate cannot be authenticated with given CA certificates

And this is the launch code: (On Win8 BTW)

Code:

ECHO Starting Vanity miner
oclvanityminer64.exe -u http://vanitypool.appspot.com -a 1CsMpcqmm9CDXU8j26E3w4iCdgt8Xbiqsi -d 0
timeout 30

Any ideas?

[jaywaka2713]

try https://vanitypool.appspot.com/getWork as your URL

[jacky4566]

Same error. I should also point out that this only happens with the 64bit version so that just adds to the confusion.

[fizzisist]

I started a thread to discuss ideas for pooled mining of vanity addresses. Please let me know if you have any ideas!

https://bitcointalk.org/index.php?topic=184132

[bit_NINj4]

Also, the private key may be leaked to your shell history file.  Kill your shell and shred your shell history file after importing.

How does one clean the shell history?