BitCrack - A tool for brute-forcing private keys

[almightyruler]

..Ty for your explanation, I can't pm u back because of messaging limits. So basically there is no way to combine really fast random hex generator to work with the calculation process? So all we can do is randomly get starting points and from there work in increments.

True, according to my knowledge after 1 year of research.

In fact, you want a gpu BrainFlayer. Everyone wants it)
Ryan refuses to write it because he is a WhiteHat.

If successful, it will be slower than sequentially calculating points.

##########################
heuristic calculate the hashrate for BrainFlayer cuda/opencl

BrainFlayer cpu sse
0,1 Mk/s - 1core i7-6820
8core - 4core real, 4core hyper-threading, so x6 instead x8
0,6 Mk/s - 8core i7-6820

You'd probably need to perform string mangling on the GPU to keep the cores working as much as possible, because the transfer of each and every candidate phrase from CPU/system memory to the GPU could end up being a severe bottleneck. Perhaps a deliberately slower algorithm like warp may work better, since the GPU will spend most of its time calculating rather than transferring to/from the host... but how common are non SHA256 brainwallets? As you state, any passphrase based search (which is essentially a set of random keys) will be slower than a sequential search.

[Telariust]

What about pollard-rho versions of this people talk in other threads, but there is no publicly available tool to try it (to my knowledge)?

https://bitcointalk.org/index.php?topic=5166284.msg52318676#msg52318676

You'd probably need to perform string mangling on the GPU to keep the cores working as much as possible, because the transfer of each and every candidate phrase from CPU/system memory to the GPU could end up being a severe bottleneck. Perhaps a deliberately slower algorithm like warp may work better, since the GPU will spend most of its time calculating rather than transferring to/from the host... but how common are non SHA256 brainwallets? As you state, any passphrase based search (which is essentially a set of random keys) will be slower than a sequential search.

I also thought so for a long time.
not really if the size does not exceed 49152

Code:

  size_t stackSize = 49152;
  err = cudaDeviceSetLimit(cudaLimitStackSize, stackSize);

..For VanitySearch, having a smaller group size is better (This is a reason why I worked a lot on this DRS62 ModInv implementation). I can double the size of the group (I will definitely do it) but not more. The GPU kernel performs one group per thread and send back hash160 to the CPU. If the group size is too large, memory transfer and allocation become a problem. Divide and rule

VanitySearch restarts the kernel about 1000 times per second (!!!), and it works fine.
opencl unknow

..and this is one of the main technical reasons why gpu BF do not.
cpu BF checks huge dictionaries in a few hours. loading such volumes into gpu is problematic.
therefore gpu BF should work using the built-in generator, e.g. brute force seed.

[Serj88]

How to prevent brute-forcing private keys?

[Archi606]

C:\Users\Mashine-5\source\repos\BitCrack2\x64\Debug>cuBitCrack.exe -i key.txt -o key1.txt
[2019-09-15.17:02:27] [Info] Compression: compressed
[2019-09-15.17:02:27] [Info] Starting at: 0000000000000000000000000000000000000000000000000000000000000001
[2019-09-15.17:02:27] [Info] Ending at:   FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFEBAAEDCE6AF48A03BBFD25E8CD0364140
[2019-09-15.17:02:27] [Info] Counting by: 0000000000000000000000000000000000000000000000000000000000000001
[2019-09-15.17:02:27] [Info] Initializing GeForce GTX 1060 6GB
[2019-09-15.17:02:27] [Info] Generating 262,144 starting points (10.0MB)
[2019-09-15.17:02:30] [Info] 10.0%
[2019-09-15.17:02:32] [Info] 20.0%
[2019-09-15.17:02:35] [Info] 30.0%
[2019-09-15.17:02:37] [Info] 40.0%
[2019-09-15.17:02:39] [Info] 50.0%
[2019-09-15.17:02:41] [Info] 60.0%
[2019-09-15.17:02:43] [Info] 70.0%
[2019-09-15.17:02:46] [Info] 80.0%
[2019-09-15.17:02:48] [Info] 90.0%
[2019-09-15.17:02:50] [Info] 100.0%
[2019-09-15.17:02:50] [Info] Done
[2019-09-15.17:02:50] [Info] Loading addresses from 'key.txt'
[2019-09-15.17:02:50] [Info] 1 addresses loaded (0.0MB)
GeForce GTX 1060 1083 / 6144MB | 1 target 1.25 MKey/s (11,796,480 total) [00:00:07]




Hello!

Why such a low rate?
There is a detailed setting, please tell me)
Maybe something is installed incorrectly

[thunderclouds]

Hi all, im currently getting only around 3 MKey/s using a GTX 1080 8GB, ive seen others getting 300 MKey/s, i have tried the card on two different PC's with a fresh install(win10), yet both give the same rate.

This is my config,

Code:

-b 36 -t 128 -p 800

I have tried to play with the numbers but crash the card if i go too high, could anyone suggest a soloution.

[iparktur]

I have a GPU brainflayer, 1 billion hashes /sec on 8 x RX570

You can send me PM if interested
[/quote]

bartekjagoda  

How to send you PM if you have "closed"?

[RPL_Ryan]

Is anyone else having issues running multiple instances?  I have a 7 card gpu rig, if initiate 2+ instances I get a computer crash. Windows 10, cl initiate.

[balskiy]

Hi, how to use cuBitCrack.exe on multiple gpu?

[iparktur]

Probably at I have very weak laptop there is no large difference on speed from change of parameters - t - p (in clBitCrack.exe)
The quantity Generating starting points depends on parameters - t - p
Than meanings of parameters - t - p by that more will be more generated Generating starting points
I think, probably, than will be more generated Generating starting points - the better?
Who knows precisely - explain.

[balskiy]

Probably at I have very weak laptop there is no large difference on speed from change of parameters - t - p (in clBitCrack.exe)
The quantity Generating starting points depends on parameters - t - p
Than meanings of parameters - t - p by that more will be more generated Generating starting points
I think, probably, than will be more generated Generating starting points - the better?
Who knows precisely - explain.

Hi, read the link https://bitcointalk.org/index.php?topic=5166284.msg52255756#msg52255756

[bartekjagoda]

I have a GPU brainflayer, 1 billion hashes /sec on 8 x RX570

You can send me PM if interested

bartekjagoda  

How to send you PM if you have "closed"?
[/quote]

I have opened my box for newbies
I usually get 160-220Mkeys per second depending on the size of addresses to check. Has been doing 100lists

[iparktur]

Probably at I have very weak laptop there is no large difference on speed from change of parameters - t - p (in clBitCrack.exe)
The quantity Generating starting points depends on parameters - t - p
Than meanings of parameters - t - p by that more will be more generated Generating starting points
I think, probably, than will be more generated Generating starting points - the better?
Who knows precisely - explain.

Hi, read the link https://bitcointalk.org/index.php?topic=5166284.msg52255756#msg52255756

I so have understood, that cuBitCrack and CUDA for processors Intel witch a video cards NVIDA
Mine laptop - AMD A6-3420M APU with AMD Radeon HD 6520G
Therefore I use OpenCL and clBitCrack
Unless cuBitCrack and CUDA will work with such configuration?

[balskiy]

Probably at I have very weak laptop there is no large difference on speed from change of parameters - t - p (in clBitCrack.exe)
The quantity Generating starting points depends on parameters - t - p
Than meanings of parameters - t - p by that more will be more generated Generating starting points
I think, probably, than will be more generated Generating starting points - the better?
Who knows precisely - explain.

Hi, read the link https://bitcointalk.org/index.php?topic=5166284.msg52255756#msg52255756

I so have understood, that cuBitCrack and CUDA for processors Intel witch a video cards NVIDA
Mine laptop - AMD A6-3420M APU with AMD Radeon HD 6520G
Therefore I use OpenCL and clBitCrack
Unless cuBitCrack and CUDA will work with such configuration?

You need clBitCrack, but the setup principle is the same as for cuBitCrack.

[iparktur]

I have a GPU brainflayer, 1 billion hashes /sec on 8 x RX570

You can send me PM if interested

bartekjagoda  

How to send you PM if you have "closed"?

I have opened my box for newbies
I usually get 160-220Mkeys per second depending on the size of addresses to check. Has been doing 100lists
[/quote]

I send PM to you

[bartekjagoda]

Hi all, im currently getting only around 3 MKey/s using a GTX 1080 8GB, ive seen others getting 300 MKey/s, i have tried the card on two different PC's with a fresh install(win10), yet both give the same rate.

This is my config,

Code:

-b 36 -t 128 -p 800

I have tried to play with the numbers but crash the card if i go too high, could anyone suggest a soloution.

Your CUDA drivers might be wrong, which version do you use?

[almightyruler]

Hey guys, newbie here glad I found this thread. I'm not a programmer or know how to brute force these bitcoin wallets.

Iv'e been working on a new way to store your private keys privately and securely out in the open. Essentially if somebody can have access to your private key, They will be unable to unlock the wallet.

In this image I put a private key that is missing 13 characters. If you can guess the password, you can have access to decrypting the characters needed to unlock the private key.

If anyone can crack this private key, they can have access to some Bitcoin I sent to it.

I'm trying to understand why you would want to store a private key "out in the open"? With 13 characters to brute force, the key is still going to be 'difficult' to break, but certainly a lot easier than if the entire key was kept private. Since the password appears to map letters and numbers to the missing base58 parts of the key, based on statistical analysis of language, the search space could be narrowed down considerably.

Can you reveal the address and sign a message to prove this is a real challenge?

[almightyruler]

Sure, the Bitcion Address is 1C2SvZUDaSr6zzTTZr7vNeCo5jbz1bwiGe


I will send 0.00007185 BTC in the next couple of minutes to prove I signed the transaction.

Yeah, but again: why?

Is this simply a fun challenge, or are you planning to use (and promote) this method of storing private keys?

As has been proven in the puzzle challenge thread, brute forcing ~76 bits (in this case) of key is not impossible.

[nc50lc]

In this image I put a private key that is missing 13 characters. If you can guess the password, you can have access to decrypting the characters needed to unlock the private key.

IDK what you're making but it's so wrong in so many levels.

First, it's missing 14 characters (not 13)  
Next, looks like the private key is in compressed WIF format which should be in base58 encoding,
means that there shouldn't be any zero, capital "o", lower-case "L" and capital "i" in the list; but there's an "O" and a "0" in the table.

Lastly, why all caps? Is your 1C2SvZUDaSr6zzTTZr7vNeCo5jbz1bwiGe address' WIF private key all-caps?

[Breadyfreddy]

Im planning on implementing this in my notebook that is launching this year. A notebook to store private keys. Shieldfolio.com

The goal is to help users store private keys privately in case somebody stumbles upon these private keys.

[almightyruler]

Im planning on implementing this in my notebook that is launching this year. A notebook to store private keys. Shieldfolio.com

The goal is to help users store private keys privately in case somebody stumbles upon these private keys.

Make a secret less secret by default just in case someone stumbles upon that secret?

How does revealing part of a private key make it more private?