BitCrack - A tool for brute-forcing private keys

[NotATether]

But I can't wait to see the speed of your dev fee program. It better be triple the speed!

When the speed bottleneck is in the CUDA brute-force algo implementation, you'd be lucky if you see a 1.5x speed boost just from optimizing software.

[1714765012]

1714765012

[1714765012]

1714765012

[1714765012]

1714765012

[1714765012]

1714765012

[1714765012]

1714765012

[WanderingPhilospher]

But I can't wait to see the speed of your dev fee program. It better be triple the speed!

When the speed bottleneck is in the CUDA brute-force algo implementation, you'd be lucky if you see a 1.5x speed boost just from optimizing software.

Yeah, I don't see sp getting more than 1,500 MKey/s while checking 20 million xpoints at the same time. A true 1,500 MKey/s...not tweaking the code to show more speed versus the actual speed.

[a.a]

Will we also see the sourcecode of those changes?

[NotATether]

Will we also see the sourcecode of those changes?

sp_ doesn't share the source code for his tools. Just see the last dozen or two pages before this one for another instance of that.

[chrysophylax]

Will we also see the sourcecode of those changes?

sp_ doesn't share the source code for his tools. Just see the last dozen or two pages before this one for another instance of that.

Well ...

He has been doing this for MANY years now, and manages to squeeze BTC out of almost everyone that messes with his 'private' binaries that are completely based on OpenSource Products.

If you go WAY back in his own thread, you will find multiple instances of him and Myself ripping into each other quite aggressively over such things. He NEVER plays fair, nor does he ever give a single shit about the people or uses of the software. Be VERY wary of him. Rune is known to be a ruthless purveyor tweaked software, not original coded software.

I learned this first hand in the very beginning when I helped with the testing when CWI had the largest GPU Farm in the Southern Hemisphere at the time.

Be VERY careful of his Malicious ways.

1% doesn't sound like much, but that was enough for him BACK THEN to make him a very rich man by tweaking other peoples Code (Namely Alexis78 - who was working with us at CWI at the time), compiling that OpenSource product after a tweak (usually a Hard Coded Intensity Setting) and distributing it as a 'Private Miner' based on Windows which he claimed was his.

As for BitCrack - this work will continue ... and it is good work so far!

#crysx

[COBRAS]

I have released a fast miner for Ethereum, Ethereum Classic, zilliqa and vertcoin.
I plan to add a fast bitcrack implementation with multigpu support, working on the latest rtx cards.
A Bitcrack implementation for ethereum is also on the drawing board.

I will take 1% commison.
Mine fee in zilliqa. 1 minute every 90 minutes.

https://bitcointalk.org/index.php?topic=5358334.0

Add option for wallet.dat password reminder !!! This is more profitable task then bitcrack.

[Andre_25]

https://github.com/richieburns/BitCrack_AMP/releases/download/v1.0.0_AMP/BitCrack_AMP.zip

Bitcrack CUDA 11.2 Compute 8.6 Win10

Would appreciate some benchmarks for RTX 30 series cards.....

Cheers

Hi, friend @richieburns
thanks for uploading the .exe file.
Could you also publish the source code of that version of cuda 11.2? I mention this since the source code that you have published, I am seeing that it still associates with cuda 10.2

Thanks....

I take advantage and ask the rest of the community.
is there any version of bitcrack, published the source code, that is under cuda 11.2 or 11.4?

thanks very much

[NotATether]

I take advantage and ask the rest of the community.
is there any version of bitcrack, published the source code, that is under cuda 11.2 or 11.4?

I don't think you will find any Bitcrack impl that is linked to CUDA 11.4 because it is so recent.

11.2 is probably more likely, but I don't know of any others other than the one you linked in your post.

[a.a]

Does it even has a performance impact? Or why is it relevant? Or is it blocking you on using latest cards?

Anyhow updating CUDA is not hard. In my OpenCL Fork I actually upgraded it to Cuda 11.3 before ripping out the cuda code.
I think you only have to modify the BitCrack.props file to upgrade the cuda version.

https://github.com/Uzlopak/BitCrackOpenCL/blob/master/BitCrack.props

[Andre_25]

Does it even has a performance impact? Or why is it relevant? Or is it blocking you on using latest cards?

Anyhow updating CUDA is not hard. In my OpenCL Fork I actually upgraded it to Cuda 11.3 before ripping out the cuda code.
I think you only have to modify the BitCrack.props file to upgrade the cuda version.

https://github.com/Uzlopak/BitCrackOpenCL/blob/master/BitCrack.props

friend, thanks for replying.
You saved me from giving up
I had made changes to multiple files with the vxproj extension, which also contained the name or path to the Cuda version, but I had not edited that specific file (.props) and now that I did it, it worked.

Thank you very much.

[Andre_25]

Hi guys, me again

Guys, even though I was able to compile BitCrack, using Cuda 11.2, I can't get it to work for me, because if I put more than 16 addresses, it tells me "Invalid Argument",

Has someone happened and corrected it?

Thanks in what can help me

[Angelo1710]

pika's random mode (which is not in brichard19's bitcrack) is not using a random number generator, it turns out that it uses an iteration count and multiplies it with all the other normal stuff, which is then ADDED to some exponent to obtaining private key samples:

Code:

// https://github.com/ZenulAbidin/BitCrack-3000/blob/master/CudaKeySearchDevice/CudaKeySearchDevice.cpp line 112
    // Generate key pairs for k, k+1, k+2 ... k + <total points in parallel - 1>
    secp256k1::uint256 privKey = _startExponent;

    if(!_randomMode) {
        exponents.push_back(privKey);
    }
// ...
// Line 272
        if(!_randomMode) {
            offset = (secp256k1::uint256((uint64_t)_blocks * _threads * _pointsPerThread * _iterations) + privateKeyOffset) * _stride;
            privateKey = secp256k1::addModN(_startExponent, offset);
        } else {
            offset = secp256k1::uint256(_iterations) * _stride;
            privateKey = exponents[privateKeyOffset];
            privateKey = secp256k1::addModN(privateKey, offset);
        }

Can you please explain a bit more how's pika's random mode is working? I though it was using RNG. Can you explain how it gets starting points compared to original bitcrack? Tnx

[tamarind3]

...
Can you please explain a bit more how's pika's random mode is working? I though it was using RNG. Can you explain how it gets starting points compared to original bitcrack? Tnx

Don't worry. It is using a RNG. NotATether did not fully understand the sourcecode and jumped to quick conclusions.

Code:

void CudaKeySearchDevice::generateStartingPoints()
{
    ...
    if(!_randomMode) {
        exponents.push_back(privKey);
    }

    for(uint64_t i = !_randomMode ? 1 : 0; i < totalPoints; i++) {

        if(_randomMode) {
             privKey = secp256k1::getRandomRange(_startExponent, _end);
        } else {
            privKey = privKey.add(_stride);
        }
        ...
    }
    ...
}

[Angelo1710]

Don't worry. It is using a RNG. NotATether did not fully understand the sourcecode and jumped to quick conclusions.

Code:

void CudaKeySearchDevice::generateStartingPoints()
{
    ...
    if(!_randomMode) {
        exponents.push_back(privKey);
    }

    for(uint64_t i = !_randomMode ? 1 : 0; i < totalPoints; i++) {

        if(_randomMode) {
             privKey = secp256k1::getRandomRange(_startExponent, _end);
        } else {
            privKey = privKey.add(_stride);
        }
        ...
    }
    ...
}

Thanks

[zahid888]

--keyspace FF7B300000000000:FF7B3FFFFFFFFFFF 16jY7qLJnxb7CHZyqBP8qca9d51gAjyXQN

[math09183]

--keyspace FF7B300000000000:FF7B3FFFFFFFFFFF 16jY7qLJnxb7CHZyqBP8qca9d51gAjyXQN

why?

[NotATether]

...
Can you please explain a bit more how's pika's random mode is working? I though it was using RNG. Can you explain how it gets starting points compared to original bitcrack? Tnx

Don't worry. It is using a RNG. NotATether did not fully understand the sourcecode and jumped to quick conclusions.

Yeah on second glance, looks like it is - I guess I never checked secp256k1lib/secp256k1.cpp at the time

https://github.com/ZenulAbidin/BitCrack-3000/blob/fdd9640347197ca834ec99de3d26c8ff0887db46/secp256k1lib/secp256k1.cpp#L910

Code:

uint256 secp256k1::getRandomRange(uint256 min, uint256 max)
{
	uint256 result;
	uint256 range = max.sub(min);

	unsigned char targetByteSize = (range.getBitRange() + 31) / 32;

	for(int i = 0; i < 8; i++) {
		if(targetByteSize > i) {
			result.v[i] = rnd.getChunk();
			if(targetByteSize > i && targetByteSize <= i + 1 && range.v[i] != 0 && result.v[i] > range.v[i]) {
				result.v[i] %= range.v[i];
			}
		}
	}
	return result.add(min);
}

[larocfra]

I was able to compile BitcrackEvo with Cuda 11.4 and currently running a Nvidia K80 with 182.64 MKey/s each processor on the card.

Thu Oct 14 19:17:44 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.63.01    Driver Version: 470.63.01    CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+================|
|   0  Tesla K80           On  | 00000000:00:10.0 Off |                 Off* |
| N/A   42C    P0   146W / 149W |  11437MiB / 12206MiB |    100%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
|   1  Tesla K80           On  | 00000000:00:11.0 Off |                 Off* |
| N/A   36C    P0   147W / 149W |  11437MiB / 12206MiB |    100%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|===============================================|
|    0   N/A  N/A     53951      C   ./cuBitCrackEvo                 11432MiB |
|    1   N/A  N/A     54088      C   ./cuBitCrackEvo                 11432MiB |
+-----------------------------------------------------------------------------+

Is there any new software available ??

[NotATether]

I was able to compile BitcrackEvo with Cuda 11.4 and currently running a Nvidia K80 with 182.64 MKey/s each processor on the card.

~
Is there any new software available ??

Regarding the fork you used, from viewing its commit history it looks like the only change other than being updated to support newer toolsets/CUDA/CL versions, is this: https://github.com/BitCrackEvo/BitCrackEvo/commit/c13fe1bd9eca709a1e9847bb18fb2865502e599c

The commit makes Bitcrack stop precalculating the private keys for points at runtime and makes the kernel calculate each point sequentially, in each iteration.

Don't you think this'll make the program slower than the original one? Its README says it's under development (and do not use in production) for a reason.

[fxsniper]

Sorry I am very very busy on day time

How can I calculate maximum bit range for bitcrack can scan with in 24 hour

please advice?

example

24 hour = 24*60=1,440*60=86,400

1 target 3.59 MKey/s (250,347,520 total) [00:01:06]
250,000,000*1,440=
250000000*1440=360,000,000,000
360000000000
log2
38.38920595031593
38 bit


1 target 75.62 MKey/s (967,680,000 total) [00:00:11]
900000000*8640=7,776,000,000,000
log2 = 7776000000000
42.82216535759204
42bit