BitCrack - A tool for brute-forcing private keys

[TrackMaD]

Need Advice on a Risk Management Question ?
So when we eventually  find the Private Key using BitCrack for any of these Bitcoin challenge transactions,
What will be the best/safest method to import the private key into your own wallet in the quickest possible time ?
eg : To an Exodus Digital Hot wallet or to a cold storage offline wallet or bitcoin core address

I heard that its possible to use transaction fees to redirect an import of private keys to another wallet
since there will be so many hackers sniffing for transactions/keys from these wallets trying to intercept transfers.
Is this possible or highly unlikely ?

Thanks 

http://www.sensoria.org.uk/wp-content/uploads/2019/07/the-adventures-of-robin-hood-1200-1200-675-675-crop-000000.jpg

[NotATether]

Need Advice on a Risk Management Question ?
So when we eventually  find the Private Key using BitCrack for any of these Bitcoin challenge transactions,
What will be the best/safest method to import the private key into your own wallet in the quickest possible time ?
eg : To an Exodus Digital Hot wallet or to a cold storage offline wallet or bitcoin core address

I heard that its possible to use transaction fees to redirect an import of private keys to another wallet
since there will be so many hackers sniffing for transactions/keys from these wallets trying to intercept transfers.
Is this possible or highly unlikely ?

Thanks 

You just have to sweep the private key into a mixer (such as Chipmixer) and then import the resulting private keys it gives you into an empty Electrum or Bitcoin Core wallet. A good mixer offers complete privacy from sniffers.

Never ever store large sums of bitcoin in a hot wallet. It's better in that case to send them to an offline wallet on an airgapped computer.

[WanderingPhilospher]

Need Advice on a Risk Management Question ?
So when we eventually  find the Private Key using BitCrack for any of these Bitcoin challenge transactions,
What will be the best/safest method to import the private key into your own wallet in the quickest possible time ?
eg : To an Exodus Digital Hot wallet or to a cold storage offline wallet or bitcoin core address

I heard that its possible to use transaction fees to redirect an import of private keys to another wallet
since there will be so many hackers sniffing for transactions/keys from these wallets trying to intercept transfers.
Is this possible or highly unlikely ?

Thanks 

You just have to sweep the private key into a mixer (such as Chipmixer) and then import the resulting private keys it gives you into an empty Electrum or Bitcoin Core wallet. A good mixer offers complete privacy from sniffers.

Never ever store large sums of bitcoin in a hot wallet. It's better in that case to send them to an offline wallet on an airgapped computer.

So by using Chipmixer, does it change the private key instantly? The fear/question has always been, if someone finds the key to #64 and moves it, someone can use the kangaroo and solve it within seconds and gain the private key and give higher trans fee and move it to their own wallet. Make sense?

[Noname400]

Need Advice on a Risk Management Question ?
So when we eventually  find the Private Key using BitCrack for any of these Bitcoin challenge transactions,
What will be the best/safest method to import the private key into your own wallet in the quickest possible time ?
eg : To an Exodus Digital Hot wallet or to a cold storage offline wallet or bitcoin core address

I heard that its possible to use transaction fees to redirect an import of private keys to another wallet
since there will be so many hackers sniffing for transactions/keys from these wallets trying to intercept transfers.
Is this possible or highly unlikely ?

Thanks 

http://www.sensoria.org.uk/wp-content/uploads/2019/07/the-adventures-of-robin-hood-1200-1200-675-675-crop-000000.jpg

who's stopping you from giving a higher commission? and don't worry about hackers.

[NotATether]

So by using Chipmixer, does it change the private key instantly?

No. The way it works is that you send all the balance to the deposit address, thereby emptying the private key. Of course, you should use a high transaction fee to reduce the chance that somebody can start and finish Kangaroo before the transaction is confirmed.

Once the transaction is confirmed then Chipmixer is holding the money at that point. It cannot be taken by a hacker. You then follow the instructions on their site to split up the balance into multiple outputs of different sizes such as 0.001, 0.002, 0.004, etc.

Then you let it do it's mixing activity (which the hackers can't track) and it will send all of the outputs to a private key they created. They will then give you the private keys to import into a wallet. If you don't trust them though you can always move the coins to a second address you have created, and that's private because hackers don't know the private keys you imported in the first place.

[zahid888]

The fear/question has always been, if someone finds the key to #64 and moves it, someone can use the kangaroo and solve it within seconds and gain the private key and give higher trans fee and move it to their own wallet. Make sense?

Better if we do it practically with some tinny amount of bitcoin..

[MeBender]

hey so I decided to upgrade BitCrack to CUDA 11.2 to see if it would improve my performance but nope, originally I was getting 216Mkey/s on my 1070 but after compiling for CUDA 11.2 it was drastically lower 4.2Mkey/s lol

Here's the forked code upgraded to CUDA 11.2 if anyone wants it: https://github.com/GonzoTheDev/BitCrack

[NotATether]

hey so I decided to upgrade BitCrack to CUDA 11.2 to see if it would improve my performance but nope, originally I was getting 216Mkey/s on my 1070 but after compiling for CUDA 11.2 it was drastically lower 4.2Mkey/s lol

Here's the forked code upgraded to CUDA 11.2 if anyone wants it: https://github.com/GonzoTheDev/BitCrack

What shocks me is that you didn't even edit the CUDA code. Apparently just upgrading the CUDA toolkit is enough to cause performance to drop 

My theory is that CUDA 11 has loosened memory alignment restrictions which causes GPUs running compute capability 6.x and lower to have crippled speeds while they align memory at runtime, however anything running at compute cap 7.x and later has a hard failure on unaligned memory (Presumably they removed those checks to optimize speed).

I'm having a hard time believing that physical GPU hardware is the problem because CUDA abstracts it all away, and besides, sometimes, replacing the OS, entire hardware, and drivers completely fixes the problem on the RTX cards (!!!).

I believe it has something to do with this:

In addition to the enhanced range of precisions, restrictions on matrix dimensions and alignment for Tensor Core acceleration have been removed. For appropriate precisions, the acceleration is now automatic, requiring no user opt-in. The heuristics for cuBLAS automatically adapt to resources when running on the GPU instances with MIG on A100.

Bearing in mind that tensor cores are just special FPU cores and that NVIDIA has to cut down on the number of arithmetic cores to make room for them, *maybe* this is where all the chaos is coming from.

[rama_rao]

Hey guys what is the ideal -b -t -p settings for 1080ti and v100 gpus?

They should be better documented or selected automatically like vanity search.

Also I was getting 1000mk/s on 1080ti on vanity search but only 300mk/s on bitcrack. What is this that slow? As far as I know vanity search only have to compare first 4-6 characters and this has to compare whole address. Is that the only reason for this slowness or something else in the implementation?

BTW, kinda offtopic. My python implementation (<50 lines) using external libs is doing only 100 keys/sec (on single cpu core) and also I noticed official c/python implementations also do around 1000 keys/sec while generating. What makes these generators a thousand/million times faster? Is it just a simple trick or years worth of dev? Give some pointers maybe.

Also while running on windows, it is not showing any cpu/gpu/memory. Why is that?

[NotATether]

Also I was getting 1000mk/s on 1080ti on vanity search but only 300mk/s on bitcrack. What is this that slow? As far as I know vanity search only have to compare first 4-6 characters and this has to compare whole address. Is that the only reason for this slowness or something else in the implementation?

I can't help you choose parameters because I don't have any of those, still trying to find a way to lease a GPU for bitcoin (turns out it's very hard to find someone who's not trying to sell you a miner  ). But the reason why VanitySearch is much faster than this is not related to the number of characters they are comparing, it's due to the fact that Jean_Luc's code generates a "jump table" of secp256k1 points (see e.g. GPU/Group.h) so he avoids actually doing the elliptic curve math at runtime, unlike brichard's code. This is despite both their engines being written in CUDA.

In VanitySearch he also writes all his math in assembly code while Bitcrack uses C for its math functions.

Several pages back somebody posted here that they used -b 112 -t 512 -p 512 for their 1080 and got similar speeds as you.

BTW, kinda offtopic. My python implementation (<50 lines) using external libs is doing only 100 keys/sec (on single cpu core) and also I noticed official c/python implementations also do around 1000 keys/sec while generating. What makes these generators a thousand/million times faster? Is it just a simple trick or years worth of dev? Give some pointers maybe.

Are your python files compiled? Try doing that so that your code doesn't have to be interpreted at runtime, using python -m compileall YOUR_PY_FILES.

Also while running on windows, it is not showing any cpu/gpu/memory. Why is that?

What do you mean? Bitcrack never printed how much CPU or memory it uses in the console. Do you mean in Task Manager? If so then you'll usually find it as a child process of Windows Command Processor (and Task Manager is going to show that using a lot of resources but it's actually the child process running inside it using all that).

[rama_rao]

Yes, I've used those values. What ever tweaks I make is only decreasing the rate. So, asked to know if there is any good doc for that.

Yeah, I was asking about task manager. Everything is normal, cpu 20%, gpu 5% usage and 6gb gpu ram which is good I guess.

What do you mean by lease? You can rent on AWS or some other services cheaper. Try google colab which has free gpus (t4 gpus I guess) but stops every 12 hrs and can run only 1 instance per account.

BTW, any plans for these softwares to be written for ASICs to make them more faster?

[NotATether]

What do you mean by lease? You can rent on AWS or some other services cheaper. Try google colab which has free gpus (t4 gpus I guess) but stops every 12 hrs and can run only 1 instance per account.

BTW, any plans for these softwares to be written for ASICs to make them more faster?

I can only pay for hosting with bitcoin which AWS does not support, and I use Google Colab is, well, awkward  to make it work for non-Python projects like Bitcrack.

ASICs that are used in bitcoin miners cannot be programmed to run private key crackers because the mining logic is hard coded into the ICs (integrated circuits) inside the ASICs. So you actually have to make a custom ASIC that only makes a table of all RIPEMD160 hashes and searches it for a particular address in order to make it work with a cracker like Bitcrack.

[jlfv777]

I want the steps on a video with explaining in order to know how can i run it i am not a programmer to understand commands and visual studio etc. even the screenshot on the first page i didn't understand anything from it.

It's not as complicated as you think and you don't have to be a programmer to be able to run the BitCrack. Basically what you need to do are :
1. Download BitCrack from https://github.com/brichard19/BitCrack/releases. If you're not sure, just download clBitCrack.exe
2. Open application called "Command Prompt" and move to directory path where clBitCrack.exe is located
3. Type clBitCrack.exe on "Command Prompt"

If you never heard or don't know how to use "Command Prompt", please read this guide https://techtipvault.com/windows/command-prompt-windows-10-cmd-guide/

P.S. i can't make screenshot since i don't use Windows OS

I didn't find clBitCrack.exe

Command Prompt said clBitCrack.exe is not recognized as an internal or external command, operable program, or batch file.

If you are running Nvidia cards, use this link to download cuBitCrack.exe


If you are using AMD cards, use this link to download clBitCrack.exe


Create a folder and place one of those .exe files in the folder, along with your input file that contains the addresses you want to search for.
So let's say you create a folder called "BitCrack"; inside there, place one of the .exe files and the input file containing addresses to search for.

Now, create a batch file. Create a new text document and add the commands you want the program to run with. For example:

Code:

cuBitCrack -d 0 -b 64 -t 256 -p 512 --keyspace 8000000000000000:FFFFFFFFFFFFFFFF -i addresses.txt -o FOUNDKEY.txt

Now save that text document as a .bat file.  When you go to save the text doc as bat file, make sure you select the "Save as type:" box as All Files and then save (for example) runbitcrack.bat....it's important to add the .bat extension or you are just saving as a text document and the program will not run.

Once that is done, double click on your new .bat file and the program will run. It may say you need another file such as vcc....; if so, google that file, download it, and place it in the same folder you created (with all these files in it).

Breakdown of .bat file commands:
cuBitCrack - this is the .exe file that you downloaded in first step. It'll either be cuBitCrack or clBitCrack, depending on your video card.
-d 0 - this is telling the program to use the video card numbered 0; if you have more than one card, you can change this to which video card you want to use
-b 64 -t 256 -p 512 - tells program what gridsize you want to use. you will have to experiment with these numbers to fine tune your card but these default numbers will get you started
--keyspace 8000000000000000:FFFFFFFFFFFFFFFF - this is the keyspace that the program will search in, change it to search the space you want to search
-i addresses.txt - the -i tells program this is your input file; this is where you save the btc addresses you want to search for
-o FOUNDKEY.txt - this -o tells the program you want to save your found keys, in this example, if you find a key, it will save the key in a text file called "FOUNDKEY" . It will save it in the same folder you are running the program from.

edit: the above is for Windows; if you are using Linux/Ubuntu, someone else will have to help you.

Gets to the load address files and dies

Unable to open 'address.txt'
Can you help me out pls.

[WanderingPhilospher]

Unable to open 'address.txt'
Can you help me out pls.

if you are using exact commands from above, the address list was titled "-i addresses.txt" (addresses, not address) but whatever the file you have your addresses in, that's what you need to specify using the -i command. Example, if your addresses are stored in apple.txt document, then you would use
-i apple.txt

[WanderingPhilospher]

Also I was getting 1000mk/s on 1080ti on vanity search but only 300mk/s on bitcrack. What is this that slow? As far as I know vanity search only have to compare first 4-6 characters and this has to compare whole address. Is that the only reason for this slowness or something else in the implementation?

I can't help you choose parameters because I don't have any of those, still trying to find a way to lease a GPU for bitcoin (turns out it's very hard to find someone who's not trying to sell you a miner  ). But the reason why VanitySearch is much faster than this is not related to the number of characters they are comparing, it's due to the fact that Jean_Luc's code generates a "jump table" of secp256k1 points (see e.g. GPU/Group.h) so he avoids actually doing the elliptic curve math at runtime, unlike brichard's code. This is despite both their engines being written in CUDA.

In VanitySearch he also writes all his math in assembly code while Bitcrack uses C for its math functions.

Several pages back somebody posted here that they used -b 112 -t 512 -p 512 for their 1080 and got similar speeds as you.

BTW, kinda offtopic. My python implementation (<50 lines) using external libs is doing only 100 keys/sec (on single cpu core) and also I noticed official c/python implementations also do around 1000 keys/sec while generating. What makes these generators a thousand/million times faster? Is it just a simple trick or years worth of dev? Give some pointers maybe.

Are your python files compiled? Try doing that so that your code doesn't have to be interpreted at runtime, using python -m compileall YOUR_PY_FILES.

Also while running on windows, it is not showing any cpu/gpu/memory. Why is that?

What do you mean? Bitcrack never printed how much CPU or memory it uses in the console. Do you mean in Task Manager? If so then you'll usually find it as a child process of Windows Command Processor (and Task Manager is going to show that using a lot of resources but it's actually the child process running inside it using all that).

You also have to realize, VanitySearch is searching more than just Point. It is searching Point +  endo1 + endo2 + symmetry ( pt.y.ModNeg; p1.y.ModNeg; and p2.y.ModNeg; It's a different animal than Bitcrack, hence what seems like super speed compared to Bitcrack. Just different designs and purposes. One is point by point the other is 6 points.

[rama_rao]

Also I was getting 1000mk/s on 1080ti on vanity search but only 300mk/s on bitcrack. What is this that slow? As far as I know vanity search only have to compare first 4-6 characters and this has to compare whole address. Is that the only reason for this slowness or something else in the implementation?

I can't help you choose parameters because I don't have any of those, still trying to find a way to lease a GPU for bitcoin (turns out it's very hard to find someone who's not trying to sell you a miner  ). But the reason why VanitySearch is much faster than this is not related to the number of characters they are comparing, it's due to the fact that Jean_Luc's code generates a "jump table" of secp256k1 points (see e.g. GPU/Group.h) so he avoids actually doing the elliptic curve math at runtime, unlike brichard's code. This is despite both their engines being written in CUDA.

In VanitySearch he also writes all his math in assembly code while Bitcrack uses C for its math functions.

Several pages back somebody posted here that they used -b 112 -t 512 -p 512 for their 1080 and got similar speeds as you.

BTW, kinda offtopic. My python implementation (<50 lines) using external libs is doing only 100 keys/sec (on single cpu core) and also I noticed official c/python implementations also do around 1000 keys/sec while generating. What makes these generators a thousand/million times faster? Is it just a simple trick or years worth of dev? Give some pointers maybe.

Are your python files compiled? Try doing that so that your code doesn't have to be interpreted at runtime, using python -m compileall YOUR_PY_FILES.

Also while running on windows, it is not showing any cpu/gpu/memory. Why is that?

What do you mean? Bitcrack never printed how much CPU or memory it uses in the console. Do you mean in Task Manager? If so then you'll usually find it as a child process of Windows Command Processor (and Task Manager is going to show that using a lot of resources but it's actually the child process running inside it using all that).

You also have to realize, VanitySearch is searching more than just Point. It is searching Point +  endo1 + endo2 + symmetry ( pt.y.ModNeg; p1.y.ModNeg; and p2.y.ModNeg; It's a different animal than Bitcrack, hence what seems like super speed compared to Bitcrack. Just different designs and purposes. One is point by point the other is 6 points.

Ohh I see. So, let's say we want to search 64 bit space, we can do that with bitcrack as usual. But with vanitysearch it doesn't make sense as it searches 6 points at a time and most probably 5 of them are out of that 64 bit space, right? Makes sense. Thanks for explanation.

Also, what makes these so fast than official implementation? It is kinda mind boggling to think someone can do 1000/10k times better than usual implementation. (Right from vanitygen, they all have very amazing perf.) Any links to docs or papers is fine. Thanks in advance.

[WanderingPhilospher]

Ohh I see. So, let's say we want to search 64 bit space, we can do that with bitcrack as usual. But with vanitysearch it doesn't make sense as it searches 6 points at a time and most probably 5 of them are out of that 64 bit space, right? Makes sense. Thanks for explanation.

Also, what makes these so fast than official implementation? It is kinda mind boggling to think someone can do 1000/10k times better than usual implementation. (Right from vanitygen, they all have very amazing perf.) Any links to docs or papers is fine. Thanks in advance.

Right, some keys will be out of the 64 bit range with VanitySearch and all keys will be in specified range with BitCrack.

What do you mean when you say "so fast than official implementation?"

[brainless]

one cuda coder bitcoinforktech disapear from 21 jan 2021, anyone know where he go, hope he will be fine, not infected by covid19, if anyone know write here about him for not online from 21 jan

https://bitcointalk.org/index.php?action=profile;u=2863059

thankx God, found he is live and maybe busy in dome project, as his listed bitcoin address in his profile
bc1qthrhmdmfawq6w4lzmjrce3reuxt06pvvnhpkj6

used on 5th feb 2021, receive and send
indicate he is live and getting his money for his work/assignments

[rama_rao]

one cuda coder bitcoinforktech disapear from 21 jan 2021, anyone know where he go, hope he will be fine, not infected by covid19, if anyone know write here about him for not online from 21 jan

https://bitcointalk.org/index.php?action=profile;u=2863059

thankx God, found he is live and maybe busy in dome project, as his listed bitcoin address in his profile
bc1qthrhmdmfawq6w4lzmjrce3reuxt06pvvnhpkj6

used on 5th feb 2021, receive and send
indicate he is live and getting his money for his work/assignments

So, basically, all the remaining puzzle problems become really hard or not rewarding enough. So, most developers are working on some other projects. On kangaroo readme it is mentioned 120 puzzle would take 2 months on 256 v100s to find a solution. I think 64 problem is also similarly hard. So, not much to do here unless the original puzzle maker adds more to the wallets again...

Making outgoing transactions on 70-120 wallets is also one way. But would be cleaned by people who have good experience with running kangaroo and good hardware resources... Until then no progress further. A lotta resources went in 64 and 120, so they might get cracked. After that no hopes...

[jlfv777]

Unable to open 'address.txt'
Can you help me out pls.

if you are using exact commands from above, the address list was titled "-i addresses.txt" (addresses, not address) but whatever the file you have your addresses in, that's what you need to specify using the -i command. Example, if your addresses are stored in apple.txt document, then you would use
-i apple.txt

That was spelling mistake, I do the same
Here's is link image
https://imgur.com/dkx89mc
https://imgur.com/oMf6XCc