Bitcoin puzzle transaction ~32 BTC prize to who solves it

[Bram24732]

People should be aware that AI can only answer things based on existing public information pulled off from the internet. When it comes to crypto and the issues surrounding this puzzle, and any available tools, AI prompts will produce 99% garbage, because it lacks real human technical expertise, the kind that isn't noob StackOverflow answers on how to draw animated monkeys via CSS, which is what AI is really good at responding, ok? Make it count straight up 1 to 100, no skips, no stops, no errors? Nope, total failure. How can you imagine it will ever be able to come up with low-level code that computes 8.1 billion H160 per second on a 4090, if such knowledge only lives in the human brain (and in some unreadable code on somebody's machine, written by a human), and not in the public data that trained an LLM to recognize cats in YouTube videos, and now pretends to make guarantees about answering simple EC questions in a very wrong way?

This explains the continued BS over the subject.

This is true, but not completely. It really depends on the operator, how the requests are phrased, and how much time one is willing to invest. After several weeks of testing, rewriting code, and changing logic, I managed to produce a solid tool with which I’m scanning 135 using CUDA. Without AI, it would have taken me months, and I might never have achieved these results. It’s been running for a few days now—let’s see what it comes up with :-)

Did you check the probabilty curve of 135 ? Asking because unlike 71, it’s not really a luck thing. Until you’ve generated a stupid amount of points, your odds of hitting are very low (lower than 71)
Might be interesting to calculate after how many days of Bruteforce does 135 get better odds than 71.

[kTimesG]

After several weeks of testing, rewriting code, and changing logic, I managed to produce a solid tool with which I’m scanning 135 using CUDA. Without AI, it would have taken me months, and I might never have achieved these results. It’s been running for a few days now—let’s see what it comes up with :-)

We probably have different definitions for "solid tool"

I guess it helped you isolate the DP matching from the distributed processing, and build the entire orchestration needed to scale up the search.

I'm using at least 10 different technologies and languages just to get to that point, over the last years... and I'm still not calling it solid.

[Niekko]

Did you check the probabilty curve of 135 ? Asking because unlike 71, it’s not really a luck thing. Until you’ve generated a stupid amount of points, your odds of hitting are very low (lower than 71)
Might be interesting to calculate after how many days of Bruteforce does 135 get better odds than 71.

We probably have different definitions for "solid tool"

I guess it helped you isolate the DP matching from the distributed processing, and build the entire orchestration needed to scale up the search.

I'm using at least 10 different technologies and languages just to get to that point, over the last years... and I'm still not calling it solid.

Fair enough, we definitely have different definitions of what "solid" means. I’m not saying it’s a system that "wins", only that at this stage it’s stable enough to run continuously and produce consistent results.

Regarding 135, I’m aware of the probability curve, and that’s exactly why I’m not treating it as a pure luck-based brute force like 71. The approach I’m using combines BSGS-style ideas with MITM techniques and ECDSA arithmetic, which changes the search dynamics compared to a classical enumeration approach. It’s been running for a few days now, so it’s still early, but the goal is to reduce reliance on randomness, without simply increasing computational power.

[parcok]

That looks cool, could you make the code available? There are only photos on GitHub.

I only have a low-spec GPU to help me find the key to puzzle #71. I divided the range 40000000000000000:7ffffffffffffffffff
which will produce the data to be tried: 1,180,591,620,717,411,303,424.
Then I divided it into 4,398,046,511,104 for 1 running batch, resulting in 270,000,000 running batches. Then I divided 270,000,000 into fbatches, where 1 fbatch contains 2,000,000 running batches. so that it produces 135 fbatch.. storing as many as 270,000,000 running batches with 135 fbatch groupings in the sql server database, and I created a search program that directly updates the status of each running batch being searched, so I can freely do leapfrog searches wherever I want, and I don't lose the log of my search results.. because everything is automatically updated to the database that I have.
If you are interested, please check out my project in the following github repository: https://github.com/bgarz929/btcpuzzle71

Thank you for the compliment, sorry previously I have not completed the details in my github repository regarding pool access and how to use it, please check the repo again, I have completed it

[parcok]

If you are interested, please check out my project in the following github repository: https://github.com/bgarz929/btcpuzzle71

Fascinating project. The screenshots look great, but I assume the code lives in a parallel universe where Puzzle 71 is solvable. Any plans to merge that universe with this repo? 

wow your assumption is too wild, if you assume the code is in a parallel universe where Puzzle 71 can be solved, then I try to bring it to this universe, please check my github repo again, yesterday I forgot to complete the details.. repo https://github.com/bgarz929/btcpuzzle71

[crytoestudo]

I thought you were going to provide the code for how you created it. I would like to read the source code, if possible, to understand the code in more detail. Is it possible to make it available? Because there's no way to analyze an already compiled file (which in your case is: xiebo).

That looks cool, could you make the code available? There are only photos on GitHub.

I only have a low-spec GPU to help me find the key to puzzle #71. I divided the range 40000000000000000:7ffffffffffffffffff
which will produce the data to be tried: 1,180,591,620,717,411,303,424.
Then I divided it into 4,398,046,511,104 for 1 running batch, resulting in 270,000,000 running batches. Then I divided 270,000,000 into fbatches, where 1 fbatch contains 2,000,000 running batches. so that it produces 135 fbatch.. storing as many as 270,000,000 running batches with 135 fbatch groupings in the sql server database, and I created a search program that directly updates the status of each running batch being searched, so I can freely do leapfrog searches wherever I want, and I don't lose the log of my search results.. because everything is automatically updated to the database that I have.
If you are interested, please check out my project in the following github repository: https://github.com/bgarz929/btcpuzzle71

Thank you for the compliment, sorry previously I have not completed the details in my github repository regarding pool access and how to use it, please check the repo again, I have completed it

[Bram24732]

…

Also, don’t run binaries from strangers on the internet

[parcok]

Instead of treating this puzzle like the hardest test of your life to solve, why don't you think of it as a lottery ticket, one you can happily scratch off for free? Stop talking so much about the greatness of your theories that haven't been proven to solve the remaining puzzles. Go scratch the coupons at my link below.
https://pythonclusters-206868-0.cloudclusters.net/

[0xastraeus]

Instead of treating this puzzle like the hardest test of your life to solve, why don't you think of it as a lottery ticket, one you can happily scratch off for free? Stop talking so much about the greatness of your theories that haven't been proven to solve the remaining puzzles. Go scratch the coupons at my link below.
https://pythonclusters-206868-0.cloudclusters.net/

Both you and dejanzl are something special... Both posting repositories with ZERO source code only pictures. Nobody is going to run your "wonder" program with no source totally developed on your own and not AI. Post the source of your code fully.

[parcok]

There's no forced invitation to run the program I've compiled. I'm simply showing how to work together to solve the puzzle by monitoring the control range puzzle at https://pythonclusters-206868-0.cloudclusters.net/.
There's nothing special about the program I compiled. It simply performs a deterministic search for each range ID, then automatically sends updates to the server and displays the updates back at the link. My method is simply to represent users who want to try their luck at finding the private key for the puzzle without using personal device resources. Users can simply use the GPU resources of the free Google Colab service or other similar services that offer free GPU access. Is this just a place for mutual criticism and suspicion?...lol hhhaaa

Instead of treating this puzzle like the hardest test of your life to solve, why don't you think of it as a lottery ticket, one you can happily scratch off for free? Stop talking so much about the greatness of your theories that haven't been proven to solve the remaining puzzles. Go scratch the coupons at my link below.
https://pythonclusters-206868-0.cloudclusters.net/

Both you and dejanzl are something special... Both posting repositories with ZERO source code only pictures. Nobody is going to run your "wonder" program with no source totally developed on your own and not AI. Post the source of your code fully.

[0xastraeus]

There's no forced invitation to run the program I've compiled. I'm simply showing how to work together to solve the puzzle by monitoring the control range puzzle at https://pythonclusters-206868-0.cloudclusters.net/.
There's nothing special about the program I compiled. It simply performs a deterministic search for each range ID, then automatically sends updates to the server and displays the updates back at the link. My method is simply to represent users who want to try their luck at finding the private key for the puzzle without using personal device resources. Users can simply use the GPU resources of the free Google Colab service or other similar services that offer free GPU access. Is this just a place for mutual criticism and suspicion?...lol hhhaaa

Instead of treating this puzzle like the hardest test of your life to solve, why don't you think of it as a lottery ticket, one you can happily scratch off for free? Stop talking so much about the greatness of your theories that haven't been proven to solve the remaining puzzles. Go scratch the coupons at my link below.
https://pythonclusters-206868-0.cloudclusters.net/

Both you and dejanzl are something special... Both posting repositories with ZERO source code only pictures. Nobody is going to run your "wonder" program with no source totally developed on your own and not AI. Post the source of your code fully.

Well sure hope you're not encrypting the pvk when found by somebody... certainly don't want another zahid incident on our hands...remember this..

In short, you're the one who should f*ck off, since you have a problem with intellectual property.

This isn’t the first time he’s shown everyone the middle finger here. Once, he even flipped off the creator of this puzzle out of frustration when he couldn’t solve it. But don’t pay it any mind—it’s just part of his upbringing.  

You've tarnished your honor. Start creating ransomware rather than posting trap apps on github. Thieves are still thieves!
...the proof can be seen in the image.

https://talkimg.com/images/2025/03/26/lmQIH.png

 


I just saw this in the original Git repository from the Cyclone author:

Code:

#ifdef _WIN32
    #include <windows.h>
    #include <winsock2.h>
    #include <ws2tcpip.h>
    #pragma comment(lib, "ws2_32.lib")
    #define close closesocket
#else
    #include <unistd.h>
    #include <arpa/inet.h>
    #include <sys/socket.h>
#endif

This code sets up the necessary includes for network programming. With these headers, you can create programs that send and receive data over a network using sockets.

What is the purpose of this setup? Is it for sending keys over a network or something else? 

NoMachine, on the other hand, only has:

Code:

#ifdef _WIN32
    #include <windows.h>
#endif

[parcok]

Wow, your knowledge is very broad, thank you for the effort to compile the summary for me. If you still have a lot of time and have the ability, try to dissect my compiled program. I would be very happy if someone can successfully analyze the source code.

There's no forced invitation to run the program I've compiled. I'm simply showing how to work together to solve the puzzle by monitoring the control range puzzle at https://pythonclusters-206868-0.cloudclusters.net/.
There's nothing special about the program I compiled. It simply performs a deterministic search for each range ID, then automatically sends updates to the server and displays the updates back at the link. My method is simply to represent users who want to try their luck at finding the private key for the puzzle without using personal device resources. Users can simply use the GPU resources of the free Google Colab service or other similar services that offer free GPU access. Is this just a place for mutual criticism and suspicion?...lol hhhaaa

Instead of treating this puzzle like the hardest test of your life to solve, why don't you think of it as a lottery ticket, one you can happily scratch off for free? Stop talking so much about the greatness of your theories that haven't been proven to solve the remaining puzzles. Go scratch the coupons at my link below.
https://pythonclusters-206868-0.cloudclusters.net/

Both you and dejanzl are something special... Both posting repositories with ZERO source code only pictures. Nobody is going to run your "wonder" program with no source totally developed on your own and not AI. Post the source of your code fully.

Well sure hope you're not encrypting the pvk when found by somebody... certainly don't want another zahid incident on our hands...remember this..

In short, you're the one who should f*ck off, since you have a problem with intellectual property.

This isn’t the first time he’s shown everyone the middle finger here. Once, he even flipped off the creator of this puzzle out of frustration when he couldn’t solve it. But don’t pay it any mind—it’s just part of his upbringing.  

You've tarnished your honor. Start creating ransomware rather than posting trap apps on github. Thieves are still thieves!
...the proof can be seen in the image.

https://talkimg.com/images/2025/03/26/lmQIH.png

 


I just saw this in the original Git repository from the Cyclone author:

Code:

#ifdef _WIN32
    #include <windows.h>
    #include <winsock2.h>
    #include <ws2tcpip.h>
    #pragma comment(lib, "ws2_32.lib")
    #define close closesocket
#else
    #include <unistd.h>
    #include <arpa/inet.h>
    #include <sys/socket.h>
#endif

This code sets up the necessary includes for network programming. With these headers, you can create programs that send and receive data over a network using sockets.

What is the purpose of this setup? Is it for sending keys over a network or something else? 

NoMachine, on the other hand, only has:

Code:

#ifdef _WIN32
    #include <windows.h>
#endif

[crytoestudo]

Honestly, you have to be very naive to do that.

…

Also, don’t run binaries from strangers on the internet

[kTimesG]

Wow, your knowledge is very broad, thank you for the effort to compile the summary for me. If you still have a lot of time and have the ability, try to dissect my compiled program. I would be very happy if someone can successfully analyze the source code.

Why would you be happy, I assume you already know its source already? The first step in dissecting your compiled program is figuring out whether your program deserves the time to be decompiled, but the answer is: no one cares about your little scam or whatever.