Bitcoin puzzle transaction ~32 BTC prize to who solves it

[Wanderingaran]

In about 10-20 seconds you will be left with nothing.

I just need someone to convince me that this is ethical. So far, no one has been able to prove to me that it’s legal. 

[kTimesG]

In about 10-20 seconds you will be left with nothing.

I just need someone to convince me that this is ethical. So far, no one has been able to prove to me that it’s legal.  

I think you're missing the principles of why Bitcoin exists: obtaining security inside an anonymous but totally insecure environment. There's absolutely nothing secure with using a private key that is known to be insecure. Ethical or legal issues are outside the point here. Also, the stupidity and naivity of using an insecure key in an insecure environment, full of unreliable participants, and yelling left and right "here's the private key, my dear ethical friends which I never met but I completely blindly trust" always has a cost.

Honestly, there is no one who can even prove that puzzles 66 and 69 were solved by them, simply because the TXs that were archived by sites such as mempool.space might have actually received an already replaced TX, from the very beginning. We'll never know.

[Cricktor]

Many paths lead to Rome, as the saying goes.

With an own node, works with a heavily pruned one just fine,  you can set it up to run a script once new transactions arrive or regularly poll also your node's mempool for a transaction which involves any of the puzzle's unsolved addresses. RPC getdescriptoractivity, properly used, might work. Or ZeroMQ stuff... I haven't tried, I'm not a coder, but it's not really rocket science.

Once a vulnerable public key is publicly exposed which can be solved fast enough, stealing bots can and will kick in and start a bidding war with increasing fees because that's how replacement of transactions with unconditional FullRBF works. Do your homework if you don't understand how FullRBF works (for those who phantasize about it's enough to broadcast a tx with just a high fee to succeed).

BTW, worse case shit happens: block 909211 took almost 95 minutes to be mined. If you're unlucky and stupid enough to publicly broadcast e.g. solution for puzzle #71 in such a scenario where the next block takes significantly longer to be mined than the average of about 10min, there's plenty of time for a lot of bots to compete.

Yes, you could be lucky that the next block is mined very few seconds after you broadcasted your transaction publicly and it reached the mining pool's mempools. You could also be very unlucky because mining is unpredictable and finding a valid block hash is a random process.

It's funny how little knowledge some seem to have about Bitcoin's current basic mechanics. Sadly, this thread is full of garbage and nonsense chit-chat.

[Akito S. M. Hosana]

Full RBF turns the mempool into the Hunger Games for puzzle solutions. And yeah, if you broadcast #71’s key during a 95-minute block delay, even my grandma’s toaster could jump in the bidding warassuming it’s running Bitcoin Core 29.0, that is.  

[mahmood1356]

Address: 1GjerJf1FeccCzvQUZVpeVmPpad3RH1ZT5
Public Key: 03af13c80e78581d870a96f112cf681db1cad6f9da26860f2c25dd9a9125b0bdfc
Range(18hex): 400000000000000000 - 7FFFFFFFFFFFFFFFFF

I have a modest rtx 2060S + GTX1060ti — solution in about 2-5 minutes. For the sake of decency, you could at least put 1 dollar there. 

Code:

~/RCKangaroo$ time ./rckangaroo -dp 16 -range 85 -start 1 -pubkey 03af13c80e78581d870a96f112cf681db1cad6f9da26860f2c25dd9a9125b0bdfc -tames tames85.dat -max 13
********************************************************************************
*                    RCKangaroo v3.0  (c) 2024 RetiredCoder                    *
********************************************************************************

This software is free and open-source: https://github.com/RetiredC
It demonstrates fast GPU implementation of SOTA Kangaroo method for solving ECDLP
Linux version
CUDA devices: 2, CUDA driver/runtime: 12.4/12.0
GPU 0: NVIDIA GeForce RTX 2060 SUPER, 7.78 GB, 34 CUs, cap 7.5, PCI 3, L2 size: 4096 KB
GPU 1: NVIDIA GeForce GTX 1660 Ti, 5.79 GB, 24 CUs, cap 7.5, PCI 8, L2 size: 1536 KB
Total GPUs for work: 2

MAIN MODE

Solving public key
X: AF13C80E78581D870A96F112CF681DB1CAD6F9DA26860F2C25DD9A9125B0BDFC
Y: C9EDCFE647622731DBE2E9E7291E8498BA0AFD44D3356C59B96F630218BDD3FB
Offset: 0000000000000000000000000000000000000000000000000000000000000001

Solving point: Range 85 bits, DP 16, start...
SOTA method, estimated ops: 2^42.702, RAM for DPs: 4.253 GB. DP and GPU overheads not included!
Max allowed number of ops: 2^46.402, max RAM for DPs: 53.044 GB
Estimated DPs per kangaroo: 57.427.
load tames...
tames loaded
GPU 0: allocated 3284 MB, 1114112 kangaroos. OldGpuMode: Yes
GPU 1: allocated 2322 MB, 786432 kangaroos. OldGpuMode: Yes
GPUs started...
MAIN: Speed: 2074 MKeys/s, Err: 0, DPs: 1418283K/109142K, Time: 0d:00h:00m/0d:00h:57m
MAIN: Speed: 2758 MKeys/s, Err: 0, DPs: 1418714K/109142K, Time: 0d:00h:00m/0d:00h:43m
MAIN: Speed: 2745 MKeys/s, Err: 0, DPs: 1419132K/109142K, Time: 0d:00h:00m/0d:00h:43m
***
MAIN: Speed: 2695 MKeys/s, Err: 0, DPs: 1424111K/109142K, Time: 0d:00h:02m/0d:00h:44m
Stopping work ...
Point solved, K: 0.066 (with DP and GPU overheads)

PRIVATE KEY: 0000000000000000000000000000000000000000000000612E8BA2E8BAC75640

real	4m35,010s
user	1m11,596s
sys	0m45,348s[s][/s]

All that effort for nothing (:  now he’ll show up with that brilliant brain and groundbreaking ideas and say:

“See? It gets found in 2 to 5 minutes, and during that time I can just pay a high fee and make a regular transfer.”

LMAO anyway, I had a lot of fun today.

Yeah, my man Spirik gotcha ya. Probably he’s gonna do that, he’s gonna rely on that time until bot crack it and steal the funds. But wait… he cracked it with a simple hardware, what about a powerful hardware which can do that faster and drop the time to a few seconds? Lmao certainly he’s gonna desapear and go back with a new account or just make up a new argument to not accept how things in our reality works. Just like the people who believes Karl Marx’s theories would actually work out.

Hello and thank you for doing this and clarifying the matter. I used to contradict this, but now I fully understand it and I apologize to my friends with whom I disagreed on this issue.

[SimonNeedsBitcoin]

How do you know the winner is transferring and broadcasting the fund? Do you get the public key immediately after the winner broadcast the transation? Does this take time?

Yes. In about 10-20 seconds you will be left with nothing.

It takes less than 2 seconds to break the private key of Puzzle 71 (once the initial TX is in the mempool), create the TX, and broadcast it, and to get it accepted by any miner.

The principle behind the puzzle bot can be explained more simply:

The bot checks a Bitcoin address for incoming transactions using a Python script.

If a transaction is found, it extracts the public key from the scriptsig field in the transaction input.

Example one-line Python command for puzzle 71:

Code:

python3 -c "import requests; import sys; address = sys.argv[1]; url = f'https://mempool.space/api/address/{address}/txs/chain'; r = requests.get(url); txs = r.json(); pubkey = next((vin['scriptsig'][-66:] for tx in txs for vin in tx['vin'] if 'scriptsig' in vin), None); print(f'Public key for address {address}: {pubkey}' if pubkey else 'Public key not found');" 1PWo3JeB9jrGwfHDNpdGK54CRas7fsVzXU

If the public key is found, it will appear like this:

Public key for address 197kFKvMHoRJPXktc8xJwMjeTuE9xijBQ: 029fd3d2479a37f40d03975cb51ce0fa18cbb709cc46b47caeaa1722cfd3683583

If not, it will return:

Public key not found

You can manually press Enter repeatedly until the public key appears.

Alternatively, you can automate it with a Python script that checks every 5 seconds indefinitely.

The public key will appear as soon as someone broadcasts a transaction from that address.

Faster Execution with a Local Node

If you're running a local Bitcoin node (e.g., with Umbrel), you can query your own mempool instead of an external API:

Code:

url = f'http://mempool.localhost/api/address/{address}/txs/chain'

This reduces latency and speeds up detection.

Once the public key is obtained, Kangaroo is used to derive the private key, typically in 2-3 seconds.

The script then replaces the original transaction in the mempool before it gets confirmed.

Example:

bitcoin-cli createrawtransaction 'INPUTS' 'OUTPUTS'
bitcoin-cli signrawtransactionwithkey 'HEX' '["PRIVATE_KEY"]'
bitcoin-cli sendrawtransaction 'SIGNED_HEX'

No Rate Limits: Local mempool API is faster and unrestricted.

Real-Time Detection: Sees transactions as soon as they enter the mempool.

Thank you for your detailed explanation. It seems that this kind of robot can really steal the bonus very quickly. Although I don’t believe it can be done in 2 seconds, it seems that it can be done within a minute or even tens of seconds. Broadcasting in a public mining pool is indeed very dangerous.

[Igor_cherkassy]

Can anyone share any key to 1PWo3JeB9jr.... Thanks.

[mahmood1356]

Can anyone share any key to 1PWo3JeB9jr.... Thanks.

649a103b78b3e74856

[kTimesG]

It seems that this kind of robot can really steal the bonus very quickly. Although I don’t believe it can be done in 2 seconds, it seems that it can be done within a minute or even tens of seconds. Broadcasting in a public mining pool is indeed very dangerous.

Good, progress.

Now, let's take it one step forward.

Having the pubKey of Puzzle #71 means solving the ECDLP in a 70-bits interval.

Algorithms such as BSGS or Kangaroo will thus require around 2**36 EC group operations, when starting from a zero setup.

Let's take RCKangaroo as a base reference (just for example sake, it is NOT an optimized solver).

With that demo program, an RTX 4090 can do around 8 Gop/s EC (group operations per second).

So, the ECDLP will be broken in around 8.5 seconds.

Let's now actually take it one step further and solve it much faster than that.

If you investigate why the ECDLP of a 70-bit interval is solved in only 2 * 2**35 steps, you will notice that 2**35 of those steps start off from a known private key. So the solving equation can be expressed like this:

intervalSize = numRequiredOpsA * numRequiredOpsB
(example: 2**70 = 2**35 * 2**35)

where you can twiggle around with the right-hand numbers.

Mathematically, the optimal (minimal) effort occurs when those two numbers have the same magnitude, because that is when the sum of the numbers is at a minimum.

However, if one wants to accelerate the solving by trading off the initial effort, we can rewrite things like this:

intervalSize = bigNumberA * smallNumberB
(example: 2**70 = 2**70 * 1)

In this example, somebody precomputes and stores all the keys of the interval, and when the public key becomes available, then the solving side only takes one step (a lookup).

D. J. Bernstein wrote a great paper on how we can take advantage of this in the ECDLP context. It turns out, that we can do something like this if we want:

bigNumberA = 2**40
smallNumberB = 2**30

In this example, we can spend a decent amount of time working out and storing data for whatever 2**40 private keys that we wish. Once we have this data prepared, when the public key is revealed, we'd only need 2**30 steps to find the associated private key.

Now, let's revisit the solve times, on a RTX 4090 (just for fun):

Precomputing time: 2**40 steps = 138 seconds
Solving time: 2**30 steps = 0.13 seconds

Maybe this should make people understand why puzzles 71 to 100+ are basically sitting ducks, and can be solved in a matter of minutes (in total).

[MrGPBit]

Can anyone share any key to 1PWo3JeB9jr.... Thanks.

649a103b78b3e74856

Public Addr: 1PWo3JeB9jRLz2NjTHsZ8uTBnqcjnu66Ak

Unfortunately, the letter R is capitalized and does not match the destination address

[viljy]

~
D. J. Bernstein wrote a great paper on how we can take advantage of this in the ECDLP context. It turns out, that we can do something like this if we want:

bigNumberA = 2**40
smallNumberB = 2**30

In this example, we can spend a decent amount of time working out and storing data for whatever 2**40 private keys that we wish. Once we have this data prepared, when the public key is revealed, we'd only need 2**30 steps to find the associated private key.

Now, let's revisit the solve times, on a RTX 4090 (just for fun):

Precomputing time: 2**40 steps = 138 seconds
Solving time: 2**30 steps = 0.13 seconds

Maybe this should make people understand why puzzles 71 to 100+ are basically sitting ducks, and can be solved in a matter of minutes (in total).

An interesting idea. And it's actually so simple, and very clever. I sometimes read threads about the puzzle, I even searched for private keys before, until I saw that you can hardly do without large financial costs.


P.S. As for the 71 range, I don't see a bot problem at all if the person who found the key uses Mara so that the transaction doesn't end up in the mempool.

[uvindele]

Can anyone share any key to 1PWo3JeB9jr.... Thanks.

7F92587DD9BD019205

[Tony8989]

Mara is a tool designed to quickly replace or front-run unconfirmed transactions by broadcasting competing ones. For the attacker, it’s effective and “safe” to use. For the victim, it means a real risk of losing funds if their transaction gets replaced before confirmation

basically Mara is one of several tools designed to do the same thing: quickly replace or front-run unconfirmed Bitcoin transactions by broadcasting competing transactions with higher fees.

What makes Mara notable is it’s often optimized for speed and automation, especially in the context of puzzle key cracking and mempool monitoring.

But fundamentally, it uses Bitcoin’s standard Replace-By-Fee (RBF) or mempool replacement rules—just like any other mempool??

[BlackAKAAngel]

my  BOT  its foolauto  


Monitor the address for new transactions. evry two secund refresh
 When a new transaction appears, it will get the txid and the amount
 it  will then extract the public key
 it will run RCKangaroo to find the private key
 As soon as the private key is found, it will
 Get the recommended transaction fee
 Create a new transaction to send the funds to your destination address, including the calculated fee.
 Sign the transaction
 Print the raw transaction hex to the console
 
and i testet work well

[Jorge54PT]

Mara is a tool designed to quickly replace or front-run unconfirmed transactions by broadcasting competing ones. For the attacker, it’s effective and “safe” to use. For the victim, it means a real risk of losing funds if their transaction gets replaced before confirmation

basically Mara is one of several tools designed to do the same thing: quickly replace or front-run unconfirmed Bitcoin transactions by broadcasting competing transactions with higher fees.

What makes Mara notable is it’s often optimized for speed and automation, especially in the context of puzzle key cracking and mempool monitoring.

But fundamentally, it uses Bitcoin’s standard Replace-By-Fee (RBF) or mempool replacement rules—just like any other mempool??

Mara doesn't publish the public key. Without it, no bot will notice any ID txid , public key, or anything else. When the origin address is 0, the all balance it's already in my new wallet then in the mempool my transaction appears saying: Public key not available or not published  something like that.
in short: bots are useless with Mara

https://www.binance.com/en/square/post/22423542983737

[SimonNeedsBitcoin]

my  BOT  its foolauto  


Monitor the address for new transactions. evry two secund refresh
 When a new transaction appears, it will get the txid and the amount
 it  will then extract the public key
 it will run RCKangaroo to find the private key
 As soon as the private key is found, it will
 Get the recommended transaction fee
 Create a new transaction to send the funds to your destination address, including the calculated fee.
 Sign the transaction
 Print the raw transaction hex to the console
 
and i testet work well

Does anyone else here have objections of designing such a robot? Is this unethical behavior commonplace?

[Cricktor]

My viewpoint is that bots are part of the game and challenge. Bots use publicly available data to compete and I don't really see anything unethical with that.

When there are developed public tools which can find a private key much faster once the public key is publicly exposed, then anyone who participates in finding puzzle solutions has to cope with that.

You may perceive it as unfair because it's hard and costly to find the solution for #71^{or any remaining below 100ish} as long as the public key is yet unknown. The puzzle creator gave his coins to be redeemed by anyone who can find a private key. Week entropy private keys are what they are: no safe protection for coins they control once the public key is exposed (and the weak entropy is known). If exposure of the public key makes the coins very vulnerable, then this is it exactly. Deal with it!

I may have referred in the past of such bots as stealer bots. After some thinking about it and my viewpoint presented above, I wouldn't declare it that way anymore. It's in my opinion no stealing.

YMMV, I'm genuinely interested how others see it. Please, try to explain if your opinion differs. I might be wrong with my own opinion and if so, I'd like to learn why!

[Tony8989]

ok so what is the solution i mean now the hard part is to get the bitcoin safe to your wallet, puzzle 71 is hard but not hard but the hard part is to send the found to your wallet
 lots of guys talk about mara but feel like mara is bots himself, and what about puzzle 67,66,65 ect found where safe send it to the wallet with not problem

[benjaniah]

ok so what is the solution i mean now the hard part is to get the bitcoin safe to your wallet, puzzle 71 is hard but not hard but the hard part is to send the found to your wallet
 lots of guys talk about mara but feel like mara is bots himself, and what about puzzle 67,66,65 ect found where safe send it to the wallet with not problem

You got it backwards, bro.
The hard part is finding the private key to the puzzle wallet address.
The easy part is creating the transaction that spends the puzzle UTXO's to your own wallet, and submitting the tx to Mara Slipstream.

[Tony8989]

Let’s get you verified
Select your residency and follow the steps
Residence
United Kingdom
Our service is not available at this time
Binance is currently reorganising its activities in the UK. Our services will be available soon. Stay tuned!