Author

Topic: Bitcoin puzzle transaction ~32 BTC prize to who solves it - page 177. (Read 244737 times)

member
Activity: 194
Merit: 14
And then check to see if you can find any pattern.
would you expect one ?

Lol. We've gone way off topic damn the son of the sun...

I think Satoshi should recharge our batteries and boost our motivation by skyrocketing the puzzles again by 10x.

Would you agree guys?
hero member
Activity: 630
Merit: 731
Bitcoin g33k
And then check to see if you can find any pattern.
would you expect one ?
copper member
Activity: 1330
Merit: 899
🖤😏
If anyone is interested in comparing base58 character match, rmd160 character match etc, I have an idea and if you are a coder you might be able to pull it off.

Basically what you need is a color encoding  tool to create colors from base58 or rmd160 hash characters, once you have your algorithm ready, you just need encode addresses or hashes with this new algo, then you should take a look at the color spectrum of the addresses you find in a certain range and compare it for example with #66 address and it's color composition, if the found similar matches were of the same colors, you might have a chance to determine whether you are close to puzzle address or not.

First thing you should do is converting adds/hashes into color bars something like this:

1111111111111111111114oLvT2

Rmd160 :

0000000000000000000000000000000000000000

Add:
11111111111111111111BZbvjr
Rmd160:
0000000000000000000000000000000000000001
Add:
11111111111111111C3CPpNyMcq
Rmd160:
0000000000000000000000000000000001234567

Or you could base the colors on a certain range, like  the above hexadecimal rmd160 converted to decimal is :
19088743, then you could mix color hues together like 1908, 1+9+8= 18, so you'd set a color for any sum of 4 digits less than 20 to red, if the sum of next 4 digits are 8+7+4+3= 22 between 22 and 25 you'd set a different color, if the sum of 4 digits is between 25 and 30, another color and so on.

Then you won't need to look at the hex/decimal values anymore, just pure color compositions of certain hashes in different bit ranges. And then check to see if you can find any pattern.
hero member
Activity: 630
Merit: 731
Bitcoin g33k
for the first time in my life i agree with citb0in
cauz' you didn't like or understand that replyGrin love it or hate it - off-topic enough.

Very boring

indeed, such requests (asking for partial matched strings) are boring even if reoccurring from time to time
newbie
Activity: 26
Merit: 0
Very boring comments, please DM to me only if anyone has one! thanks
member
Activity: 194
Merit: 14
Has anyone ever found a key matching: 13zb1hQbWVsc ? just would be interesting to see!
what for ? why should anyone be interested in finding a partly matching string of a puzzle BTC address ? it's completely useless

for the first time in my life i agree with citb0in
hero member
Activity: 630
Merit: 731
Bitcoin g33k
Has anyone ever found a key matching: 13zb1hQbWVsc ? just would be interesting to see!
what for ? why should anyone be interested in finding a partly matching string of a puzzle BTC address ? it's completely useless
newbie
Activity: 26
Merit: 0
Has anyone ever found a key matching: 13zb1hQbWVsc ? just would be interesting to see!
member
Activity: 93
Merit: 16
Alek, for the win. Long time no hear/see.
Glad you’ve made it back.

What does your new program do exactly? The VanitySearch one.

Myself, no one wanted to combine forces to attack 130, so I’ve just been tinkering on different things.
Nothing new, not any closer than I was months ago.
Thanks! This is a fork, it’s not mine programm, the program is optimized for a puzzle, but I removed everything unnecessary from the code and introduced a new random generator - no predictable random, as it was before.
member
Activity: 282
Merit: 20
the right steps towerds the goal

 There is no pattern here. But there is a time when a puzzle was created. I even went back in time and generating numbers from 2015. Every second of that year.


https://bitcointalksearch.org/topic/m.62144570

Almost six months ago, I undertook various tasks of a similar stuffs..
member
Activity: 503
Merit: 38
@nomachine what is the problem with the original code?

Quote
if a non-deterministic source (e.g. a hardware device) is not available to the implementation. In this case each std::random_device object may generate the same number sequence.


Everything is great with the code. Even std::random_device works great for me. I am experimenting with different generators and speeds. It's not even an ordinary platform. I'm currently on an ARM machine.



 There is no pattern here. But there is a time when a puzzle was created. I even went back in time and generating numbers from 2015. Every second of that year.

Code:
add = "13zb1hQbWVsc2S7ZTZnP2G4undNNpdh5so"
# Specify the start date and time
start_datetime = datetime(2015, 1, 13, 19, 7, 14)
current_datetime = start_datetime

# Specify the end date and time
end_datetime = datetime(2015, 1, 15, 19, 7, 14)

while current_datetime <= end_datetime:
    # Convert the current datetime to a Unix timestamp
    timestamp = int(current_datetime.timestamp())

    random.seed(timestamp)
    key = random.randint(start_range, end_range)
    public_key = private_key_to_public_key(key)
    bitcoin_address = public_key_to_address(public_key, compressed=True)
    public_key = public_key_to_hex(public_key)

    # Increment the current datetime by one second for the next timestamp
    current_datetime += timedelta(seconds=1)

you won't believe what I've tried so far.
copper member
Activity: 1330
Merit: 899
🖤😏
WP, for the android, no lung no ear, glad you made it back from where ever you were. Lol.


Ahhh, Joes? Jules? Sup.

Try this, I just messed up when trying to divide each target by different scalar, so I can't tell whether this is dividing 1 target by odd and the other target by even values or not, if you could try it with scalar mod n instead of points, do try and also share it with us, I have no time to do it right now.

Code:
import gmpy2 as mpz
from gmpy2 import powmod

# Define the EllipticCurve class
class EllipticCurve:
    def __init__(self, a, b, p):
        self.a = mpz.mpz(a)
        self.b = mpz.mpz(b)
        self.p = mpz.mpz(p)

    def contains(self, point):
        x, y = point.x, point.y
        return (y * y) % self.p == (x * x * x + self.a * x + self.b) % self.p

    def __str__(self):
        return f"y^2 = x^3 + {self.a}x + {self.b} mod {self.p}"

# Define the Point class
class Point:
    def __init__(self, x, y, curve):
        self.x = mpz.mpz(x)
        self.y = mpz.mpz(y)
        self.curve = curve

    def __eq__(self, other):
        return self.x == other.x and self.y == other.y and self.curve == other.curve

    def __ne__(self, other):
        return not self == other

    def __add__(self, other):
        if self.curve != other.curve:
            raise ValueError("Cannot add points on different curves")

        # Case when one point is zero
        if self == Point.infinity(self.curve):
            return other
        if other == Point.infinity(self.curve):
            return self

        if self.x == other.x and self.y != other.y:
            return Point.infinity(self.curve)

        p = self.curve.p
        s = 0
        if self == other:
            s = ((3 * self.x * self.x + self.curve.a) * powmod(2 * self.y, -1, p)) % p
        else:
            s = ((other.y - self.y) * powmod(other.x - self.x, -1, p)) % p

        x = (s * s - self.x - other.x) % p
        y = (s * (self.x - x) - self.y) % p

        return Point(x, y, self.curve)

    def __sub__(self, other):
        if self.curve != other.curve:
            raise ValueError("Cannot subtract points on different curves")

        # Case when one point is zero
        if self == Point.infinity(self.curve):
            return other
        if other == Point.infinity(self.curve):
            return self

        return self + Point(other.x, (-other.y) % self.curve.p, self.curve)

    def __mul__(self, n):
        if not isinstance(n, int):
            raise ValueError("Multiplication is defined for integers only")

        n = n % (self.curve.p - 1)
        res = Point.infinity(self.curve)
        addend = self

        while n:
            if n & 1:
                res += addend

            addend += addend
            n >>= 1

        return res

    def __str__(self):
        return f"({self.x}, {self.y}) on {self.curve}"

    @staticmethod
    def from_hex(s, curve):
        if len(s) == 66 and s.startswith("02") or s.startswith("03"):
            compressed = True
        elif len(s) == 130 and s.startswith("04"):
            compressed = False
        else:
            raise ValueError("Hex string is not a valid compressed or uncompressed point")

        if compressed:
            is_odd = s.startswith("03")
            x = mpz.mpz(s[2:], 16)

            # Calculate y-coordinate from x and parity bit
            y_square = (x * x * x + curve.a * x + curve.b) % curve.p
            y = powmod(y_square, (curve.p + 1) // 4, curve.p)
            if is_odd != (y & 1):
                y = -y % curve.p

            return Point(x, y, curve)
        else:
            s_bytes = bytes.fromhex(s)
            uncompressed = s_bytes[0] == 4
            if not uncompressed:
                raise ValueError("Only uncompressed or compressed points are supported")

            num_bytes = len(s_bytes) // 2
            x_bytes = s_bytes[1 : num_bytes + 1]
            y_bytes = s_bytes[num_bytes + 1 :]

            x = mpz.mpz(int.from_bytes(x_bytes, byteorder="big"))
            y = mpz.mpz(int.from_bytes(y_bytes, byteorder="big"))

            return Point(x, y, curve)

    def to_hex(self, compressed=True):
        if self.x is None and self.y is None:
            return "00"
        elif compressed:
            prefix = "03" if self.y & 1 else "02"
            return prefix + hex(self.x)[2:].zfill(64)
        else:
            x_hex = hex(self.x)[2:].zfill(64)
            y_hex = hex(self.y)[2:].zfill(64)
            return "04" + x_hex + y_hex

    @staticmethod
    def infinity(curve):
        return Point(-1, -1, curve)

# Define the ec_mul function
def ec_mul(point, scalar, base_point):
    result = Point.infinity(point.curve)
    addend = point

    while scalar:
        if scalar & 1:
            result += addend

        addend += addend
        scalar >>= 1

    return result

# Define the ec_operations function
def ec_operations(start_range, end_range, target_1, target_2, curve, divide_1_by_odd=True, divide_1_by_even=True, divide_2_by_odd=True, divide_2_by_even=True):
    # Define parameters for secp256k1 curve
    n = mpz.mpz("0xfffffffffffffffffffffffffffffffebaaedce6af48a03bbfd25e8cd0364141")
    G = Point(
        mpz.mpz("0x79be667ef9dcbbac55a06295ce870b07029bfcdb2dce28d959f2815b16f81798"),
        mpz.mpz("0x483ada7726a3c4655da4fbfc0e1108a8fd17b448a68554199c47d08ffb10d4b8"),
        curve
    )

    for i in range(start_range + (start_range%2), end_range, 3):
        # divide target 1 by odd or even numbers
        if i%2 == 0 and not divide_1_by_even:
            continue
        elif i%2 == 1 and not divide_1_by_odd:
            continue
        try:
            # calculate inverse modulo of i
            i_inv = powmod(i, n-2, n)

            # divide the targets by i modulo n
            result_1 = ec_mul(target_1, i_inv, G)
            result_2 = ec_mul(target_2, i_inv, G)

            # divide target 2 by odd or even numbers
            if i%2 == 0 and not divide_2_by_even:
                continue
            elif i%2 == 1 and not divide_2_by_odd:
                continue

            # subtract the results
            sub_result = result_2 - result_1

            # write the results to separate files
            print(f"{i}-{sub_result.to_hex()}")

        except ZeroDivisionError:
            pass


if __name__ == "__main__":
    # Set the targets and range for the operations
    curve = EllipticCurve(
        mpz.mpz(0),
        mpz.mpz(7),
        mpz.mpz("0xFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFEFFFFFC2F")
    )

    target_1 = Point.from_hex("037564539e85d56f8537d6619e1f5c5aa78d2a3de0889d1d4ee8dbcb5729b62026", curve)

    target_2 = Point.from_hex("03633cbe3ec02b9401c5effa144c5b4d22f87940259634858fc7e59b1c09937852", curve)
   
    start_range = 0
    end_range = 256
   
    ec_operations(start_range, end_range, target_2, target_1, curve)

Haters gonna keep hatin', bitches gonna keep bitchin. Beware of the old lady ahead, her whining kills.😉
hero member
Activity: 862
Merit: 662
@nomachine what is the problem with the original code?

Quote
if a non-deterministic source (e.g. a hardware device) is not available to the implementation. In this case each std::random_device object may generate the same number sequence.

For Linux the best source of entropy is the internal RNG.


I strongly advice not use that code provide by @nomachine if you are using the program to generate your vanity addresses.


If the code is not going to be used for that and you are pretty sure that your machine have a RNG Hardware device the it is ok to use it
member
Activity: 503
Merit: 38


- we run keyhunt in BSGS mode on 8 CPU threads with about 16GB RAM allocation and get about 80 PKey/s.

Is Kangaroo with multi-GPU usage more advantageous and more likely to get a hit, even though keyhunt with CPU usage in BSGS can process 80 PKey/s?



Try changing the random generator in both. Play what works best in your environment.
example

keyhunt.cpp

replace
Code:
#include
to
Code:
#include
replace  seed random generator with minstd_rand rng
Code:
#if defined(_WIN64) && !defined(__CYGWIN__)
    // Any Windows secure random source goes here
    rseed(clock() + time(NULL) + rand());
#else
    std::random_device rd;
    std::minstd_rand rng(rd());
    
    // Seed the random number generator
    rseed(rng());
#endif

I have from 80  to ~177 Pkeys/s (177940438134284990 keys/s) Grin
These apps can't get better if we don't all work on them and test them.


An windows version ?

Code:
#if defined(_WIN64) && !defined(__CYGWIN__)
    // On Windows, use a combination of clock, time, and rand as the seed
    return std::minstd_rand(static_cast(clock() + time(nullptr) + rand()));
#else
    // On non-Windows platforms (including Linux), use std::random_device to seed
    std::random_device rd;
    return std::minstd_rand(rd());


This code should work on both Windows and Linux.
I also use -O3 flag during compilation.
Aggressive optimizations can greatly improve performance,
they may also increase compilation time, and in some cases, they can make debugging more challenging because the optimized code may differ significantly from the original source code.
member
Activity: 206
Merit: 16


- we run keyhunt in BSGS mode on 8 CPU threads with about 16GB RAM allocation and get about 80 PKey/s.

Is Kangaroo with multi-GPU usage more advantageous and more likely to get a hit, even though keyhunt with CPU usage in BSGS can process 80 PKey/s?



Try changing the random generator in both. Play what works best in your environment.
example

keyhunt.cpp

replace
Code:
#include
to
Code:
#include
replace  seed random generator with minstd_rand rng
Code:
#if defined(_WIN64) && !defined(__CYGWIN__)
    // Any Windows secure random source goes here
    rseed(clock() + time(NULL) + rand());
#else
    std::random_device rd;
    std::minstd_rand rng(rd());
    
    // Seed the random number generator
    rseed(rng());
#endif

I have from 80  to ~177 Pkeys/s (177940438134284990 keys/s) Grin
These apps can't get better if we don't all work on them and test them.


An windows version ?
full member
Activity: 1232
Merit: 242
Shooters Shoot...
Alek, for the win. Long time no hear/see.
Glad you’ve made it back.

What does your new program do exactly? The VanitySearch one.

Myself, no one wanted to combine forces to attack 130, so I’ve just been tinkering on different things.
Nothing new, not any closer than I was months ago.
member
Activity: 503
Merit: 38


- we run keyhunt in BSGS mode on 8 CPU threads with about 16GB RAM allocation and get about 80 PKey/s.

Is Kangaroo with multi-GPU usage more advantageous and more likely to get a hit, even though keyhunt with CPU usage in BSGS can process 80 PKey/s?



Try changing the random generator in both. Play what works best in your environment.
example

keyhunt.cpp

replace
Code:
#include
to
Code:
#include
replace  seed random generator with minstd_rand rng
Code:
#if defined(_WIN64) && !defined(__CYGWIN__)
    // Any Windows secure random source goes here
    rseed(clock() + time(NULL) + rand());
#else
    std::random_device rd;
    std::minstd_rand rng(rd());
    
    // Seed the random number generator
    rseed(rng());
#endif

I have from 80  to ~177 Pkeys/s (177940438134284990 keys/s) Grin
These apps can't get better if we don't all work on them and test them.
member
Activity: 239
Merit: 53
New ideas will be criticized and then admired.
Hey all, correct me if I'm wrong please: Since the rewards got multiplied by 10X nothing has been found! Correct?
only #125
newbie
Activity: 4
Merit: 0
Hey all, correct me if I'm wrong please: Since the rewards got multiplied by 10X nothing has been found! Correct?
jr. member
Activity: 75
Merit: 5
Hi all,

to what extent can the performance and the hit probability of these two tools be compared and what significance do they then have?

For example:

- we run Kangaroo in multi-GPU mode on several modern GPUs and achieve a displayed speed of 15 GKey/s

- we run keyhunt in BSGS mode on 8 CPU threads with about 16GB RAM allocation and get about 80 PKey/s.

Is Kangaroo with multi-GPU usage more advantageous and more likely to get a hit, even though keyhunt with CPU usage in BSGS can process 80 PKey/s?

From this numerical example, one would think that keyhunt with the 80 PKey/s would have to be worlds better. But is that the case or is it not possible to compare this at all and the given speed rates are completely useless for the comparison?


the easiest example I can give to you for this question is a "web" and a "gunshot"
now picture both having the same momentum
Kangaroo would be considered a "web" moving at the same momentum as the "gunshot" from BSGS
and what do webs do? it spreads, covering so much area and what do gunshots do? it moves on a straight line
Now the preference is you choice and your available resources
they're both super fast like I've explained earlier.
let's say you have a single 4090 and 16 cpu threads and you combined both and you got the speed of 5GKey/s
and you choose not to use the 4090 but you rather use the 16 cpu thread on your machine and you have 128GB RAM and you chose -k 7000 and you achieved the speed of 1 EKey/s you got so much momentum on both processes but based on my calculations, you would be 90% slower using the latter compared to the former because they are both having ETAs which can easily be calculated
Mind you this has nothing to do with what I am saying, it has everything to do with your preference and resources
Jump to: