Bitcoin puzzle transaction ~32 BTC prize to who solves it

[kTimesG]

1MVDYgVaSN6ApM5LJXriH8cFZkgfCia6Dh   xxxxxxxxxx?

Unless you spent at least 2500 dollars in the last 14 days, I don't think you have the key for that one.

[bibilgin]

Unless you spent at least 2500 dollars in the last 14 days, I don't think you have the key for that one.

Are you up for the 1 btc bet?

I wrote before.
I have 3 4070 ti super and 2 4070 graphics cards.

My only expense is Electricity..

In fact, the company I work for pays for electricity.

Actually, I wrote it before. I can find average PROBABILITY intervals.

[Desyationer]

I'm curious about how much the RTX 5090 outperforms the RTX 4090 in Bitcoin private key cracking speed. In programs like bitcrack, keyhunt, and vanbitcracken, the 4090 reaches around 4 Gkeys/s. So far, I haven’t found specific data on the RTX 5000 series for this niche test. However, I came across a post on the Hashcat forum suggesting that in one mode, the 5090 might be up to three times faster than the 4090. The best way to get a definitive answer is probably to ask on that forum—some users likely already have 5000-series GPUs, whether it's a 5090, 5080, or even a 5070.

[kTimesG]

Unless you spent at least 2500 dollars in the last 14 days, I don't think you have the key for that one.

Are you up for the 1 btc bet?

I wrote before.
I have 3 4070 ti super and 2 4070 graphics cards.

My only expense is Electricity..

In fact, the company I work for pays for electricity.

Actually, I wrote it before. I can find average PROBABILITY intervals.

Then post the SHA-256 of the public key of that address, no need for any bets to prove you own the private key for that address. You can't? Oh well, that's probably because you would need 40 RTX 4090 GPUs running for 2 weeks at a rate of 7 Gkeys/s in order to stumble upon such an address with 58 matched bits. But I'm sure that with your super-quick budget GPUs at your disposal you got it in like, 2 hours.

I'm curious about how much the RTX 5090 outperforms the RTX 4090 in Bitcoin private key cracking speed. In programs like bitcrack, keyhunt, and vanbitcracken, the 4090 reaches around 4 Gkeys/s. So far, I haven’t found specific data on the RTX 5000 series for this niche test. However, I came across a post on the Hashcat forum suggesting that in one mode, the 5090 might be up to three times faster than the 4090. The best way to get a definitive answer is probably to ask on that forum—some users likely already have 5000-series GPUs, whether it's a 5090, 5080, or even a 5070.

RTX 5090: ~9600 h160/s when searching a specific key interval. Or, 27 GK/s if you only want to look at the public keys

[Desyationer]

I'm curious about how much the RTX 5090 outperforms the RTX 4090 in Bitcoin private key cracking speed. In programs like bitcrack, keyhunt, and vanbitcracken, the 4090 reaches around 4 Gkeys/s. So far, I haven’t found specific data on the RTX 5000 series for this niche test. However, I came across a post on the Hashcat forum suggesting that in one mode, the 5090 might be up to three times faster than the 4090. The best way to get a definitive answer is probably to ask on that forum—some users likely already have 5000-series GPUs, whether it's a 5090, 5080, or even a 5070.

RTX 5090: ~9600 h160/s when searching a specific key interval. Or, 27 GK/s if you only want to look at the public keys

Almost 10! – An impressive result! It turns out that the 5090 surpasses the 4090 by almost 2.5 times. But why? After all, the difference between the 4090 and 3090 wasn’t that significant, just like between the 2080 Ti and 1080 Ti. At the very least, it's comparable to the difference between the 1080 Ti and 4090. I’ve had experience with the 1080 Ti, 2080 Ti, 3080 Ti, and now the 4090, but I’ve never seen such drastic jumps—at most, a performance increase of about 1.5 times for mkeys between 3080ti and 4090

[kTimesG]

Almost 10! – An impressive result! It turns out that the 5090 surpasses the 4090 by almost 2.5 times. But why? After all, the difference between the 4090 and 3090 wasn’t that significant, just like between the 2080 Ti and 1080 Ti. At the very least, it's comparable to the difference between the 1080 Ti and 4090. I’ve had experience with the 1080 Ti, 2080 Ti, 3080 Ti, and now the 4090, but I’ve never seen such drastic jumps—at most, a performance increase of about 1.5 times.

No, it's not surpassing 2.5 times. An RTX 4090 reaches 6.9 GK/s. IDK what you're using, but Keyhunt-CUDA supposedly goes beyond 6 GK/s, and that one isn't even fully optimized.

RTX 4090: 128 SM cores
RTX 5090: 170 SM cores

If we compare SM by SM speed increase (for computing and hashing BTC pub keys), it's somewhere around a 10% speedup. So RTX 5090 has no dramatic effect, just more SMs and more power requirements.

The bottleneck is the SHA and RIPEMD hashing, which eat up most of the time. I didn't look if there are ways to improve those - if somehow they can be transformed to multiplications instead of logical operations, the speed may increase.

[nomachine]

Both SHA-256 and RIPEMD-160 can be parallelized using SIMD instructions like AVX2 or AVX-512 for CPUs. Check the cpuminer GitHub repository.

For extreme performance, ASICs or FPGAs are designed to perform hashing at extremely high speeds. Using GPUs for this is like throwing money out the window.

[bibilgin]

Then post the SHA-256 of the public key of that address, no need for any bets to prove you own the private key for that address. You can't? Oh well, that's probably because you would need 40 RTX 4090 GPUs running for 2 weeks at a rate of 7 Gkeys/s in order to stumble upon such an address with 58 matched bits. But I'm sure that with your super-quick budget GPUs at your disposal you got it in like, 2 hours.

You are the one who thinks I don't have the key.

I have no intention of proving it to you.
But I am sending the SHA256 of the Hex code for others to see.

72f170a0d41c6199029510a7c09b297c5e2e499020889ba298a4880e0310dc05

So that you can think about PROBABILITIES without using the word "I don't think so" again.

I would love to have a hardware like that, 40 RTX 4090 GPU with 7 Gkeys/s speed that you mentioned. But it doesn't exist.

Maybe I found it as a result of my calculation on PROBABILITY? If the answer to that is "I don't think so", forget about it. Go fishing. lol

[kTimesG]

You are the one who thinks I don't have the key.

I have no intention of proving it to you.
But I am sending the SHA256 of the Hex code for others to see.

72f170a0d41c6199029510a7c09b297c5e2e499020889ba298a4880e0310dc05

How about you stop trolling? You're also contradicting yourself lol, you say you have no intention to prove anything (I wonder why that is?) and immediately proceed to post something that should prove you do have the key.

I would love to apologize to you in the most honest way possible, but first you should proof that you own the key. You do realize that posting the SHA-256 of the public key does not disclose the actual private key, right? But it just allows us to actually verify you are not simply bullshitting around here?

The SHA you posted belongs to BTC address 1DYhRqregvyqwZG7AYaCGpZucsUgJFkRgt - what does this have to do with anything?

[bibilgin]

How about you stop trolling? You're also contradicting yourself lol, you say you have no intention to prove anything (I wonder why that is?) and immediately proceed to post something that should prove you do have the key.

I would love to apologize to you in the most honest way possible, but first you should proof that you own the key. You do realize that posting the SHA-256 of the public key does not disclose the actual private key, right? But it just allows us to actually verify you are not simply bullshitting around here?

The SHA you posted belongs to BTC address 1DYhRqregvyqwZG7AYaCGpZucsUgJFkRgt - what does this have to do with anything?

Are you seriously writing your articles in CHAT GPT?

Do you also have a reading problem?

1- I shared the SHA256 of the hex code.

2- Actually I don't want to share it with you but I shared it so that other friends can see that it is wrong to say I DON'T THINK.

[kTimesG]

How about you stop trolling? You're also contradicting yourself lol, you say you have no intention to prove anything (I wonder why that is?) and immediately proceed to post something that should prove you do have the key.

I would love to apologize to you in the most honest way possible, but first you should proof that you own the key. You do realize that posting the SHA-256 of the public key does not disclose the actual private key, right? But it just allows us to actually verify you are not simply bullshitting around here?

The SHA you posted belongs to BTC address 1DYhRqregvyqwZG7AYaCGpZucsUgJFkRgt - what does this have to do with anything?

Are you seriously writing your articles in CHAT GPT?

Do you also have a reading problem?

1- I shared the SHA256 of the hex code.

2- Actually I don't want to share it with you but I shared it so that other friends can see that it is wrong to say I DON'T THINK.

OK, I'm tired dealing with you and your useless posts. But you can be proud of one thing, since you failed to prove that you own a made-up address - you're now the first user in my ignore list. You had too many chances to simply post the SHA of the pubKey, but it's clear that you have zero (squat) clues about what I asked, lol. Happy hashing.

[Desyationer]

Almost 10! – An impressive result! It turns out that the 5090 surpasses the 4090 by almost 2.5 times. But why? After all, the difference between the 4090 and 3090 wasn’t that significant, just like between the 2080 Ti and 1080 Ti. At the very least, it's comparable to the difference between the 1080 Ti and 4090. I’ve had experience with the 1080 Ti, 2080 Ti, 3080 Ti, and now the 4090, but I’ve never seen such drastic jumps—at most, a performance increase of about 1.5 times.

No, it's not surpassing 2.5 times. An RTX 4090 reaches 6.9 GK/s. IDK what you're using, but Keyhunt-CUDA supposedly goes beyond 6 GK/s, and that one isn't even fully optimized.

RTX 4090: 128 SM cores
RTX 5090: 170 SM cores

If we compare SM by SM speed increase (for computing and hashing BTC pub keys), it's somewhere around a 10% speedup. So RTX 5090 has no dramatic effect, just more SMs and more power requirements.

The bottleneck is the SHA and RIPEMD hashing, which eat up most of the time. I didn't look if there are ways to improve those - if somehow they can be transformed to multiplications instead of logical operations, the speed may increase.

I’m curious, which program manages to achieve almost 7Gkey on a 4090? In the most optimized version for CUDA 12.6, I get less than 5, and in older versions like BitCrack or KeyHunt-CUDA, the performance is even lower—around 4 or less.

[Akito S. M. Hosana]

Both SHA-256 and RIPEMD-160 can be parallelized using SIMD instructions like AVX2 or AVX-512 for CPUs. Check the cpuminer GitHub repository.

How to separate RIPEMD-160 8way (last version) from chpuminer to work in VanitySearch, Keyhunt or  Cyclone 

[nomachine]

Both SHA-256 and RIPEMD-160 can be parallelized using SIMD instructions like AVX2 or AVX-512 for CPUs. Check the cpuminer GitHub repository.

How to separate RIPEMD-160 8way (last version) from chpuminer to work in VanitySearch, Keyhunt or  Cyclone  

This is very similar to a CPU miner version:
https://github.com/Dookoo2/AVX2-SHA256-and-RIPEMD160/tree/main/ripemd160_avx2

Since it processes 8 hashes simultaneously, the scripts must be adapted to generate keys in batches of 8. (not very easy for beginners)

Each thread gets a copy of an initial key, modifies it, and hashes it:

Code:

#pragma omp parallel
{
    int threadId = omp_get_thread_num();
    unsigned char keysBatch[8][64] = {0};
    unsigned char hashesBatch[8][20];

    for (int i = 0; i < 8; ++i) {
        memcpy(keysBatch[i], initialKeyBytes, 32);
        incrementByteArray(keysBatch[i], 32, threadId * 100 + i);
    }

    ripemd160avx2::ripemd160avx2_32(
        keysBatch[0], keysBatch[1], keysBatch[2], keysBatch[3],
        keysBatch[4], keysBatch[5], keysBatch[6], keysBatch[7],
        hashesBatch[0], hashesBatch[1], hashesBatch[2], hashesBatch[3],
        hashesBatch[4], hashesBatch[5], hashesBatch[6], hashesBatch[7]
    );

    std::cout << "Thread " << threadId << " computed hashes\n";
}

The method leverages AVX2 and OpenMP for high-speed RIPEMD-160 hashing.

You can even create a shared library (.so) from the optimized AVX2 RIPEMD-160 C++ implementation and use it in Python instead of hashlib.  

[Akito S. M. Hosana]

Thanks.
Does anyone know what's happening with privatekeys.pw ?
I can't access it at all.

[southrockstar]

Thanks.
Does anyone know what's happening with privatekeys.pw ?
I can't access it at all.

My two cents: someone it trying to find the private key for puzzle #68 by scanning and scraping the web pages. lol

[bibilgin]

OK, I'm tired dealing with you and your useless posts. But you can be proud of one thing, since you failed to prove that you own a made-up address - you're now the first user in my ignore list. You had too many chances to simply post the SHA of the pubKey, but it's clear that you have zero (squat) clues about what I asked, lol. Happy hashing.

Look bro, I guess you are young.

1- You are the one who said I DO NOT THINK I have the key to this wallet.

2- I shared the SHA 256 of the Hex Code and everyone saw it.

3- I will share the HEX code on the day 68 is cracked. You will see the proof and apologize.

4- You or someone else wants to change to another wallet with the prefix 1MVDYgVaSN6. I will provide the proof. YOU will apologize again.

In conclusion, you will apologize in any case. So do not succumb to your EGO any more.

If you want, I can show the hex code to someone I know whose name is on the forum, he will confirm it to everyone and you can APOLOGIZE again.

Now, think twice before responding to my comments. Or do not respond at all.

We could actually get along well, but when you overcome your EGO.

[kTimesG]

Look bro, I guess you are young.

You guess wrong.

1- You are the one who said I DO NOT THINK I have the key to this wallet.

Yes, because this is what I think. Prove me I'm wrong - post the hash!

It takes 2 seconds to hash the compressed public key of the address you "xxxxxxx"-ed and post the SHA-256 hash, instead of continuosly trolling non-sense about who thinks what or what their age is.

You know damn well that posting the SHA256 does NOT enable anyone to derive your precious private key.

But you also know you cannot do that because, well, you don't have the private key. So instead, you post non-sense, like "SHA of hex code" that helps no one with nothing, and proves nothing to anyone.

You are full of shit, essentially.

2- I shared the SHA 256 of the Hex Code and everyone saw it.

Totally useless information. And what the f**k is a "hex code" anyway? Do you even know like basic computer science? You said you studied hashes for many years, but you can't even use normal terminology to let people understand what the hell you are referring to.

3- I will share the HEX code on the day 68 is cracked. You will see the proof and apologize.

4- You or someone else wants to change to another wallet with the prefix 1MVDYgVaSN6. I will provide the proof. YOU will apologize again.

Too little too late. You can prove NOW that you have the key to that address without ever having to bother that anyone can find the private key. Why don't you do it ASAP and I'll apologize.

In conclusion, you will apologize in any case. So do not succumb to your EGO any more.

If you want, I can show the hex code to someone I know whose name is on the forum, he will confirm it to everyone and you can APOLOGIZE again.

Instead of consuming the keyboard any longer with useless attacks, why don't you give us the SHA 256 of the public key, so we can end this stupid debate about EGO and whatever? You started it when you "xxxxx"-ed the freaking key.

Now, think twice before responding to my comments. Or do not respond at all.

We could actually get along well, but when you overcome your EGO.

Yes, I thought about it 10 times before I said that I think you DO NOT have the private key for that address. Here's why:

1. You have stated that you can only hash something like 7 GK/s with all of you equipment
2. It takes 40 RTX 4090 GPUs running for 2 weeks to find such an address.
3. You don't have such equipment, and I'm very suspicious that you used vast.ai to rent 40 RTX 4090 for 2 weeks. You also claimed that such an equipment does not exist.
4. You insist on mathematics that is irrelevant for finding that address.
5. All the other addresses you posted are in the 45-ish bit length prefix match.
6. It takes a mere amount of days to find all of your posted addresses, because their prefix is only in the 45-ish prefix length. THIS IS WHY YOUR ADDRESS WITH KEYS ARE VALID.
6. Meanwhile, you then posted some made-up address (prove me wrong, post the SHA256 of the public key) that has a 58-bit prefix. That is around 1000-2000 times harder to find.
6. You didn't had 40 days to find such a prefix with your equipment.
7. The user who posted those two addresses BEFORE your useless post, posted addresses that had a 48 and 52 bits prefix. They were both superior to your shitty addresses, since they are much harder to find.

Post the SHA-256 hash of the public key, or admit the address was made up.

Stop the shit with EGO and showing it to special people. No one can steal your precious key, but everyine can verify that you actually have the key to that address.

Otherwise, please shut up and go back to your prefix collection routine.

[bibilgin]

......

I am writing briefly.

The Public Key you requested, converted to SHA256.

ffb3b8a632a6dfbaa52c7497d84df8df5d53737c9ed5fdb7dd8387283a88a635

Also, as a note, let me say this.

The number of wallets I found with the prefix 1MVDYgVaSN is 17.  (I do not share these knowingly.)

Now you apologize, we can close the topic.
I expect you to respect my opinion and my PROBABILITY theorem.

First, remember that these discussion topics were started by your answers. I am not the one who wants to argue with you.

I have said it before. "Every brave man eats yogurt differently."
The most important thing is RESPECT.

[mcdouglasx]

......

I am writing briefly.

The Public Key you requested, converted to SHA256.

ffb3b8a632a6dfbaa52c7497d84df8df5d53737c9ed5fdb7dd8387283a88a635

Also, as a note, let me say this.

The number of wallets I found with the prefix 1MVDYgVaSN is 17.  (I do not share these knowingly.)

Now you apologize, we can close the topic.
I expect you to respect my opinion and my PROBABILITY theorem.

First, remember that these discussion topics were started by your answers. I am not the one who wants to argue with you.

I have said it before. "Every brave man eats yogurt differently."
The most important thing is RESPECT.

Just to clarify, the hash is computed on the bytes, as shown in the following example, not directly on the hex string or str, because it would be a different hash.

Code:

import hashlib
import base58

def hash160(pubkey: bytes) -> bytes:
    sha256_pubkey = hashlib.sha256(pubkey).digest()
    print("pubkey (hash):", sha256_pubkey.hex())
    ripemd160 = hashlib.new('ripemd160', sha256_pubkey).digest()
    return ripemd160

def get_addr(pubkey_hex: str) -> str:
    pubkey_bytes = bytes.fromhex(pubkey_hex)
    hash160_bytes = hash160(pubkey_bytes)
    versioned_payload = b'\x00' + hash160_bytes
    checksum = hashlib.sha256(hashlib.sha256(versioned_payload).digest()).digest()[:4]
    address_bytes = versioned_payload + checksum
    return base58.b58encode(address_bytes).decode()

pubkey_hex = "0279be667ef9dcbbac55a06295ce870b07029bfcdb2dce28d959f2815b16f81798"
address = get_addr(pubkey_hex)
print("Legacy:", address)