Keyhunt - development requests - bug reports

[ccinet]

What do you expect, you run a program find a key and getting a millionair ? RTFM! No pain no gain. This is nothing for "From Zero To Hero" fanboys. Do your maths. RTFM.

Take it easy. I am a Newbie getting to know Keyhunt. I am at 5 Petakeys/s right now. So what if I get the millions? what's it to you?

5 Petakeys wow!
What are you hunting puzzle 130 or 66?
What range are you scanning?

[1714252715]

1714252715

[1714252715]

1714252715

[AndrewWeb]

5 Petakeys wow!
What are you hunting puzzle 130 or 66?
What range are you scanning?

I am trying out 7 Petakeys/s now and think I need some more RAM

For now puzzle 130. I have BitCrack running for puzzle 66 on a another computer.

I am trying out different ranges, it's not easy to decide what range. I have to read some more about what ranges are more probable than others ? do you have any suggestions what or where to read ? links ?

[ccinet]

5 Petakeys wow!
What are you hunting puzzle 130 or 66?
What range are you scanning?

I am trying out 7 Petakeys/s now and think I need some more RAM

For now puzzle 130. I have BitCrack running for puzzle 66 on a another computer.

I am trying out different ranges, it's not easy to decide what range. I have to read some more about what ranges are more probable than others ? do you have any suggestions what or where to read ? links ?

Personally I believe that whoever made the puzzles did not use a random method, I think the location of the known puzzles in relation to their rank demonstrates this. A while ago I did a linear regression study and I was surprised by the result. I expected the average difference between the predicted and actual location to be approximately 50%, which is what is expected for a linear regression prediction on random data, however the average is 27.81%

5 Petakeys wow!
What are you hunting puzzle 130 or 66?
What range are you scanning?

I am trying out 7 Petakeys/s now and think I need some more RAM

For now puzzle 130. I have BitCrack running for puzzle 66 on a another computer.

I am trying out different ranges, it's not easy to decide what range. I have to read some more about what ranges are more probable than others ? do you have any suggestions what or where to read ? links ?

Personally I believe that whoever made the puzzles did not use a random method, I think the location of the known puzzles in relation to their rank demonstrates this. A while ago I did a linear regression study and I was surprised by the result. I expected the average difference between the predicted and actual location to be approximately 50%, which is what is expected for a linear regression prediction on random data, however the average is 27.81%

I dare to predict that puzzle 66 is at  ~72% of its range.

[albert0bsd]

Personally I believe that whoever made the puzzles did not use a random method, I think the location of the known puzzles in relation to their rank demonstrates this. A while ago I did a linear regression study and I was surprised by the result. I expected the average difference between the predicted and actual location to be approximately 50%, which is what is expected for a linear regression prediction on random data, however the average is 27.81%

I dare to predict that puzzle 66 is at  ~72% of its range.

💩💩💩

Bullshit!

The author already said that it was the result of a random keys padding with bits to match the expected range.

[Baskentliia]

hi ALBERTO;

For N value in Keyhunt Program
Example: Nx100000000 = 4294967296 keys
Does it mean that it scans every 4294967296 key sequentially? Is it true ?
It selects a random key and scans 4294967296 wallets sequentially, and then it selects a random key again and scans 4294967296 wallets sequentially.

Did I understand correctly? Does it work like this? Keyhunt N value?

[AndrewWeb]

I dare to predict that puzzle 66 is at  ~72% of its range.

Ok I'll try that. I to think it's somewhere around there 

I now got 67 Pkeys/s hunting Puzzle 130 and 40 Mkeys/s on Puzzle 66

I need a new computer to get 1 or 2 Ekeys/s like this guy,

https://bitcointalk.org/index.php?topic=5322040.msg62268526#msg62268526

[ccinet]

I dare to predict that puzzle 66 is at  ~72% of its range.

Ok I'll try that. I to think it's somewhere around there  

Hey if you find something don't forget about me  
1DvdiYvRr7pzHsYRJiXYdroQNZUqKxLAzf

I now got 67 Pkeys/s hunting Puzzle 130 and 40 Mkeys/s on Puzzle 66

I need a new computer to get 1 or 2 Ekeys/s like this guy,

https://bitcointalk.org/index.php?topic=5322040.msg62268526#msg62268526

With bsgs mode it is only a matter of ram to reach those speeds, I don't see it as easy with puzzle 66.
Anyway, I don't think those who solved 120 and 125 have done a full range scan or used a single computer, it would be interesting to know what program and what values ​​they used, but obviously this must be confidential information.
Unfortunately I have the problem that I like numbers and I understand magnitudes, and honestly every time I see the infinity that we face I want to look for my weapon!! ha ha

Personally I believe that whoever made the puzzles did not use a random method, I think the location of the known puzzles in relation to their rank demonstrates this. A while ago I did a linear regression study and I was surprised by the result. I expected the average difference between the predicted and actual location to be approximately 50%, which is what is expected for a linear regression prediction on random data, however the average is 27.81%

I dare to predict that puzzle 66 is at  ~72% of its range.

💩💩💩

Bullshit!

The author already said that it was the result of a random keys padding with bits to match the expected range.

Hello Alberto.
However, is the author willing to provide more information? I don't think so, otherwise I would have exposed more public keys. What it has done is increase the incentive so that more people hit the security.
After all, what does the author want to prove? I imagine that the security of bitcoin...

[albert0bsd]

what does the author want to prove? I imagine that the security of bitcoin...

Yes he want to prove the bitcoint security

in his own words:

I am the creator.

You are quite right, 161-256 are silly.  I honestly just did not think of this.  What is especially embarrassing, is this did not occur to me once, in two years.  By way of excuse, I was not really thinking much about the puzzle at all.

I will make up for two years of stupidity.  I will spend from 161-256 to the unsolved parts, as you suggest.  In addition, I intend to add further funds.  My aim is to boost the density by a factor of 10, from 0.001*length(key) to 0.01*length(key).  Probably in the next few weeks.  At any rate, when I next have an extended period of quiet and calm, to construct the new transaction carefully.

A few words about the puzzle.  There is no pattern.  It is just consecutive keys from a deterministic wallet (masked with leading 000...0001 to set difficulty).  It is simply a crude measuring instrument, of the cracking strength of the community.

Finally, I wish to express appreciation of the efforts of all developers of new cracking tools and technology.  The "large bitcoin collider" is especially innovative and interesting!

Now please if you want to continue the discution of that topic please use that "Re: Bitcoin puzzle transaction ~32 BTC prize to who solves it" Topic, not this, this only for Keyhunt related issues.

Any new post that is not keyhunt will be reported as Offtopic or SPAM

hi ALBERTO;

For N value in Keyhunt Program
Example: Nx100000000 = 4294967296 keys
Does it mean that it scans every 4294967296 key sequentially? Is it true ?
It selects a random key and scans 4294967296 wallets sequentially, and then it selects a random key again and scans 4294967296 wallets sequentially.

Did I understand correctly? Does it work like this? Keyhunt N value?

Yes it is exactly like you describe above

[Baskentliia]

hello alberto
In keyhunt addresses mode, we can change the -n value as we wish and perform sequential random searches of any size we want. However, in keyhunt bsgs mode, since this -n value works together with the k value, we cannot exceed a certain number.
What can we do for this?
Example: N 100000000000= It scans 17592186044416 keys sequentially and then randomly selects a range again and scans wallets up to 17592186044416 sequentially. Is it possible to increase this number?
After scanning 1000000000000000000 wallets, I want to choose a random range again and scan 1000000000000000000 wallets sequentially. I hope I was able to explain.

[albert0bsd]

hello alberto
In keyhunt addresses mode, we can change the -n value as we wish and perform sequential random searches of any size we want. However, in keyhunt bsgs mode, since this -n value works together with the k value, we cannot exceed a certain number.
What can we do for this?
Example: N 100000000000= It scans 17592186044416 keys sequentially and then randomly selects a range again and scans wallets up to 17592186044416 sequentially. Is it possible to increase this number?
After scanning 1000000000000000000 wallets, I want to choose a random range again and scan 1000000000000000000 wallets sequentially. I hope I was able to explain.

I don't see what is the problem here, just use the number that you want. Did you already tried?

[Baskentliia]

hello alberto
In keyhunt addresses mode, we can change the -n value as we wish and perform sequential random searches of any size we want. However, in keyhunt bsgs mode, since this -n value works together with the k value, we cannot exceed a certain number.
What can we do for this?
Example: N 100000000000= It scans 17592186044416 keys sequentially and then randomly selects a range again and scans wallets up to 17592186044416 sequentially. Is it possible to increase this number?
After scanning 1000000000000000000 wallets, I want to choose a random range again and scan 1000000000000000000 wallets sequentially. I hope I was able to explain.

I don't see what is the problem here, just use the number that you want. Did you already tried?

For Keyhunt bsgs mod
I want to set -N to a high value but the program freezes because there is not enough RAM.
So for 16 GB ram K max = 1024 N max = 100000000000
I want to increase the N value but I can't.
For example, the value of N
I want to do 400000000000000 but the computer freezes.
Cannot run High N value on low system

[albert0bsd]

OK, i got you, yes the limit is the ram.
Then why no you just select your specific range and run it sequentially even with the small N it should works?

[Baskentliia]

OK, i got you, yes the limit is the ram.
Then why no you just select your specific range and run it sequentially even with the small N it should works?

So I wanted to ask if there is a solution, my friend.
My computer scans 17592186044416 wallets in order, then selects a random order again and scans 17592186044416 wallets in order. I just asked if we can increase the number of wallets scanned.
I want it to scan 100000000000000000 wallets in order, then choose a random order again, and then scan 100000000000000000 wallets again. But since the amount of RAM is not enough, I wanted to ask you. Thank you for your answers. This program is amazing. I hope we will see new versions in the future.

[albert0bsd]

So I wanted to ask if there is a solution

I already told you a possible work around, select a specific range of the number of keys that you want to scan and do it sequentially instead of random.

If that doesn't work for you then nothing will work for you, MAKE YOUR OWN TESTS.

[GTX1060x2]

If anyone is using keyhunt on an AMD processor, try compiling it with AOCC.
I got a +13% performance increase on Zen2 architecture compared to GCC-12, 13 and 14.
It would be interesting to see tests on Zen4 architecture.
https://www.amd.com/en/developer/aocc.html

[Lugh1Man]

Hi all, Alberto,
Has anyone done tests of KeyHunt (and shared the results) with very large amount of ram - say 512GB - 1TB? I'm thinking of doing this myself to at least document the speed gains (relationship between the n and k values, and the speed reported). To my understanding, adding more ram will significantly increase the speed because you can use a higher n and k values. I believe adding more RAM is much more beneficial to adding more CPU power, but of course both are important.
Edit: I'm talking about BSGS mode, of course.

Please delete this if it's off-topic for this thread.

[albert0bsd]

Please delete this if it's off-topic for this thread.

it is not a off-topic.

Yes there are some users who use up to 1TB of ram, as far i remember those get some tens of Exakeys/s but it is not enough for high bit  puzzles

[tkredbaron]

hey Alberto,
thanks for your outstanding work
what about the idea run key hunter on bitminer/antminer hardware?

[albert0bsd]

what about the idea run key hunter on bitminer/antminer hardware?

That is not possible

Miner Hardware only is capable to do sha256.

And key hunt process need sha256, rmd160, eliptic curve operations and  some other math operations.

[FlleOWA]

There are 2 processors installed on the motherboard. The program only works with one. I specify all cores, but only one processor is used. The program can't work with multiple processors?