BitCrack - A tool for brute-forcing private keys

[NotATether]

Can possible port BitCrack full version to BitCrack Lite version for small version cpu and port again to mobile device, android for convert old mobile device unused to be BitCrack android

That's a bad idea. Continuous CPU usage of phones while they're charging will deplete the battery's maximum capacity. Besides a phone doesn't have enough memory to run it anyway, even when I use -b 1 -t 32 -p 1 I'm still using about 100MB+ of video card memory (not RAM, remember that. Embedded GPU chips don't have that much VRAM).

[willi9974]

Interesting tool. Is it possible to feed the tool with a list of millions of BTC addresses and the tool will try to find a private key to one of the BTC addresses in the list? Or is it only possible to specify an address and limit the search radius like with the puzzle searches e.g. puzzle #64?

[fxsniper]

Interesting tool. Is it possible to feed the tool with a list of millions of BTC addresses and the tool will try to find a private key to one of the BTC addresses in the list? Or is it only possible to specify an address and limit the search radius like with the puzzle searches e.g. puzzle #64?

BitCrack is very very great tools
but not yet have power enough for scan 265 bit key

yes, you can try use BitCrack with real bitcoin I try it , it works
you can try on list of million bitcoin have balance
I testing by start from not so far from private key millions key, bitcrack found key (key for test by generate)
but problem is range between bitcoin is over 1 quintilliard not easy to found key

bitcrack no limited scan, unlimited,
problem is range too high and very very large too much

for puzzle #64 bitcrack still scan very long time a half year
because individual scan for group scan is just 1/9 of all range

[sp_]

Well, no one is going to mine a bitcoin using the cards that are used for the puzzle so you have to break it down mining alt coins like eth, and convert earnings to bitcoin to get an estimate. Pretty sure more is made from mining AND it's guaranteed (if in a eth pool, or solo if you have enough juice) versus you may run cards 6 months, a year, 2 years, and never find puzzle key(s).

The problem with ethereum mining is that the memory requrements need cards with over 4gigs of memory. <4gb cards have few profitable coins to choose from.
A rx 570 4GB card earn around $1.7 a day mining vertcoin with the spminer,  (570 khash)
a rx 570 8GB card earn around $2.90 a day mining ethereum.

A few months ago the profit was as low as $0.25 and could barely cover the power cost.

- Bitcrack doesn't need an internet connection
- Bitcrack provide anonymous mining revenue in bitcoin

The puzzle transactions are hard, but not impossible to solve. Some mining farms have 100 000 2gb/3gb/4gb cards that cannot mine ethereum anymore.

how long will it take to crack puzzle #64 with
100 000 1060 gtx 3gb@366MKeys with the sp-mod?

2^64 / (366 000 000 * 100 000) = 504009 seconds (140 hours)

[pacnload]

Is there any way to generate all privat keys with compressed and uncompressed and save them to txt file with 100MB Splited size?

starting from:
000000000000000000000000000000000000000000000001A838B13505B26867
ending on:
000000000000000000000000FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF

[fxsniper]

Is there any way to generate all privat keys with compressed and uncompressed and save them to txt file with 100MB Splited size?

starting from:
000000000000000000000000000000000000000000000001A838B13505B26867
ending on:
000000000000000000000000FFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFFF

Just using python script running you can find easy

I know you want to use very fast key generate like bitcrack
not yet have someone help to modify to save to file
 

[pacnload]

Just using python script running you can find easy

I know you want to use very fast key generate like bitcrack
not yet have someone help to modify to save to file
 

i have not mutch knowledge about python

[WanderingPhilospher]

Just using python script running you can find easy

I know you want to use very fast key generate like bitcrack
not yet have someone help to modify to save to file
 

i have not mutch knowledge about python

It really doesn't matter.  You would have to have tens of thousands of GPUs or CPUs just to get through that range. Next problem is I/O...writing to disk/file.  You could have a GPU that searches 1000 MKey/s but when you use it to write every key to file it will drop to 0 and you will be lucky to write maybe 200k in a minute.  so divide 2^160 by 200k = 7.30750818665451459101842416358141509827966271488e+42 minutes with one GPU.

Let's say you got lucky and 1,000,000 people gifted you a GPU. So now you have 1 million GPUs. Now divide 2^160 / 200k*1000000 = 7.30750818665451459101842416358141509827966271488e+36 minutes.  That's a whole lot of years, right?

[NotATether]

The puzzle transactions are hard, but not impossible to solve. Some mining farms have 100 000 2gb/3gb/4gb cards that cannot mine ethereum anymore.

how long will it take to crack puzzle #64 with
100 000 1060 gtx 3gb@366MKeys with the sp-mod?

2^64 / (366 000 000 * 100 000) = 504009 seconds (140 hours)

I'm not really interested in bringing this up anymore now that I have my own fork but let me just point out that as long as there's only a windows version then you're never going to be able to pool that many GPUs together, because NVIDIA/AMD drivers and PCI cards limit the amount of GPUs you can attach at once and there is no MPI-kind of code for serving other machines Bitcrack shares like Kangaroo has (as primitive as it may be).


PS. I just got a GPU attached to my Windows server VM with PCI passthrough. I should be able to make a compiled Bitcrack now...

[fxsniper]

on think about puzzle #64 and other
private key may be on some where people not yet scan

calculate time scan 1 day from start and skip it may be save time
I think every start from start point will be scan same

some where not from start and middle(half)

[WanderingPhilospher]

The puzzle transactions are hard, but not impossible to solve. Some mining farms have 100 000 2gb/3gb/4gb cards that cannot mine ethereum anymore.

how long will it take to crack puzzle #64 with
100 000 1060 gtx 3gb@366MKeys with the sp-mod?

2^64 / (366 000 000 * 100 000) = 504009 seconds (140 hours)

I'm not really interested in bringing this up anymore now that I have my own fork but let me just point out that as long as there's only a windows version then you're never going to be able to pool that many GPUs together, because NVIDIA/AMD drivers and PCI cards limit the amount of GPUs you can attach at once and there is no MPI-kind of code for serving other machines Bitcrack shares like Kangaroo has (as primitive as it may be).

I can run 18 GPUs on a Windows machine. How many more do you need on 1 PC? 

How many can you run on one Linux machine?

[fxsniper]

some idea

Can someone Fork BitCrack to version for puzzle #64 only

BitCrackPuzzle64.exe

some detail
Create database from firebase free version (1GB storage)
pool scanned database

split scan from 100,000,000 keys to 1 slot  (or if too much may be = 10,000,000)
so, will got 92233720369 slot
random 1 from 92233720369 slot and scan

if random slot number not have store on database then begin scan that slot
but if random slot number have store on database already, then scan new one slot

second slot will start from 9223372036854775808+100000000 = 9223372036954775808
slot number 999 will = 9223372036854775808+100000000*999 = 9223372136654775808

when scan complete slot sent request to record slot on firebase database
that mean that slot already scan

when other people scan will be check same system and skip slot not found to ther slot

options
make option -slot (number)  for can custom order to scan

if number slot have on record
give error message "slot 999 already scanned"

firebase c++
https://firebaseopensource.com/projects/firebase/quickstart-cpp/

just sn idea not serious

now we work some same thing with burn resource out  (many people do same job same keyspace but not fould
this system will be help to skip address that already scan

[WanderingPhilospher]

some idea

split scan from 100,000,000 keys to 1 slot  (or if too much may be = 10,000,000)
so, will got 92233720369 slot
random 1 from 92233720369 slot and scan

if random slot number not have store on database then begin scan that slot
but if random slot number have store on database already, then scan new one slot

second slot will start from 9223372036854775808+100000000 = 9223372036954775808
slot number 999 will = 9223372036854775808+100000000*999 = 9223372136654775808

when scan complete slot sent request to record slot on firebase database
that mean that slot already scan

when other people scan will be check same system and skip slot not found to ther slot

options
make option -slot (number)  for can custom order to scan

if number slot have on record
give error message "slot 999 already scanned"

firebase c++
https://firebaseopensource.com/projects/firebase/quickstart-cpp/

just sn idea not serious

now we work some same thing with burn resource out  (many people do same job same keyspace but not fould
this system will be help to skip address that already scan

Why reinvent the wheel? There is already a pool with 11-15% of the ranges scanned.  The pool is broken down into 2^36 ranges but you can go up to 2^40 range if you have higher end GPUs so time is not wasted doing small range and having to regen up for the next range.

[willi9974]

where is the pool? what software is needed for this?

[NotATether]

How many can you run on one Linux machine?

It's not about how many you can run on one machine, Windows 10 is extremely bad at preserving uptime.

Say sp_ does add the code for pooling Bitcrack one day (though I doubt it), if somebody fires up a cluster of say 100 Windows boxes with 18 GPUs in each, its only a matter of time before Windows restarts itself for an automatic update.

And just like that, you lose hashing power without even being aware of it. To negate this, you must either 1) Buy Windows server on all those boxes (you see, you can't just run it unlicensed because after 6 months of running like that they will start powering themselves off automatically per anti-piracy measures) and make Group Policy halt all the updates, and this already puts you at a net loss because Windows Server already costs a couple hundred per key.

Or 2) you pay for some third party monitoring software that alerts you if a box restarts, which also puts you at a net loss before you even started the search.


Meanwhile, in VS2019 land...



These CUDA and OpenCL include directories are becoming a real problem to deal with, I'll see if there's any way to abstract them in the solution properties (and not in the project properties or I will have to do this 18 times).

[WanderingPhilospher]

Windows restarts itself for an automatic update.

And just like that, you lose hashing power without even being aware of it. To negate this, you must either 1) Buy Windows server on all those boxes (you see, you can't just run it unlicensed because after 6 months of running like that they will start powering themselves off automatically per anti-piracy measures) and make Group Policy halt all the updates, and this already puts you at a net loss because Windows Server already costs a couple hundred per key.

Nah, first thing you do is turn off auto updates. Increase pagefile, adjust power options/sleep mode, auto power up after powerloss, startup program upon restart, etc.

You can have free version of windows, only thing you can't do is adjust personal settings, which I'm not sure what that is.

Anyway, been mining for years with zero issues with Windows.  More problems with servers/pools/payouts from pools, etc. and the occasional driver issue.

But I agree with you in the fact it is better to have everyone be able to use program/pool, Windows and Linux users.  I really do need to learn more of Linux versus having the Ubuntu subsystem on my Windows machines.

[WanderingPhilospher]

where is the pool? what software is needed for this?

http://ttdsales.com/64bit/login.php
You can use Bitcrack.

[pacnload]

My thinking was to create a lot of private keys from hex and save them to files then you only need to search with some programm all the files to find the BTC Adress inside this file and next to the adress is written the wif key to access the BTC Adress.

First i have try to crawl the website privatekeys.pw but they block crowling then i found a script like privatekeys.pw but it also wasnt possible to save this websites because of ajax that must load first then i have installed it on my own server but also no way because of ajax. Last i wanted to use iMacros to let klick him the website and download it and go to next website but there are to mutch data.

i know to scan the full hex codes it is impossible maybe it is impossible one time. google has created in cooperation with nasa a quantum computer maybe they can do this.

i also have breaked my head to look as a way to get the solution to the puzzle all i have found out is that the numbers 1 3 7 and Maybe letter F has any important meaning. i also found the numbers 2 4 8 1 but that cames from the number 2^30 2^31 and something:

2^1 = 2
2^2 = 4
2^3 = 8
2^4 = 16

2^31 = 2147483648
2^32 = 4294967296
2^33 = 8589934592
2^34 = 17179869184

also if you compare the 2^1 numbers with the solution you will find out that is every time an -1 counting

2^1 = 2  = 0000000000000000000000000000000000000000000000000000000000000001
2^2 = 4 = 0000000000000000000000000000000000000000000000000000000000000003
2^3 = 8 = 0000000000000000000000000000000000000000000000000000000000000007
2^4 = 16 = 0000000000000000000000000000000000000000000000000000000000000008 wrong but F (maybe Fail)
2^5 = 32 = 0000000000000000000000000000000000000000000000000000000000000015 here is the 16 from 2^4 -1
2^6 = 64 = 0000000000000000000000000000000000000000000000000000000000000031 here is the 32 from 2^6 -1
2^7 = 128 = 000000000000000000000000000000000000000000000000000000000000004c
2^7 = 128 = 0000000000000000000000000000000000000000000000000000000000000076 (in decimal) if you show in the last keys you have the 1 3 and here in decimal ist the 7

like you see we found the 1 3 7 and here again in the description with hex:

1...1
2...3
4...7
8...f
10...1f

maybe someone see something i dont see but i dont think that the created of the puzzle wanted to scan every hex to find the solution.

[_Counselor]

My thinking was to create a lot of private keys from hex and save them to files then you only need to search with some programm all the files to find the BTC Adress inside this file and next to the adress is written the wif key to access the BTC Adress.

First i have try to crawl the website privatekeys.pw but they block crowling then i found a script like privatekeys.pw but it also wasnt possible to save this websites because of ajax that must load first then i have installed it on my own server but also no way because of ajax. Last i wanted to use iMacros to let klick him the website and download it and go to next website but there are to mutch data.

Do you imagine how much space is required for such amount of keys? Even if you will store just one byte from address, puzzle 64 space will require 2^63 bytes, or 8 388 608 Terabytes.

All that databases like privatekeys.pw generates key lists on the fly, and don't stores them.

[sp_]

Say sp_ does add the code for pooling Bitcrack one day (though I doubt it), if somebody fires up a cluster of say 100 Windows boxes with 18 GPUs in each, its only a matter of time before Windows restarts itself for an automatic update.

Bitcrack is a simple terminal program. I think it's possible to run it on linux in Wine.