Topic: VanBitCracken - a program to use for 32 BTC challenge (supports RTX 30xx cards) - page 8. (Read 5673 times)

Btw any idea why program is not working when you try to search for both compressed and uncompressed addresses? Getting error when I try and program stops.

Tnx, now I understand. Had no idea your -r feature does that, will use it from now on, that is exactly what I wanted.

You say "...define a keyspace and it goes randomly through whole keyspace "it" (not sure what you meant by it) on each point, not just creating starting points and going in +1 increments."
I do not understand how this is random. It sounds like you want it to go check each point in the keyspace or do random jumps through the keyspace. I get confused when you say through the whole keyspace. What I have does the following:

top range/keyspace/beginning range you want to start with
800000
so that is a 24 bit top range

You define if you want random keys generated in all of 24 bit range by selecting subrange of 23; now it will generate random points from 8000000 through FFFFFF
if you wanted to search subrange of 16 bit you select subrange of 16; now it will generate random points from 8000000 through 80FFFF

You control how often it generates random points using the -r (rekey) feature.

But this program is designed to generate and regenerate random points in a given keyspace as often as you like.  It will generate your GPU grid size random points and go sequentially from each random point. Once it rekeys, it generates new random points and goes sequentially. Let's say your GPU gets 100 MKey/s; you could put -r 100 and the program would generate new random points every 1 second.

The difference is; bitcrack generates random starting points and goes sequentially until keyspace has been searched; the program eventually ends.

My program just keeps generating random points and it will never end, until you close the program down. It will never start with random points and walk them sequentially until whole keyspace is searched.

Tested program, it's nice. About 2x faster then pika's random version of bitcrack. I would love to have -r random feature here also. If possible that every point that is created is random in defined keyspace, not just starting points that are random like in bitcrack. So you define your keyspace and it goes randomly through whole keyspace it on each point, not just creating starting points and going in +1 increments.

Yes that's correct, say I set a keyspace, it generates random points to search within that keyspace.

I may have what you already want.

I just need to make sure it is what it is.

Please reexplain what you are wanting to do. Is it just to create random points over and over inside a defined keyspace?

If it's something I have or can do, no need for donation, I do it to tinker, to actually get something to work.

I can donate a few dollars in BTC for the random implementation, but sadly I don't have much. 

Yes that is correct!
I think a few others were requesting this above with the -r feature.

So are you just wanting to search in a keyspace of something like 80000:8FFFF and have a program that generates random points all in between that keyspace?

Heres the version of BitCrack that has the source code to generate random points.

Hopefully it helps in speeding things along for you and getting -r implemented.

https://github.com/neutron220/BitCrack

TBH, most of what you suggested is above my skillset.

I am currently looking for source code of the random version of VBC which randomly generates new points, so if you have grid size of 10*100 on your GPU, then the program will create 1000 random points and start searching incrementally from each random point, but you can adjust how often it re randomizes and starts the process all over again. So in theory you could set it to re randomize every second.

But that will be on my task list first. I had a project to create a solo mining pool on an old PC (running geth and acting as server) and see if I could mine some blocks; and I did that. So I have a few other personal tasks to complete and then I'll start on random feature.

Solo mining pool test with actual results; completed:
https://bitcointalksearch.org/topic/can-my-old-pc-run-gethact-as-a-server-and-let-me-solo-minefollow-along-5346020

Eth miner / mining comparison tests up and running with real live results; completed:
https://bitcointalksearch.org/topic/version-2-of-gminer-versus-t-rex-miner-actually-mining-real-eth-on-3070s-5346433

Knocking things off the todo list, one by one.
I can say this, once mining cools down, I'll be back to experimenting with these key finding programs 100%

Masks are hard for users to find the optimal one, no? A bitmask will eventually translate down to a stride amount anyway so maybe instead of a bitmask, there could be some "oscillators" built inside the program each of which implements a different stride amount for each iteration. Kind of like how analog frequency is variable but has different patterns.

awesome program, thank you for your hard work...  Signed up just to post these suggestions...

I know in its current form, its just a start from one point and 1by1 increment....

1) like many the -stride function (aka increment up by 100,000 instead of 1)
2) ability to check for the REVERSE of the hash
3) -random function with some options
      a) ability to also create a "mask" that will increment each position a certain number (for instance 000000001100111000111000110045980813040) would increment each position by the number in that position forwards
          and backwards
      b) again to do the reverse with the above options
      c) if possible restrict the number of duplicates in a row...example... -max-duplicates 3 means in the key no more than 3 of the same letters/numbers in a row... (not sure if this can be controlled with "random" but thought I would ask)

4) ability to input your own hex keys as a file input/stdin
5) on the command line display have the ability to see the status/pause/etc.. kinda like hashcat

Thank you again for your hard work.

stride mode please not work

Where is the sourcecode?

Thanks will try again

You know, I'm not sure if JLP did or did not make any patches for Bitcrack.

For the x and y CUDA grids, that's a good question.  It always seems if you increase the y, the card performs better from a speed standpoint.

I will PM you the Makefile and you can take a look at it.

Download exe and place in a folder. Create a batchfile and place in same folder. 

If you downloaded version 1.1

Batch file:
(remember to save as .bat)

If you downloaded version 1.0
(remember to save as .bat)

Double click batch file and let it run.

Reference with post on thread bitcoin puzzle 100 update

it is good idea for mod output to csv format and have other data information


https://bitcointalksearch.org/topic/m.56696467

Great job. I can write the Makefile for you if it doesn't exist.

So apparently JeanLucPONS was making patches to Bitcrack at one point, is that correct?

The README file has this gem in it:


My question is is there a benefit to being able to adjust the Y grid size? CUDA usually sets the limit for the x and y dimensions to the same number so I'm just wondering if there's a possibility that CUDA is more "tuned" for large Y/small x or small Y/large x or equal dimensions or something like that.