Original file [Request]

[ItsLeon15]

I have a tominecon.7z file but I'm not sure if it's the original one so I was wondering if you might, have it? I also agree with unnaturalistic as this should be multi-threaded as well as less IO reads.

[Bonnie39]

You can use s3md5 to check if the ETag of your tominecon.7z matches that of the original 2011 version. Make sure to specify the multipart chunk size as 64MB when running s3md5. The expected outcome is e10d225cd041b43221a99ebea3c764e8-2.

I am currently working on making this program utilize multiple threads (and allowing the user to specify the amount of threads they wish to use). An update should be out soon.

[ItsLeon15]

Thanks, I'm getting this result which seems to be the same as the etag you mentioned. Been running it for a while now and have only gotten to 805,000, I have a beefier machine lying around which has 48 threads and 512GB RAM which I would want to use but it's Linux not Windows which is a shame.

root@ubuntu:~# ./s3md5 64 tominecon.7z
e10d225cd041b43221a99ebea3c764e8-2

[Bonnie39]

I just pushed an update that allows you to specify the number of threads you want to use when running this. This multiplies brute forcing speed by the number of threads you choose to use.

Unfortunately since I used Windows-specific libraries, I can't directly compile this to run on Linux and would have to make a few tweaks. I do have plans to make a Linux version eventually though.

[natefabian18]

I was just looking into this and found this project. Im a linux person and my c++ skills havent been touched since I was back in college but I have some computing power I can donate so if you can get a linux compatible version around I would appreciate it.

[ItsLeon15]

Managed to get it to work on Linux on my machine and it is going really quickly around 1,000 per 6 seconds (Using 48 threads). If you want, I can fork the repository with the working code?

Edit 1: No password found between 0 - 1,680,000 Edit 2: No password found between 1,680,000 - 2,745,780 Edit 2: No password found between 2,745,780 - 4,500,000

[natefabian18]

I would say throw up the fork. Should we open an issue to keep track of known not it ranges.

[Bonnie39]

I'm working on an update that will allow the user to specify where to start so that known incorrect passwords can be skipped

[SondreDj]

You should also spread out the threads so they dont end up doing the same work multiple times. It seems to me that the threads start only 1 number apart. (I might be wrong here, im not that good at c++)

[Bonnie39]

You should also spread out the threads so they dont end up doing the same work multiple times. It seems to me that the threads start only 1 number apart. (I might be wrong here, im not that good at c++)

Each thread starts 1 number apart but the next combination each thread tries increments by the number of threads.

So thread 1 tries 000, thread 2 tries 001, thread 3 tries 002, etc. Then in the next pass thread 1 tries 003, thread 2 tries 004, and thread 3 tries 005.

[theogbob]

I highly recommend you guys check out my post here and with my current setup and the script i provided im able to get around 3 million passwords per minute guessed and doing some GPU cloud rental (expensive so i didnt run it for long, 18 4090's) i was able to get 43 million per minute. Id also recommend checking out my issue here to see how we can improve this script

[megapro17]

expensive so i didnt run it for long, 18 4090's

around 3 million passwords per minute guessed

What a fuck. Single 4090 can get 2400000 hashes per minute using hashcat

[agent932]

can you post a link to the file?