Search found 110 matches

by Bitweasil
Tue Mar 17, 2009 2:11 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: MD5, and 2 constants
Replies: 14
Views: 9113

Re: MD5, and 2 constants

Does the inner md5 return an ASCII string (32 bytes, 1-0a-f) or a binary string (16 bytes, 0x00-0xff)?
by Bitweasil
Mon Feb 16, 2009 8:20 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers
Replies: 17
Views: 11589

Re: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers

*nods* I'm not aiming at single hash performance. Also, FWIW, InsidePro Extreme GPU is faster on lists of MD5 hashes. They're all Windows only, which doesn't suit my requirements. This is being written, primarily, for my needs. The fact that it's cross-platform (and Linux compatible) is more of a si...
by Bitweasil
Mon Feb 16, 2009 6:03 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers
Replies: 17
Views: 11589

Re: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers

My brute forcer currently doesn't support LM, but I'm working on it. The next release will include: - No limit on # of hashes cracked (I've run 5M through the testing versions) - Substantially better performance on large hash lists - Per-position charset support (different character sets for each po...
by Bitweasil
Tue Feb 10, 2009 9:48 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: What hash to support next?
Replies: 51
Views: 27002

Re: What hash to support next?

By implementing SHA1, you'll also make it possible to actually calculate the file out of a torrent file. A torrent is divided into pieces, each equal in size (Except the last, which is equal to the remainder of the file). A typical piece size is 256 KB. But what if we made a torrent, with pieces 16...
by Bitweasil
Tue Feb 10, 2009 5:37 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: LM Implementation
Replies: 5
Views: 4512

Re: LM Implementation

The tricky bit is in the function, DES() - that's not quite trivial to implement in CUDA yet.
by Bitweasil
Tue Feb 10, 2009 2:57 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: And I was like OMFGWHAT?!
Replies: 25
Views: 42261

Re: And I was like OMFGWHAT?!

Feeling like this is the wrong place to tell you that, but I am looking forward to that release! :) Check some benchmarks over on my forum with the 1M & 5M lists. I'm also working on some additional stuff - not sure if it will go in this release or not, but per-position-charset support (different c...
by Bitweasil
Mon Feb 09, 2009 9:31 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: And I was like OMFGWHAT?!
Replies: 25
Views: 42261

Re: And I was like OMFGWHAT?!

All the other CUDA MD5-Hash crackers work fine for me. I even found one which is not that slow, but also not as fast as BarsWF, called CUDA Multiforcer . Wait for the 0.7 release. :) Speedups are very, very significant on larger hash lists. Like, a few order of magnitude. It's a bit faster on singl...
by Bitweasil
Sun Feb 01, 2009 9:32 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: How high can you get
Replies: 260
Views: 177149

Re: How high can you get

mrb wrote:A 4870 X2 card with the right drivers, and the latest CAL 1.3 SDK will achieve about 2100 Mhash/sec with BarsWF, outperforming every Nvidia card.
Having barrel shifters in hardware helps more than a little bit with most crypto algorithms...
by Bitweasil
Tue Jan 27, 2009 6:07 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: What hash to support next?
Replies: 51
Views: 27002

Re: What hash to support next?

yep, that was the point, very fast :D :D :D personally i'm interested into DES & freeBSD MD5 (as i voted) also BlowFish cause it's very slow, even slower then freeBSD MD5 :( DES is a bit of a challenge to implement on GPUs, as it uses a lot of resources to do it quickly (bitslicing). It may have to...
by Bitweasil
Mon Jan 26, 2009 8:19 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers
Replies: 17
Views: 11589

Re: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers

Yes. There is sufficient hardware allocated for this project. Plus a good amount of GPU power available locally, and I'm working on getting time on a somewhat local Tesla cluster (48 GPUs) for this. The current target is NTLM 1-9 full US charset, plus extensions out to 12+ characters for smaller sub...
by Bitweasil
Mon Jan 26, 2009 7:53 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers
Replies: 17
Views: 11589

Re: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers

I've not kept up with that - what's expected generate time and size on that table?
by Bitweasil
Mon Jan 26, 2009 5:37 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: x
Replies: 2
Views: 3451

Re: MS SQL BarsWF

Post a request over at cryptohaze.com - I'm working on adding algorithms, and my base is already cross-platform (Windows/Linux/Mac).

//EDIT: lol, it stores the uppercase version? What is it with MS & storing uppercase representations of passwords? LM does the same thing.
by Bitweasil
Mon Jan 26, 2009 5:08 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers
Replies: 17
Views: 11589

Re: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers

LM, for what ? when frt finishes creating the huge tables, you won't be needing a bruteforcer anymore. (ah...for 100-99.97% maybe..) Will the LM tables support all possible foreign characters? I believe they are only supporting a few foreign characters. And I need the algorithm anyway for other thi...
by Bitweasil
Mon Jan 26, 2009 3:47 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers
Replies: 17
Views: 11589

Re: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers

Yes, it's slower than BarsWF on a single hash and will likely remain so - it's focus is on performance with common numbers of hashes (in the 10-1000 range) - not extreme performance at either end. What other algorithms would you like to see? I intend to add variable hash length support soon, which w...
by Bitweasil
Mon Jan 26, 2009 6:22 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers
Replies: 17
Views: 11589

Windows/Linux/Mac CUDA MD4/MD5/NTLM brute forcers

http://www.cryptohaze.com/bruteforcers.php Currently supports cracking: MD4, MD5, NTLM. Platforms supported: Windows XP (32 & 64 bit), Windows Vista (32 & 64 bit). Untested on Windows 7 but probably will work. Linux (32 & 64 bit) Mac OS X Requires an nVidia 8000 series or higher GPU, CUDA drivers. P...
by Bitweasil
Thu Jan 22, 2009 11:13 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

Ok my 64 bit system is up now. I got this error: $ ./NTLM-Multiforcer Launching 64-bit binary ./NTLM-Multiforcer-64: /usr/lib/gcc/x86_64-pc-linux-gnu/4.1.2/libstdc++.so.6: version `GLIBCXX_3.4.9' not found (required by ./NTLM-Multiforcer-64) even if I have the library in the path. That's probably b...
by Bitweasil
Thu Jan 22, 2009 8:12 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

I'll be making a point release later tonight to fix the bug I found. And probably releasing the MD5 version as well, since the bug I found with it affected both. I'm still working on the "How to get my Unix code to compile cleanly under Windows" problem, though. Hopefully a few hours of cursing at V...
by Bitweasil
Thu Jan 22, 2009 3:24 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

And... found a bug. I wouldn't use a non-multiple-of-8 charset right now :) I'll fix it and release a new version this evening. ... and maybe release a MD5 multiforcer while I'm at it. Still working on the Windows code - Visual Studio 2008 doesn't like my C code and doesn't include a lot of headers ...
by Bitweasil
Thu Jan 22, 2009 2:11 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

New version released: 32/64 bit Linux binaries (with launcher): http://www.cryptohaze.com/releases/NTLM-Multiforcer-Linux-0.5.tar.bz2 Support at http://www.cryptohaze.com/forum/ I'll write up more docs later - going bowling now. Performance is right around 25B hashes per second on a GTX260 with a ha...
by Bitweasil
Wed Jan 21, 2009 9:04 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

I'm going to benchmark - I just haven't finished that block of code yet.
by Bitweasil
Wed Jan 21, 2009 8:07 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

I really don't understand why people still insist on 32-bit OSes...

I also need to finish my benchmarking code to tune block size, as I suspect my current GTX260 optimized settings would not work well on an 8600.
by Bitweasil
Wed Jan 21, 2009 7:33 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

synthesis wrote:So I will change my Geforce 7950 on my Gentoo desktop pc with a 8xxx one to try your code :D
Are you running a 32-bit or 64-bit install? If it's 64-bit, I could send you some code to play with currently - it's not quite done, but it's very usable right now.
by Bitweasil
Wed Jan 21, 2009 6:16 am
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

Found more performance, need to write a chunk of benchmark code (as right now it's tuned for things like a GTX260, and won't play nice with a GUI on anything much slower than an 8800 - testing on my 8600 was somewhere between funny and painful with the GTX260 block/thread counts). I'm running about ...
by Bitweasil
Mon Jan 19, 2009 2:05 pm
Forum: BarsWF, Cryptography, Security, GPGPU and supercomputing
Topic: CUDA NTLM multi-hash bruteforcer
Replies: 29
Views: 38064

Re: CUDA NTLM multi-hash bruteforcer

Even if the project name is "NTLM Multiforce" do you plan to support other algorithms or make similar programs to crack md5, *nix md5 etc? :D It would be great to study these sources... Yes, I intend to include support for other algorithms. They're likely not going to be nearly as optimized as Bars...