oclhashcat-plus is an awesome tool, but I would love to see support for sha512(unix). This is pretty standard for the shadow file of Linux distributions these days. I have yet to see a GPU cracking tool that supports this algorithm for Stream devices. Thank you for the amazing work you are doing. I hope this makes it into a future release.
|
sha512 (unix) in oclhashcat-plus
|
01-04-2012, 10:10 AM
(01-04-2012, 10:04 AM)wizard Wrote: oclhashcat-plus is an awesome tool, but I would love to see support for sha512(unix). This is pretty standard for the shadow file of Linux distributions these days. I have yet to see a GPU cracking tool that supports this algorithm for Stream devices. Thank you for the amazing work you guys are doing. I hope this makes it into a future release. +1 
01-04-2012, 10:53 AM
the worst problem with sha512 is the lacking support for true 64 bit integers. sha512crypt is even more worse, since it is doing both, lots of iterations and generating huge datablock that need transformation. i dont expect sha512crypt being much faster on gpu than on cpu. maybe x 2 or x 3 at best. i think we have to wait for better 64 bit int support on gpus.
01-04-2012, 11:12 AM
(01-04-2012, 10:53 AM)atom Wrote: the worst problem with sha512 is the lacking support for true 64 bit integers. sha512crypt is even more worse, since it is doing both, lots of iterations and generating huge datablock that need transformation. i dont expect sha512crypt being much faster on gpu than on cpu. maybe x 2 or x 3 at best. i think we have to wait for better 64 bit int support on gpus. Ah, that's too bad. Thanks for the quick response. 
01-04-2012, 03:28 PM
some sha512(unix) benches here ? i have got a q6xx quadcore 4x4.3 ghz , while cracking with bf my result is 390 p/sec , maybee a gast i7 oc was able to handle 1000 p/sec ??? comone , dont talk about hard algo , let me see results...
01-12-2012, 07:08 AM
I get around 1.4k c/s on my Phenom II 965 Black not overclocked.
After thinking about it for a bit, even if it only 2-3 times faster on GPU, it would be nice to have it in the tool regardless. If it wouldn't be too much work to implement the algorithm for GPU cracking, I think it would be worth it. It would be interesting to see how much faster it actually is.
09-07-2012, 05:12 PM
|
|
« Next Oldest | Next Newest »
|