[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 25 Jul 2012 18:12:00 +0800
From: myrice <qqlddg@...il.com>
To: john-dev@...ts.openwall.com
Subject: Re: Result of hard core password generation on 7970
Solar
On Tue, Jul 24, 2012 at 8:52 PM, myrice <qqlddg@...il.com> wrote:
> Hi, Solar,
>
> Here is the result, with password generation on [a-zA-Z][a-zA-Z]
> The code is here on PG-test branch:
> https://github.com/qqldd/myrice-JtR/tree/PG-test
>
> raw-md5-opencl with PG
> 1
> guesses: 0 time: 0:00:04:52 0.00% c/s: 196758K
>
> 1000
> guesses: 2 time: 0:00:01:44 0.00% c/s: 30660M
>
> 1000000
> Unable to get, will reduce the max_keys_per_crypt number or divided it
> to 1000s kernel's run(every run with fixed 1000 loaded hashes), so
> loaded hashes could take advantages of constant memory.
Here is the new rough result
1: with 2048*8, ~900M c/s, but with 2048*16, it is ~500M c/s
1000: with 2048*8, ~45G = ~45M, with 2048*16, ~90G = ~ 45M
1M: still cannot get, I reduce the global work size to 128 and only
append[a-e][a-e], it is very slow. And the kernel cannot finished with
[a-z][a-z].
I am think about compare, with 1M hashes, the loop inside one thread
increased to 26*26*1M = 676M, it is very large for a thread.
Thanks
myrice
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
