V47 bugs Nvidia - Printable Version +- hashcat Forum (https://hashcat.net/forum) +-- Forum: Developer (https://hashcat.net/forum/forum-39.html) +--- Forum: Beta Tester (https://hashcat.net/forum/forum-31.html) +--- Thread: V47 bugs Nvidia (/thread-1462.html) Pages:
1
2
|
V47 bugs Nvidia - mastercracker - 08-15-2012 System: Windows 7 64 bit, Driver 285.62 GTX470, oclhashcat-plus beta 47 General problem: - Something got wrong with -a 3. It does not work for all length on the following algos: 0, 10, 11, 20, 21, 30, 40, 100, 101, 110, 111, 112, 120, 121, 122, 130, 131, 132, 140, 141, 300, 900, 1000, 1100, 1400, 1410, 1420, 1500, 1700, 1710, 1720, 1722, 2400, 2600, 2611, 2711, 2811, 3000 and 3100. As you can see, heavy algo seems to be untouched by this problem. - The progress and reject counter are wrong, see attached picture. Specific problems -m 130 sha1(unicode($pass).$salt), -a 1 and -a 7 fails for all length. -m 131 MSSQL(2000), -a 1 and -a 7 fails for all length. -m 132 MSSQL(2005), -a 1 and -a 7 fails for all length. -m 1710 sha512($pass.$salt), -a 1, -a 6 and -a 7 fails for all length. -m 1800 sha512crypt, SHA512(Unix), length 15 in -a 1 and -a 6 fails. Length 4 and 8 in -a 7 fails. -m 2711 only dictionary (-a 0) attack works. -m 2811 only dictionary (-a 0) attack works. - I get the ERROR: cuStreamSynchronize() 999 on the following attack line for -m 1800 and -m 2100: cudahashcat-plus64.exe -a 3 -d 1 -m 1800 -o found.txt --bf-min 4 --bf-max 4 -1 ?s?d?u?l hash.txt ?u?d?s?d RE: V47 bugs Nvidia - atom - 08-17-2012 Thanks mastercracker! Quote:-m 2711 only dictionary (-a 0) attack works. This is interessting. The same happens on AMD, so maye there is a shared bug. RE: V47 bugs Nvidia - atom - 08-17-2012 v48 is up but this is only compiled with cuda 5.0. there should be no changes to the stated bugs. there is no need to test again before i fixed them. cuda 5.0 however produces massive performance increases, see here: https://twitter.com/hashcat/status/236218225082388480 RE: V47 bugs Nvidia - atom - 08-21-2012 fixed: Quote:-m 141 Nothing works New version is up, pls test RE: V47 bugs Nvidia - atom - 08-21-2012 how can i reproduce this? Quote:- The progress and reject counter are wrong, see attached picture. RE: V47 bugs Nvidia - mastercracker - 08-22-2012 For further Nvidia testing, it seems that I need to update the driver I have. I currently get the CuModuleLoad() 209 error. I will try updating it when I'll get back home. RE: V47 bugs Nvidia - atom - 08-23-2012 Yep, you need 304.x on Linux, not sure which version to use on windows. RE: V47 bugs Nvidia - mastercracker - 08-24-2012 (08-21-2012, 01:32 PM)atom Wrote: how can i reproduce this?It happens for example with this command line: cudahashcat-plus64.exe -a 1 -d 1 -m 3200 -o found.txt hash.txt dic1.txt dic1.txt So you can try combination attack on blowfish and see if you get the same thing. Otherwise I can provide you with the hash.txt and dic1.txt files. This is for NVidia with my GTX470 but the same happens with AMD HD5750. Also note that for the dictionary attack (-a 0), both counters work fine in both systems (AMD and NVIDIA). RE: V47 bugs Nvidia - Rolf - 08-25-2012 You need to download a 883MB package just to install the beta driver needed to run latest betas of cudacats. Crazy if you ask me. RE: V47 bugs Nvidia - M@LIK - 08-25-2012 @Rolf:: You sure? I grabbed the latest beta driver from here (304.79, Only ~200Mb): http://www.nvidia.com/Download/Find.aspx Forget to say, it feels "buggy" for me, even the WindowsExperinceIndex rating went down :( |