Strange Problem With oclHashcat-plus

I have been using hashcat for long time but now i face a problem which i cant explain.

I have some 70.000 IPB hashes in file ipb.txt . All are valid.

I'am using that command[oclhashcat 0.8]

oclHashcat-plus64.exe --hash-type 2811 --remove --outfile C:/ipb.txt.out --outfile-format 2 C:/ipb.txt E:\WordLists\A_passfound_rules.dic

The attack starts with success but it cant find any password although that i have some passwords that fits the hashes. If i try the CPU version it succefully cracks some hashes.

Also, another strange thing is that , if i use --hash-type 2611 (in the same IPB LIST) the attacks start again. I chose the wrong algorithm but somehow it validate the hashes.

What i am doing wrong??


even if i try this command:

oclHashcat-plus64.exe --hash-type 2611 --remove --outfile C:/ipb.txt.out --outfile-format 2 a810600eebe27c334589c7e0c3795effAngryuC3t E:\WordLists\A_passfound_rules.dic

I enter an IPB hash, I selected vbulletin < 3.8.6 but the attack it starts again with wrong algorithm.

Status.......: Running
Input.Mode...: File (E:\WordLists\A_passfound_rules.dic)
Hash.Target..: a810600eebe27c334589c7e0c3795effAngryuC3t
Hash.Type....: vBulletin < v3.8.5
Time.Running.: 0 secs
Time.Left....: 0 secs
Time.Util....: 554.2ms/484.0ms Real/CPU, 689.6% idle
Speed........:  4434.5k c/s Real, 39811.5k c/s GPU
Recovered....: 0/1 Digests, 0/1 Salts
Progress.....: 2533907/4868474 (52.05%)
Rejected.....: 76307/2533907 (3.01%)
HW.Monitor.#1:  0% GPU, 59c Temp


Windows 7 Home Premium 64 bit
Latest Catalyst Version
This is a known issue with ATI Catalyst version 12.4 and ATI Radeon HD 5xxx and 6xxx series.

In the case of IPB, you cannot crack the multiple IPB hashes in the list.

I don't see a reason why you would choose a wrong algorithm when you know the hash type already. Both the hashes are formatted in a similar manner and the length of the salts differ.

let me find the link for you regarding the discussion on issue with cracking multiple IPB hashes. I have posted several combinations of attacks there along with the results.

Ok, so here is the link to the post where I have posted results of several combinations of attacks:
I am selecting wrong algorithm just to show you that hash validation isnt working. (It doesnt say "length exception ... etc etc).

vbulletin salt length (hash type 2611) is 3 length and ipb is 5 length. It should say that it isnt valid.

Ok so AMD has the bug for this.

Thanks for the answer.
its exactly as NeonFlash said
Guys i have to add something.

I have the latest catalyst 12.4 and i tried with oclHashcat 0.7 and it is working! So its not a driver issue. Right?
It is a driver issue. It is not important which version you use, it is the version I use when I compile the kernel.