oclHashcat-plus v0.09 RC1
#1
Hey Beta-testers,

after all the bugfixing its time for the 1st (and hopefully the only one) release candidate for oclHashcat-plus v0.09.

The last beta version before RC1 was b75.

There will be no more new features - only bug fixes if there are any. If there are no major bugs found I plan to release on Friday.

All other beta related threads has been closed. Please post everything in this thread. You might want to check all the other files that come with the archive, too, like docs/ etc. Maybe there are typos etc.

I have a good feeling about this v0.09 but you are free to destroy it with bug-reports Smile

PS linux users: This version has been compiled on old GLIBC Smile

Many thanks to you guys for invest of time in testing everything!

--
atom
#2
Cool, many thanks atom. We all own u big time for the amount of time and effort u put into this. My hat goes off 2 u sir Smile

I'll do some testing tnite hopefully.
Owner of: www.md5decrypter.co.uk
#3
Updated graph attached with v0.09rc1


Attached Files
.xlsx   oclhc-plus_graph.xlsx (Size: 13.77 KB / Downloads: 3)
Owner of: www.md5decrypter.co.uk
#4
I found some typos in the help. Well, you might argue if these are typos or not but i think they all should have their first char be uppercase to be conform with all other lines.

--disable-potfile do not write potfile
--cpu-affinity=STR locks to CPU devices, seperate with comma
-j, --rule-left=RULE single rule applied to each word from left dict
-k, --rule-right=RULE single rule applied to each word from right dict
sch0.org
#5
(09-04-2012, 09:03 PM)blandyuk Wrote: Updated graph attached with v0.09rc1

Thanks! Can I add them to release notes?

https://hashcat.net/misc/plus89_mh18.png
https://hashcat.net/misc/plus89_mh915.png
#6
regarding the chart, how about a bar graph? i think it better emphasize an increase in performance, rather than a line graph that's sloping downwards...
#7
Go ahead, although I now use 1 5870 and 1 5850 now Smile will have 3 x 5870s soon and will look at upgrading more after that.
Owner of: www.md5decrypter.co.uk
#8
Just a few notes before we dive into the hardcore stuff:
  • docs\user_manuals.txt(line_3) is old enough to be removed, or just tag it with "Outdated".
  • --help(line_35), you can add " --outfile-format=NUM Define outfile-format for recovered hash, see reference below".
  • I'm not sure about this one but you need to check --outfile-format=4, its outputs is always padded with zeros. Try:
    Code:
    echo -n 00 | hc64p --outfile-format=4 b4b147bc522828731f1a016bfa72c073
#9
Full testing on Nvidia with my GTX460 was flawless. I just realized the inclusion of -j and -k. This is good news. I will have to think on how to exploit it properly in my attack scheme. I might be wrong but I also remember discussion about being able to crack empty passwords. Is that implemented yet or in a future release? Finally, the new distributed feature should be explained somewhere in the docs and/or be ready and published on the wiki at the release date.

Beside that, great job. For me it's good to go.
#10
(09-04-2012, 09:16 PM)kartan Wrote: I found some typos in the help. Well, you might argue if these are typos or not but i think they all should have their first char be uppercase to be conform with all other lines.

--disable-potfile do not write potfile
--cpu-affinity=STR locks to CPU devices, seperate with comma
-j, --rule-left=RULE single rule applied to each word from left dict
-k, --rule-right=RULE single rule applied to each word from right dict

fixed, thanks!