Comparison with an older version!

[wagh]

I have pdfcrack v.011 installed (from SF) for several years.
Recently, I installed the -mp version and compared it with 0.11 on some PDF files. Here are some observations / comments:

1. Version 0.11 supported multi-threaded (whereas this feature was missing from 0.15 & 0.16 -- Have not tried intermediate versions)
2. Ver. 0.11 is almost 3 times faster than -mp version on almost all the files I tried.
3. Ver 0.11 supports cracking owner password, if you provide user pass with -p switch; whereas -mp version gives the following:
   Multithread operation only supports user-password search.
4. -mp version supports pattern, which is not possible in 0.11.

Am I missing something or giving some wrong parameters?

Disclaimer:
I am writing this just as a comment and do not intend to raise this as an issue.

[shreepads]

Please see #24

…

On Wed 1 Aug, 2018, 3:17 PM VInay Wagh, ***@***.***> wrote: I have pdfcrack *v.011* installed (from SF <https://sourceforge.net/projects/pdfcrack/>) for several years. Recently, I installed the *-mp* version and compared it with 0.11 on some PDF files. Here are some observations / comments: 1. Version 0.11 supported multi-threaded (whereas this feature was missing from 0.15 & 0.16 -- Have not tried intermediate versions) 2. Ver. 0.11 is almost 3 times faster than *-mp* version on almost all the files I tried. 3. Ver 0.11 supports cracking owner password, if you provide user pass with *-p* switch; whereas *-mp* version gives the following: *Multithread operation only supports user-password search.* 4. *-mp* version supports pattern, which is not possible in 0.11. Am I missing something or giving some wrong parameters? Disclaimer: I am writing this just as a *comment* and do not intend to raise this as an *issue*. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#27>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AMGDunqAx8Aiq5gVBUu76kGm70jka2Noks5uMXkdgaJpZM4VqLy4> .

[wagh]

I dont see the relation with #24! Could you please explain?

• I am already using the branch baseopenmp.
• The parameters were exactly same for both the versions: v.0.11 and -mp.
• The performance comparison is (comment 2 in original post) when there is no pattern known.
• Most of the sample passwords were numeric (with -c0987654321). -- Interestingly though, the order in which this set is mentioned matters a lot!!!

[shreepads]

" Viktor I've only implemented multi threading for the new pattern option while leaving the existing methods untouched. The pattern can cover anything that could be done with the existing options including variable length and char types. "

…

On Wed 1 Aug, 2018, 4:23 PM VInay Wagh, ***@***.***> wrote: I dont see the relation with #24 <#24>! Could you please explain? - I am already using the branch baseopenmp. - The parameters were exactly same for both the versions: v.0.11 and -mp. - The performance comparison is (comment 2 in original post) when there is no pattern known. - Most of the sample passwords were numeric (with *-c0987654321*). -- Interestingly though, the order in which this set is mentioned matters a lot!!! — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#27 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AMGDuvA6GJANO-x2x1YyuJZCEVebuVs_ks5uMYiNgaJpZM4VqLy4> .