hashcat Forum › Deprecated; Ancient Versions › Very old oclHashcat-lite Support v
0.10 a litle bit slower

Threaded Mode
0.10 a litle bit slower
adolfotregosa Offline
Junior Member
**
Posts: 6
Threads: 1
Joined: Dec 2011
#1
07-02-2012, 10:57 PM (This post was last modified: 07-02-2012, 10:59 PM by adolfotregosa.)
I am NOT complaining, ok ??

Just to ask if it is supposed because of new code implementation or I'm I doing something wrong ?

thank you

print from 0.9 vs 0.10


Attached Files Thumbnail(s)
       
Find
KT819GM Offline
Senior Member
****
Posts: 344
Threads: 2
Joined: Aug 2011
#2
07-03-2012, 06:09 AM
You need to turn off markov, because stats file which *lite now uses have nothing to do with SL3.
Find
adolfotregosa Offline
Junior Member
**
Posts: 6
Threads: 1
Joined: Dec 2011
#3
07-03-2012, 11:09 AM
(07-03-2012, 06:09 AM)KT819GM Wrote: You need to turn off markov, because stats file which *lite now uses have nothing to do with SL3.

ok, but If i understood correctly, you don't have that option.

No biggie! I can still use 0.9. It has been not less that perfect all this time ! Never has failed on me.

thank you
Find
KT819GM Offline
Senior Member
****
Posts: 344
Threads: 2
Joined: Aug 2011
#4
07-03-2012, 11:30 AM
Code:
* Markov:

       --markov-hcstat               Specify hcstat file to use, default is hashcat.hcstat
       --markov-disable              Disables markov-chains, emulates classic brute-force
       --markov-classic              Enables classic markov-chains, no per-position enhancement
  -t,  --markov-threshold=NUM        Threshold when to stop accepting new markov-chains

So try --markov-disable on your command line.
Find
adolfotregosa Offline
Junior Member
**
Posts: 6
Threads: 1
Joined: Dec 2011
#5
07-03-2012, 11:41 AM (This post was last modified: 07-03-2012, 11:41 AM by adolfotregosa.)
(07-03-2012, 11:30 AM)KT819GM Wrote:
Code:
* Markov:

       --markov-hcstat               Specify hcstat file to use, default is hashcat.hcstat
       --markov-disable              Disables markov-chains, emulates classic brute-force
       --markov-classic              Enables classic markov-chains, no per-position enhancement
  -t,  --markov-threshold=NUM        Threshold when to stop accepting new markov-chains

So try --markov-disable on your command line.

I did look for it but I've missed that one, thank you !
But still slower. It is no problem, 0.09 works fine for sl3.

I understand that because of new code this can happen and real life usage it would only take some extra minutes. It was just for "benchmark".
Find
forumhero Offline
Senior Member
****
Posts: 313
Threads: 44
Joined: Aug 2011
#6
07-03-2012, 05:30 PM
have you tried updating drivers or compare by using different OS like linux?
Find
adolfotregosa Offline
Junior Member
**
Posts: 6
Threads: 1
Joined: Dec 2011
#7
07-03-2012, 05:33 PM
(07-03-2012, 05:30 PM)forumhero Wrote: have you tried updating drivers or compare by using different OS like linux?

has a matter of fact when i saw less speed I updated from 12.4 to 12.7 beta just to be sure it wasn't drivers. Haven't tried linux because I don't use it.
Find
atom Offline
Administrator
*******
Posts: 5,185
Threads: 230
Joined: Apr 2010
#8
07-03-2012, 05:37 PM
(07-03-2012, 11:30 AM)KT819GM Wrote: So try --markov-disable on your command line.

This wont help. The speed remain. Speed drop from 0.09 to 0.10 is not caused by word-generator. Its because AMD did some changes to the JIT compiler.
Website Find
yayo050909 Offline
Junior Member
**
Posts: 4
Threads: 3
Joined: Sep 2012
#9
09-19-2012, 03:22 PM
(07-03-2012, 05:37 PM)atom Wrote:
(07-03-2012, 11:30 AM)KT819GM Wrote: So try --markov-disable on your command line.

This wont help. The speed remain. Speed drop from 0.09 to 0.10 is not caused by word-generator. Its because AMD did some changes to the JIT compiler.

I don`t understand, what is best for sl3 markov-disable or markov-hcstat.
I am downloading the file sl3.hcstat.
Find