hashcat Forum
[ASK] About hashcat benchmark - Printable Version

+- hashcat Forum (https://hashcat.net/forum)
+-- Forum: Deprecated; Ancient Versions (https://hashcat.net/forum/forum-46.html)
+--- Forum: Very old oclHashcat-plus Support (https://hashcat.net/forum/forum-23.html)
+--- Thread: [ASK] About hashcat benchmark (/thread-2099.html)



[ASK] About hashcat benchmark - unname2012 - 02-25-2013

Hello all,i have a system with :
CPU : Intel(R) Core(TM) i7-3770 CPU @ 3.40GHz
GPU : HD7970 x 2
RAM : 8GB
OS : Ubuntu 12.10 (64)

i have benchmarked using oclHashcat64 and DES Algorithm ,then compare with (PC3: Ubuntu 12.04.1, 64 bitCatalyst 13.11x AMD hd6990stock core clockoclHashcat-plus v0.13 ) --> 78941k c/s
but i don't know why ? my SPEED GPU just 380k/s ? it's right or wrong ?
and this's a picture : http://postimage.org/image/q3a9sz9t7/
Thank


RE: [ASK] About hashcat benchmark - epixoip - 02-25-2013

first, you should be comparing your speed to "PC1", not "PC3."

second, you are benchmarking with 202 hashes, instead of one hash. this will slow down your speed by a factor of 202. 388k * 202 = 78376k.

also, please post your command line and text output next time instead of a screenshot.


RE: [ASK] About hashcat benchmark - unname2012 - 02-25-2013

ok,thank bro,i have a question that should i use max performance of GPU (such as --gpu-loops=1024,--gpu-accel=160) or default and if i use dictionary to crack instead of bruteforce , is it faster ?

Thank!


RE: [ASK] About hashcat benchmark - epixoip - 02-25-2013

using -n 160 -u 1024 will yield better performance than the defaults, yes.

if the password is in the dictionary you're using, then of course dictionary would be faster than brute force. checking more plains per second doesn't necessarily translate into "faster."

but as far as acceleration goes, hashcat uses an on-gpu plaintext candidate generator, so brute force can be accelerated more than dictionary attacks can. you will have to use rules to get full performance with dictionary attacks.


RE: [ASK] About hashcat benchmark - unname2012 - 02-25-2013

1/
follow you said like above
using -n 160 -u 1024 will yield better performance than the defaults, yes.
if i use -n 160 -u 1024 , it really damage for hardware ?

2/
The operator system affect to GPU speed ? i mean i,with my current hardware system , if i using linux (ubuntu) faster than window ?

Thank


RE: [ASK] About hashcat benchmark - epixoip - 02-25-2013

no, it will not damage your hardware.

no, the operating system does not affect speed.


RE: [ASK] About hashcat benchmark - unname2012 - 02-25-2013

(02-25-2013, 02:54 PM)epixoip Wrote: no, the operating system does not affect speed.

i have a question
like my system above,i crack NTLM ,GPU Speed is 8750 M/s (default,not max perfomance) , looking for PC 1 again, GPU Speed is 7500 M/s , what's different ?

Thank


RE: [ASK] About hashcat benchmark - epixoip - 02-25-2013

ghz edition? overlclocking?


RE: [ASK] About hashcat benchmark - unname2012 - 02-25-2013

(02-25-2013, 03:25 PM)epixoip Wrote: ghz edition? overlclocking?
what are you want to say about ?


RE: [ASK] About hashcat benchmark - epixoip - 02-26-2013

your speed is 8750 M/s, the reference benchmark is 7500 M/s.. that 7500 M/s is using the stock clock of 925 Mhz. so if you have a ghz edition card, or are overclocking, then that explains why your card is faster. make sense?