Page 2 of 2 FirstFirst 12
Results 11 to 20 of 20

Thread: Airolib-ng VS. Aircrack-ng

  1. #11
    Developer
    Join Date
    Mar 2007
    Posts
    6,126

    Default

    well this is surprising. test was done on a q6600 quad core with the 1,000,000 wordlist that comes with the 33 gig tables

    r00t@infected ~/Downloads $ time /usr/bin/aircrack-ng -w final-wordlist.txt -e buzza wpa.cap

    took
    real 4m27.130s
    user 17m6.288s
    sys 0m5.961s


    echo buzza | airolib-ng testdb -import essid -
    airolib-ng testdb --import passwd final-wordlist.txt
    airolib-ng testdb -batch
    /usr/bin/aircrack-ng -r testdb -e buzza wpa.cap

    I saved these lines in a script and that whole operation took.

    r00t@infected ~/Downloads $ time ./airo.sh

    real 57m36.802s
    user 57m27.454s
    sys 0m2.391s

  2. #12
    Good friend of the forums Eatme's Avatar
    Join Date
    Aug 2009
    Location
    Socks5
    Posts
    308

    Default

    hold on, did you test to see how long airolib-ng took to compute VS aircrack-ng running the whole list without a successful crack ?

    looks like you tested airolib-ng using aircrack-ng..idk maybe im reading it wrong. care to elaborate more ?
    Wiffy-Auto-Cracker - was the best thing that ever happen to me. :) Wo0oT :)
    AWUSO36H_500mW_5dBi Antenna

  3. #13
    Developer
    Join Date
    Mar 2007
    Posts
    6,126

    Default

    Quote Originally Posted by Eatme View Post
    hold on, did you test to see how long airolib-ng took to compute VS aircrack-ng running the whole list without a successful crack ?

    looks like you tested airolib-ng using aircrack-ng..idk maybe im reading it wrong. care to elaborate more ?
    I don't understand what you mean. Neither of the cracks were successful because the passphrase was not in the dictionary. I think you may be confused about how airolib works.

  4. #14
    Developer
    Join Date
    Mar 2007
    Posts
    6,126

    Default

    This is how long it takes to compute a database and the run it through aircrack using the 1 million password list.

    Code:
                                    Aircrack-ng 1.0
    
    
                       [00:00:09] 994836 keys tested (108705.66 k/s)
    
    
                           Current passphrase: zyromski                   
    
    
          Master Key     : 90 3F 44 7E AB CB 0B F0 59 58 6A 9D 23 30 9C 01 
                           7C AB 5F 3E FA 45 5B DB 4C EE 79 5C 66 B8 2F FD 
    
          Transient Key  : F4 F8 A6 19 89 2A 42 6E 5E 27 C9 A1 CB CD EE BD 
                           4C 9D 2A 8E 8C 96 E8 90 44 C9 D8 50 73 9C 55 A0 
                           F7 A1 68 C2 94 53 7C 07 4A 24 68 12 87 E8 F4 C2 
                           B4 7D A3 F6 1C 30 3A 43 25 9A 0C AF 82 3B 5D D9 
    
          EAPOL HMAC     : 32 95 BC 82 AC 7F 69 5C 86 F0 DC C6 B9 7F 14 45 
    
    
    Quitting aircrack-ng...
    
    real	57m36.802s
    user	57m27.454s
    sys	0m2.391s
    And this is how long it takes to run the same list just through aircrack

    Code:
                                     Aircrack-ng 1.0
    
    
                       [00:04:26] 996256 keys tested (3816.56 k/s)
    
    
                           Current passphrase: }terfinn                   
    
    
          Master Key     : 62 E9 4F 9B F2 28 9B 14 E3 3B CA 63 66 69 7A A0 
                           C8 F2 FC AF 58 BE B6 53 71 0F FD 5B BB 50 2E 03 
    
          Transient Key  : 98 E6 9F D9 1E E1 53 B7 30 4C 69 87 66 B5 D7 4B 
                           2F 9F 9C 61 7F 11 E3 A0 7B 7E 7F 92 FE C5 B7 E2 
                           1D 3E D8 53 D3 B3 0B 4F 3C DA 2C FA 56 55 EE 61 
                           02 2C 62 41 E3 B3 E8 E3 2B 80 71 78 F3 7F 02 9A 
    
          EAPOL HMAC     : 89 7E 5F F9 92 27 81 60 42 AE AD 02 63 FC A2 02 
    
    Passphrase not in dictionary 
    
    
    
    Quitting aircrack-ng...
    
    real	4m27.130s
    user	17m6.288s
    sys	0m5.961s
    This big difference is SSE2 which aircrack-ng code uses but airolib-ng's does not. So when I use airolib-ng I can only compute about 200-300 pmk/s however with aircrack-ng I can test 3500-4000 keys per second.

    Here is a link for sse2 if you have never heard of it. http://en.wikipedia.org/wiki/SSE2
    sse2 is what Edgan worked so hard to incorperate into Pyrit which gives us the CPU boost we now have.

  5. #15
    Good friend of the forums Eatme's Avatar
    Join Date
    Aug 2009
    Location
    Socks5
    Posts
    308

    Default

    lol.... no purehate

    i was saying that because that was what you were sappose to do as in not put the pw in the list, i was just making sure..lol

    Also, you weren;t suppose to test the computed list in aircrack (if you did it just for your sake then ok) you were only suppose to calculate the time it takes to "make the computed list" VS running a regualar/non computed list in aircrack...sorry if i wasn't clear enough on the first page.

    so my next question is, how long did it take to compute the whole list...?

    compare this time VS. "how long it takes to run the same list just through aircrack"

    im guessing running the non computed list through aircrack will finish before airolib computes it..this was my whole theory.

    EDIT:
    If Aircrack-ng can run the whole list (without cracking a pw, even tho this will take even longer without crackin it) then this will make it clear to the users that doesn't need to save hash/pre-computed files...in case they just wanna run a tested essid with a large pwl, they can just run it through Aircrack-ng instead of computing it since they wont need the table later on..other wise its better to do it vise versa.

    Edit:
    nvm, i kept reading the times wrong.. correct if im right..

    1.It took almost an 1hr (57min) to compute the list?
    2.It took 4min to run the non computed list ?

    If yes, then my theory was right all along..thanks for testing this, ill give it a try and post my results with my hardware. This could get interesting on different hardware setups IMO.
    Wiffy-Auto-Cracker - was the best thing that ever happen to me. :) Wo0oT :)
    AWUSO36H_500mW_5dBi Antenna

  6. #16
    Developer
    Join Date
    Mar 2007
    Posts
    6,126

    Default

    Also, you weren;t suppose to test the computed list in aircrack (if you did it just for your sake then ok) you were only suppose to calculate the time it takes to "make the computed list" VS running a regualar/non computed list in aircrack...sorry if i wasn't clear enough on the first page.
    Once again you are confused I think. What would be the point in only testing the time it took to compute the list?

    Well I tried to help you understand and post the results but now you have thoroughly confused me with what you are talking about. Plus your spelling and typing sucks so its hard to understand.

    It would be easy to subtract aircracks time from the airolib figures

    108705.66 / 60 and then subtract that from 57m36s although that is pointless.

    So the real point is that it takes 20 times as long to compute the database and run it through aircrack as it does to just use aircrack by it self. The only upside to using airolib is that you have a table left over.

  7. #17
    Good friend of the forums Eatme's Avatar
    Join Date
    Aug 2009
    Location
    Socks5
    Posts
    308

    Default

    Quote Originally Posted by pureh@te View Post
    So the real point is that it takes 20 times as long to compute the database and run it through aircrack as it does to just use aircrack by it self. The only upside to using airolib is that you have a table left over.
    no im not. Im trying to prove that by using Aircrack-ng alone is better then computing a pwl using airolib-ng unless you want a saved table for later use.

    Quote Originally Posted by pureh@te View Post
    What would be the point in only testing the time it took to compute the list?
    Because if you dont want to save a table for later use, which i explain in the above post also, its better to just run Aircrack-ng alone.

    I know how Airolib-ng works, i just look at it as, its only scanning a pwl against a essid(s) without verifying the correct pw(s).

    So if you use Airolib+Aircrack, you're basically running the list twice, but the good thing is that on the 2nd time it would run much faster and you get to save the table.

    I was just trying to point something out to the people that will or have questions about which one to use..and why. I know its some people out there that don't need or want tables for later use, so why go through the list twice.

    My opinion is, shit you get the best of both worlds by saving tables..you just gotta have patience.
    Wiffy-Auto-Cracker - was the best thing that ever happen to me. :) Wo0oT :)
    AWUSO36H_500mW_5dBi Antenna

  8. #18
    My life is this forum Snayler's Avatar
    Join Date
    Jan 2010
    Posts
    1,418

    Default

    Quote Originally Posted by Eatme View Post
    I know how Airolib-ng works, i just look at it as, its only scanning a pwl against a essid(s) without verifying the correct pw(s).
    So if you use Airolib+Aircrack, you're basically running the list twice, but the good thing is that on the 2nd time it would run much faster and you get to save the table.
    I was just trying to point something out to the people that will or have questions about which one to use..and why. I know its some people out there that don't need or want tables for later use, so why go through the list twice.
    I give up trying to understand the point of all this... You're not making any sense...

  9. #19
    Good friend of the forums Eatme's Avatar
    Join Date
    Aug 2009
    Location
    Socks5
    Posts
    308

    Default

    sure i am. All im saying is, you will save time in cracking a pw using aircrack-ng alone.

    Ex:
    If both of use wanted to crack a pw at the same time.

    Im using: Aircrack-ng
    &
    You using: Airolib-ng+Aircrack-ng

    The size of pwl doesn't matter at this time. But lets say the correct pw is "construction"

    while ur computing the pwl, my aircrack-ng will be running...everyone knows i will crack construction way before you even started cracking..lol

    The point is the time. Thats all im saying

    Another thing i wanted to point out is that, since the pw is "construction" it will be at the top of the list since it starts with the letter "c". If the pwl wasn't sorted in alphabetically it will take longer.

    BUT, since this is for testing lets pretend we didnt know what the pw was, and the list wasn't sorted out alphabetically and the correct pw happened to be on the 5th line of 1 million lines. that would crack even faster, probably 1 sec or less.

    So for this theroy, a pwl thats not sorted out alphabetically, wouldn't we have a better/faster chance of crackin a pw VS. someone using a sorted out pwl ???

    Sorry for all the BS theory's, but im the type that thinks outside the box. Bare with me
    Wiffy-Auto-Cracker - was the best thing that ever happen to me. :) Wo0oT :)
    AWUSO36H_500mW_5dBi Antenna

  10. #20
    Very good friend of the forum TAPE's Avatar
    Join Date
    Jan 2010
    Location
    Europe
    Posts
    599

    Default

    wow..


    Although in fact it wouldn't be an uninteresting test, of course it makes no sense to have an unstructured approach.

    Bit like going fishing.
    Me using hook and a line,
    You throwing a rock in the water and hoping the scared fish will jump in your basket.

Page 2 of 2 FirstFirst 12

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •