Page 3 of 16 FirstFirst 1234513 ... LastLast
Results 21 to 30 of 156

Thread: pyrit CUDA nvidia Tutorial + Nvidia overclock instructions

  1. #21
    Member
    Join Date
    Aug 2007
    Posts
    468

    Default

    Great work pureh@te,

    Regretting I sent for an ATI 4870 card now my new system.

  2. #22
    Junior Member SWFu64's Avatar
    Join Date
    Jan 2010
    Posts
    97

    Default

    Quote Originally Posted by BOFH139 View Post
    Great work pureh@te,

    Regretting I sent for an ATI 4870 card now my new system.
    You might want to check this out:

    http://znuh.blogspot.com/2009/01/wpa...nd-radeon.html

    Seems as though the GTX 295 hits 8628.76 PMKs/s and the HD 4870 9650 PMKs/s (10000 OC'ed!) only going by other peoples results from the above URL's.

    GPUs have always interested me in this field and I'm about to buy a quad core and high end GPU system, maybe we could get together a thread with speeds we are hitting with our GPUs?
    "I do not know with what weapons World War III will be fought, but World War IV will be fought with sticks and stones."

    Albert Einstein

  3. #23
    Just burned his ISO
    Join Date
    Jan 2008
    Posts
    3

    Default

    Pyrit on Ati works in BT3. Install Ati driver: hxxp://ati.amd.com/support/drivers/linux/previous/linux-rf-cat812.html. Download it: hxxp://uloz.to/1270984/amd_brook.rar. Unrar and install two files with installpkg. Open /etc/ld.so.conf and add line "/usr/local/amdbrook/sdk/lib". Compile Pyrit with command: ./setup.py HAVE_STREAM build and ./setup.py install. Sorry my very bad english.

  4. #24
    Just burned his ISO
    Join Date
    Feb 2009
    Posts
    7

    Default

    Great tutorial.It works better than expected.Thank you(both)

  5. #25
    Just burned his ISO
    Join Date
    Dec 2007
    Posts
    16

    Default crappy or bad install

    Hey dont know if this is an issue, only thing throwing me off is the HUGE diference between 8800 gt score and mine 8600M gs.
    Code:
    root@bt:/pentest/password/pyrit# python pyrit.py benchmark
    The Pyrit commandline-client (C) 2008 Lukas Lueg 
    This code is distributed under the GNU General Public License v3
    
    The ESSID-blobspace seems to be empty; you should create an ESSID...
    Available cores: 'Standard CPU', 'Nvidia CUDA'
    Testing CPU-only core 'Standard CPU' (2 CPUs)...  358.41 PMKs/s
    
    Testing GPU core 'Nvidia CUDA' (Device 'GeForce 8600M GS')...  654.62 PMKs/s
    your thoughts. Bad installation or got scammed on the video card

  6. #26
    Developer
    Join Date
    Mar 2007
    Posts
    6,126

    Default

    Quote Originally Posted by Araquera View Post
    Hey dont know if this is an issue, only thing throwing me off is the HUGE diference between 8800 gt score and mine 8600M gs.
    Code:
    root@bt:/pentest/password/pyrit# python pyrit.py benchmark
    The Pyrit commandline-client (C) 2008 Lukas Lueg 
    This code is distributed under the GNU General Public License v3
    
    The ESSID-blobspace seems to be empty; you should create an ESSID...
    Available cores: 'Standard CPU', 'Nvidia CUDA'
    Testing CPU-only core 'Standard CPU' (2 CPUs)...  358.41 PMKs/s
    
    Testing GPU core 'Nvidia CUDA' (Device 'GeForce 8600M GS')...  654.62 PMKs/s
    your thoughts. Bad installation or got scammed on the video card
    Well thats for sure a bad number. It is a mobile card though so it may not preform that well. I would guess some of your cores in your card may be bad or some thing along those lnes.

  7. #27
    Just burned his ISO
    Join Date
    Dec 2007
    Posts
    16

    Default

    Well, this card is actually under-clocked, under windows I have it OCed to a stable point. Nvidia website indicates the clock to be at 600/1200/480, and mine is at 500/1000/400. been searching a little bit on how to OC under linux haven't found something concrete, yet, I haven't obsessed over it yet but I guess ill start

  8. #28
    Developer
    Join Date
    Mar 2007
    Posts
    6,126

    Default

    That reminds me....

    How to overclock your nvidia card in Linux......

    run this commnad: apt-get install nvidia-settings

    That will install the nvidia-settings interface. Now in order to overclock you have to enable it like this:
    (make sure to do it for each card if you have more than 1)

    nano /etc/X11/xorg.conf

    and find the section that looks like this:

    Section "Device"
    Identifier "Videocard1"
    Driver "nvidia"
    VendorName "NVIDIA Corporation"
    BoardName "GeForce 8800 GT"
    BusID "PCI:3:0:0"
    Screen 1
    Option "AddARGBGLXVisuals" "true"
    Option "Coolbits" "1"
    Option "RenderAccel" "true"
    EndSection

    Add the coolbits option and then restart X and open nvidia-settings and you should have a overclock option like this:


    Thats it, Switch it around however you want.

    Alternatively you can also use the command line tool nvclock: apt-get install nvclock
    Code:
    r00t@pwnsauce ~ $ nvclock
    NVClock v0.7
    
    Using NVClock you can overclock your Nvidia videocard under Linux and FreeBSD.
    Use this program at your own risk, because it can damage your system!
    
    Usage: ./NVClock [options]
    
    Overclock options:
       -c  --card number            Number of the card to overclock
       -m  --memclk speed           Memory speed in MHz
       -n  --nvclk speed            Core speed in MHz
    
       -r  --reset          Restore the original speeds
    
    Other options:
       -d  --debug                  Enable/Disable debug info
       -f  --force                  Force a speed, NVClock won't check min/max speeds
       -h  --help                   Show this help info
       -i  --info                   Print detailed card info
       -s  --speeds                 Print current speeds in MHz

  9. #29
    Just burned his ISO
    Join Date
    Dec 2007
    Posts
    16

    Default

    Thx Purehate, found nvclock, but having a little bit of issues installing the GTK version, gtk+-2.0 >= 2.4.0... checking for x11... configure: error: "X11 required for nvcontrol support" this shows up after ./configure.

    but Im gonna stick with your first suggestion for now

  10. #30
    Just burned his ISO
    Join Date
    Jan 2010
    Posts
    2

    Default

    Quote Originally Posted by devto View Post
    Same errore, same card.

    Its the CUDA_INSTALL.. It should be --force... instead of just one dash. This fixed it for me.

Page 3 of 16 FirstFirst 1234513 ... LastLast

Posting Permissions

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