Forum Ubuntu-it

Codice: Seleziona tutto

sudo apt-get install nvidia-glx-190

Codice: Seleziona tutto

sudo /etc/init.d/gdm stop

Codice: Seleziona tutto

sudo sh nvidia.run

Codice: Seleziona tutto

sudo reboot

Codice: Seleziona tutto

sudo chmod u+x nomefile.run

Codice: Seleziona tutto

sudo apt-get install binutils build-essential linux-headers-$(uname -r) pkg-config xserver-xorg-dev

Codice: Seleziona tutto

wget http://it.download.nvidia.com/XFree86/Linux-x86/195.36.08/NVIDIA-Linux-x86-195.36.08-pkg1.run

Codice: Seleziona tutto

wget http://it.download.nvidia.com/XFree86/Linux-x86_64/195.36.08/NVIDIA-Linux-x86_64-195.36.08-pkg2.run

Codice: Seleziona tutto

sudo apt-get --purge autoremove

Codice: Seleziona tutto

sudo stop gdm

Codice: Seleziona tutto

sudo chmod u+x NVIDIA-Linux-*

Codice: Seleziona tutto

sudo sh NVIDIA-Linux-* -a -n -N

Codice: Seleziona tutto

sudo start gdm

Codice: Seleziona tutto

sh NVIDIA-Linux-* --advanced-options

NVIDIA-Linux-x86-190.53-pkg1.run [options]

This program will install the NVIDIA Accelerated Graphics Driver
for Linux-x86 190.53 by unpacking the embedded tarball and executing
the ./nvidia-installer  installation utility.

--info
  Print embedded info (title, default target directory) and exit.

--lsm
  Print embedded lsm entry (or no LSM) and exit.

--pkg-history
  Print the package history of this file and exit.

--list
  Print the list of files in the archive and exit.

--check
  Check integrity of the archive and exit.

-x, --extract-only
  Extract the contents of /media/Dati/Linux/Appunti/Nvidia/NVIDIA-Linux-x86-190.53-pkg1.run, but do not
  run 'nvidia-installer'.

--add-this-kernel
  Build a precompiled kernel interface for the currently running
  kernel and repackage the .run file to include this newly built
  precompiled kernel interface.  The new .run file will be placed
  in the current directory and the string "-custom" appended
  to its name, unless already present, to distinguish it from the
  original .run file.

--apply-patch [Patch]
  Apply the patch 'Patch' to the kernel interface files included
  in the .run file, remove any precompiled kernel interfaces
  and then repackage the .run file.  The new .run file will be
  placed in the current directory and the string "-custom"
  appended to its name, unless already present, to distinguish it
  from the original .run file.

--keep
  Do not delete target directory when done.

--target [NewDirectory]
  Extract contents in 'NewDirectory'


The following arguments will be passed on to the ./nvidia-installer
utility:

COMMON OPTIONS:

  -a, --accept-license
      Bypass the display and prompting for acceptance of the
      NVIDIA Software License Agreement.  By passing this option
      to nvidia-installer, you indicate that you have read and
      accept the License Agreement contained in the file
      'LICENSE' (in the top level directory of the driver
      package).

  --update
      Connect to the NVIDIA FTP server '
      ftp://download.nvidia.com ' and determine the latest
      available driver version.  If there is a more recent driver
      available, automatically download and install it.  Any
      other options given on the commandline will be passed on to
      the downloaded driver package when installing it.

  -v, --version
      Print the nvidia-installer version and exit.

  -h, --help
      Print usage information for the common commandline options
      and exit.

  -A, --advanced-options
      Print usage information for the common commandline options
      as well as the advanced options, and then exit.


ADVANCED OPTIONS:

  -a, --accept-license
      Bypass the display and prompting for acceptance of the
      NVIDIA Software License Agreement.  By passing this option
      to nvidia-installer, you indicate that you have read and
      accept the License Agreement contained in the file
      'LICENSE' (in the top level directory of the driver
      package).

  --update
      Connect to the NVIDIA FTP server '
      ftp://download.nvidia.com ' and determine the latest
      available driver version.  If there is a more recent driver
      available, automatically download and install it.  Any
      other options given on the commandline will be passed on to
      the downloaded driver package when installing it.

  -v, --version
      Print the nvidia-installer version and exit.

  -h, --help
      Print usage information for the common commandline options
      and exit.

  -A, --advanced-options
      Print usage information for the common commandline options
      as well as the advanced options, and then exit.

  -i, --driver-info
      Print information about the currently installed NVIDIA
      driver version.

  --uninstall
      Uninstall the currently installed NVIDIA driver.

  --sanity
      Perform basic sanity tests on an existing NVIDIA driver
      installation.

  -e, --expert
      Enable 'expert' installation mode; more detailed questions
      will be asked, and more verbose output will be printed;
      intended for expert users.  The questions may be suppressed
      with the '--no-questions' commandline option.

  -q, --no-questions
      Do not ask any questions; the default (normally 'yes') is
      assumed for all yes/no questions, and the default string is
      assumed in any situation where the user is prompted for
      string input.  The one question that is not bypassed by
      this option is license acceptance; the license may be
      accepted with the commandline option '--accept-license'.

  -s, --silent
      Run silently; no questions are asked and no output is
      printed, except for error messages to stderr.  This option
      implies '--ui=none --no-questions --accept-license'.

  --x-prefix=X-PREFIX
      The prefix under which the X components of the NVIDIA
      driver will be installed; the default is '/usr/X11R6'
      unless nvidia-installer detects that X.Org >= 7.0 is
      installed, in which case the default is '/usr'.  Only under
      rare circumstances should this option be used.

  --xfree86-prefix=XFREE86-PREFIX
      This is a deprecated synonym for --x-prefix.

  --x-module-path=X-MODULE-PATH
      The path under which the NVIDIA X server modules will be
      installed.  If this option is not specified,
      nvidia-installer uses the following search order and
      selects the first valid directory it finds: 1) `X
      -showDefaultModulePath`, 2) `pkg-config
      --variable=moduledir xorg-server`, or 3) the X library path
      (see the '--x-library-path' option) plus either 'modules'
      (for X servers older than X.Org 7.0) or 'xorg/modules' (for
      X.Org 7.0 or later).

  --x-library-path=X-LIBRARY-PATH
      The path under which the NVIDIA X libraries will be
      installed.  If this option is not specified,
      nvidia-installer uses the following search order and
      selects the first valid directory it finds: 1) `X
      -showDefaultLibPath`, 2) `pkg-config --variable=libdir
      xorg-server`, or 3) the X prefix (see the '--x-prefix'
      option) plus 'lib' on 32bit systems, and either 'lib64' or
      'lib' on 64bit systems, depending on the installed Linux
      distribution.

  --opengl-prefix=OPENGL-PREFIX
      The prefix under which the OpenGL components of the NVIDIA
      driver will be installed; the default is: '/usr'.  Only
      under rare circumstances should this option be used.  The
      Linux OpenGL ABI
      (http://oss.sgi.com/projects/ogl-sample/ABI/) mandates this
      default value.

  --opengl-libdir=OPENGL-LIBDIR
      The path relative to the OpenGL library installation prefix
      under which the NVIDIA OpenGL components will be installed.
      The default is 'lib' on 32bit systems, and 'lib64' or 'lib'
      on 64bit systems, depending on the installed Linux
      distribution.  Only under very rare circumstances should
      this option be used.

  --installer-prefix=INSTALLER-PREFIX
      The prefix under which the installer binary will be
      installed; the default is: '/usr'.  Note: please use the
      '--utility-prefix' option instead.

  --utility-prefix=UTILITY-PREFIX
      The prefix under which the NVIDIA utilities
      (nvidia-installer, nvidia-settings, nvidia-xconfig,
      nvidia-bug-report.sh) and the NVIDIA utility libraries will
      be installed; the default is: '/usr'.

  --utility-libdir=UTILITY-LIBDIR
      The path relative to the utility installation prefix under
      which the NVIDIA utility libraries will be installed.  The
      default is 'lib' on 32bit systems, and 'lib64' or 'lib' on
      64bit systems, depending on the installed Linux
      distribution.

  --documentation-prefix=DOCUMENTATION-PREFIX
      The prefix under which the documentation files for the
      NVIDIA driver will be installed.  The default is: '/usr'.

  --kernel-include-path=KERNEL-INCLUDE-PATH
      The directory containing the kernel include files that
      should be used when compiling the NVIDIA kernel module. 
      This option is deprecated; please use
      '--kernel-source-path' instead.

  --kernel-source-path=KERNEL-SOURCE-PATH
      The directory containing the kernel source files that
      should be used when compiling the NVIDIA kernel module. 
      When not specified, the installer will use
      '/lib/modules/`uname -r`/build', if that directory exists. 
      Otherwise, it will use '/usr/src/linux'.

  --kernel-output-path=KERNEL-OUTPUT-PATH
      The directory containing any KBUILD output files if either
      one of the 'KBUILD_OUTPUT' or 'O' parameters were passed to
      KBUILD when building the kernel image/modules.  When not
      specified, the installer will assume that no separate
      output directory was used.

  --kernel-install-path=KERNEL-INSTALL-PATH
      The directory in which the NVIDIA kernel module should be
      installed.  The default value is either
      '/lib/modules/`uname -r`/kernel/drivers/video' (if
      '/lib/modules/`uname -r`/kernel' exists) or
      '/lib/modules/`uname -r`/video'.

  --proc-mount-point=PROC-MOUNT-POINT
      The mount point for the proc file system; if not specified,
      then this value defaults to '/proc' (which is normally
      correct).  The mount point of the proc filesystem is needed
      because the contents of '<proc filesystem>/version' is used
      when identifying if a precompiled kernel interface is
      available for the currently running kernel.  This option
      should only be needed in very rare circumstances.

  --log-file-name=LOG-FILE-NAME
      File name of the installation log file (the default is:
      '/var/log/nvidia-installer.log').

  --tmpdir=TMPDIR
      Use the specified directory as a temporary directory when
      downloading files from the NVIDIA ftp site; if not given,
      then the following list will be searched, and the first one
      that exists will be used: /tmp, /tmp, ., /home/peppe.

  -m, --ftp-mirror=FTP-MIRROR
      Use the specified FTP mirror rather than the default '
      ftp://download.nvidia.com ' when downloading driver
      updates.

  -l, --latest
      Connect to the NVIDIA FTP server '
      ftp://download.nvidia.com ' (or use the ftp mirror
      specified with the '--ftp-mirror' option) and query the
      most recent Linux-x86 driver version number.

  -f, --force-update
      Forces an update to proceed, even if the installer thinks
      the latest driver is already installed; this option implies
      '--update'.

  --ui=UI
      Specify what user interface to use, if available.  Valid
      values for UI are 'ncurses' (the default) or 'none'. If the
      ncurses interface fails to initialize, or 'none' is
      specified, then a simple printf/scanf interface will be
      used.

  -c, --no-ncurses-color
      Disable use of color in the ncurses user interface.

  --no-opengl-headers
      Normally, installation will install NVIDIA's OpenGL header
      files.  This option disables installation of the NVIDIA
      OpenGL header files.

  --force-tls=FORCE-TLS
      NVIDIA's OpenGL libraries are compiled with one of two
      different thread local storage (TLS) mechanisms: 'classic
      tls' which is used on systems with glibc 2.2 or older, and
      'new tls' which is used on systems with tls-enabled glibc
      2.3 or newer.  nvidia-installer will select the OpenGL
      libraries appropriate for your system; however, you may use
      this option to force the installer to install one library
      type or another.  Valid values for FORCE-TLS are 'new' and
      'classic'.

  -k, --kernel-name=KERNEL-NAME
      Build and install the NVIDIA kernel module for the
      non-running kernel specified by KERNEL-NAME (KERNEL-NAME
      should be the output of `uname -r` when the target kernel
      is actually running).  This option implies
      '--no-precompiled-interface'.  If the options
      '--kernel-install-path' and '--kernel-source-path' are not
      given, then they will be inferred from KERNEL-NAME; eg:
      '/lib/modules/KERNEL-NAME/kernel/drivers/video/' and
      '/lib/modules/KERNEL-NAME/build/', respectively.

  -n, --no-precompiled-interface
      Disable use of precompiled kernel interfaces.

  --no-runlevel-check
      Normally, the installer checks the current runlevel and
      warns users if they are in runlevel 1: in runlevel 1, some
      services that are normally active are disabled (such as
      devfs), making it difficult for the installer to properly
      setup the kernel module configuration files.  This option
      disables the runlevel check.

  --no-abi-note
      The NVIDIA OpenGL libraries contain an OS ABI note tag,
      which identifies the minimum kernel version needed to use
      the library.  This option causes the installer to remove
      this note from the OpenGL libraries during installation.

  --no-rpms
      Normally, the installer will check for several rpms that
      conflict with the driver (specifically: NVIDIA_GLX and
      NVIDIA_kernel), and remove them if present.  This option
      disables this check.

  -b, --no-backup
      During driver installation, conflicting files are backed
      up, so that they can be restored when the driver is
      uninstalled.  This option causes the installer to simply
      delete conflicting files, rather than back them up.

  -N, --no-network
      This option instructs the installer to not attempt to
      connect to the NVIDIA ftp site (for updated precompiled
      kernel interfaces, for example).

  --no-recursion
      Normally, nvidia-installer will recursively search for
      potentially conflicting libraries under the default OpenGL
      and X server installation locations.  With this option set,
      the installer will only search in the top-level
      directories.

  -K, --kernel-module-only
      Install a kernel module only, and do not uninstall the
      existing driver.  This is intended to be used to install
      kernel modules for additional kernels (in cases where you
      might boot between several different kernels).  To use this
      option, you must already have a driver installed, and the
      version of the installed driver must match the version of
      this kernel module.

  --no-kernel-module
      Install everything but the kernel module, and do not remove
      any existing, possibly conflicting kernel modules.  This
      can be useful in some DEBUG environments.  If you use this
      option, you must be careful to ensure that a NVIDIA kernel
      module matching this driver version is installed
      seperately.

  --no-x-check
      Do not abort the installation if nvidia-installer detects
      that an X server is running.  Only under very rare
      circumstances should this option be used.

  --precompiled-kernel-interfaces-path=PRECOMPILED-KERNEL-INTERFA
  CES-PATH
      Before searching for a precompiled kernel interface in the
      .run file, search in the specified directory.

  --precompiled-kernel-interfaces-url=PRECOMPILED-KERNEL-INTERFAC
  ES-URL
      If no precompiled kernel interfaces are found within the
      driver package or provided on the file system by the Linux
      distribution, check the specified URL for updates.  NVIDIA
      does not intend to provide updated precompiled kernel
      interfaces, but system administrators might use this for
      distributing precompiled kernel interfaces in a local area
      network.

  -X, --run-nvidia-xconfig
      nvidia-installer can optionally invoke the nvidia-xconfig
      utility.  This will update the system X configuration file
      so that the NVIDIA X driver is used.  The pre-existing X
      configuration file will be backed up.  At the end of
      installation, nvidia-installer will ask the user if they
      wish to run nvidia-xconfig; the default response is 'no'. 
      Use this option to make the default response 'yes'.  This
      is useful with the '--no-questions' or '--silent' options,
      which assume the default values for all questions.

  --force-selinux=FORCE-SELINUX
      Linux installations using SELinux (Security-Enhanced Linux)
      require that the security type of all shared libraries be
      set to 'shlib_t' or 'textrel_shlib_t', depending on the
      distribution. nvidia-installer will detect when to set the
      security type, and set it using chcon(1) on the shared
      libraries it installs.  If the execstack(8) system utility
      is present, nvidia-installer will use it to also clear the
      executable stack flag of the libraries.  Use this option to
      override nvidia-installer's detection of when to set the
      security type.  Valid values for FORCE-SELINUX are 'yes'
      (force setting of the security type), 'no' (prevent setting
      of the security type), and 'default' (let nvidia-installer
      decide when to set the security type).

  --selinux-chcon-type=SELINUX-CHCON-TYPE
      When SELinux support is enabled, nvidia-installer will try
      to determine which chcon argument to use by first trying
      'textrel_shlib_t', then 'texrel_shlib_t', then 'shlib_t'. 
      Use this option to override this detection logic.

  --no-sigwinch-workaround
      Normally, nvidia-installer ignores the SIGWINCH signal
      before it forks to execute commands, e.g. to build the
      kernel module, and restores the SIGWINCH signal handler
      after the child process has terminated.  This option
      disables this behavior.

  --no-cc-version-check
      The NVIDIA kernel module should be compiled with the same
      compiler that was used to compile the currently running
      kernel. The layout of some Linux kernel data structures may
      be dependent on the version of gcc used to compile it. The
      Linux 2.6 kernel modules are tagged with information about
      the compiler and the Linux kernel's module loader performs
      a strict version match check. nvidia-installer checks for
      mismatches prior to building the NVIDIA kernel module and
      aborts the installation in case of failures. Use this
      option to override this check.

  --no-distro-scripts
      Normally, nvidia-installer will run scripts from
      /usr/lib/nvidia before and after installing or uninstalling
      the driver.  Use this option to disable execution of these
      scripts.

Codice: Seleziona tutto

:~$ nvidia-xconfig -A

nvidia-xconfig:  version 1.0  (buildmeister@builder63)  Fri Aug 14 17:54:58 PDT
2009
  The NVIDIA X Configuration Tool.

  This program is used to manipulate X configuration files, specifically to
  enable NVIDIA X driver functionality.

  Copyright (C) 2005 - 2008 NVIDIA Corporation.


  In its normal operation, nvidia-xconfig finds the system X configuration file
  (or generates a new X configuration if it cannot find the system file), makes
  sure the configuration is usable by the NVIDIA X driver, applies any updates
  requested on the commandline, and writes the new configuration to file.

  Please see the NVIDIA README for a description of NVIDIA X configuration file
  options.


nvidia-xconfig [options]

  -c XCONFIG, --xconfig=XCONFIG
      Use XCONFIG as the input X config file; if this option is not specified,
      then the same search path used by the X server will be used to find the X
      configuration file.

  -o OUTPUT-XCONFIG, --output-xconfig=OUTPUT-XCONFIG
      Use OUTPUT-XCONFIG as the output X configuration file; if this option is
      not specified, then the input X configuration filename will also be used
      as the output X configuration filename.

  -s, --silent
      Run silently; no messages will be printed to stdout, except for warning
      and error messages to stderr.

  -t, --tree
      Read the X configuration file, print to stdout the X configuration data
      in a tree format, and exit.

  -v, --version
      Print the nvidia-xconfig version and exit.

  -h, --help
      Print usage information for the common commandline options and exit.

  -A, --advanced-help
      Print usage information for the common commandline options as well as the
      advanced options, and then exit.

  --acpid-socket-path=ACPID-SOCKET-PATH, --no-acpid-socket-path
      Set this option to specify an alternate path to the Linux ACPI daemon
      (acpid)'s socket, which the NVIDIA X driver will use to connect to
      acpid.

  --add-argb-glx-visuals, --no-add-argb-glx-visuals
      Enables or disables support for OpenGL rendering into 32-bit ARGB windows
      and pixmaps.

  --allow-glx-with-composite, --no-allow-glx-with-composite
      Enable or disable the "AllowGLXWithComposite" X configuration option.

  --bandwidth-test, --no-bandwidth-test
      Disable or enable the "NoBandWidthTest" X configuration option.

  --cool-bits=COOL-BITS, --no-cool-bits
      Enable or disable the "Coolbits" X configuration option. Setting this
      option will enable support in the NV-CONTROL X extension for manipulating
      GPU clock settings. Default value is 0.
      WARNING: this may cause system damage and void warranties.

  --composite, --no-composite
      Enable or disable the "Composite" X extension.

  --connect-to-acpid, --no-connect-to-acpid
      Enable or disable the "ConnectToAcpid" X configuration option.  If this
      option is set, the NVIDIA X driver will attempt to connect to the Linux
      ACPI daemon (acpid).  Set this option to off to prevent the X driver from
      attempting to connect to acpid.

  --constant-dpi, --no-constant-dpi
      Enable or disable the "ConstantDPI" X configuration option, which
      controls whether the NVIDIA X driver maintains a constant dots per inch
      (DPI) value by recomputing the reported size in millimeters of the X
      screen when XRandR changes the size in pixels of the X screen.

  --custom-edid=CUSTOM-EDID, --no-custom-edid
      Enable or disable the  "CustomEDID" X configuration option; setting this
      option forces the X driver to use the EDID specified.This option is a
      semicolon-separated list of pairs of display device names and filename
      pairs; e.g "CRT-0:\tmp\edid.bin". Note that a display device name must
      always be specified even if only one EDID is specified. 

  --dac-8bit, --no-dac-8bit
      Most Quadro parts by default use a 10 bit color look up table (LUT) by
      default; setting this option to TRUE forces these graphics chips to use
      an 8 bit (LUT).

  -d DEPTH, --depth=DEPTH
      Set the default depth to DEPTH; valid values for DEPTH are 8, 15, 16, 24,
      and 30.

  --disable-glx-root-clipping, --no-disable-glx-root-clipping
      Disable or enable clipping OpenGL rendering to the root window via the
      "DisableGLXRootClipping" X configuration option.

  --damage-events, --no-damage-events
      Use OS-level events to notify the X server when a direct-rendering client
      has performed rendering that needs to be composited to the screen. 
      Improves performance when using GLX with the composite extension.

  --dynamic-twinview, --no-dynamic-twinview
      Enable or disable support for dynamically configuring TwinView.

  --preserve-driver-name
      By default nvidia-xconfig changes the  display  driver  to "nvidia" for
      all configured X screens; this option preserves the existing driver name
      of each X screen.

  --enable-acpi-hotkeys, --no-enable-acpi-hotkeys
      The "EnableACPIHotkeys" option can be specified to override the NVIDIA X
      driver's default decision to enable or disable ACPI display change hotkey
      events.

  -a, --enable-all-gpus
      Configure an X screen on every GPU in the system.

  --exact-mode-timings-dvi, --no-exact-mode-timings-dvi
      Forces the initialization of the X server with the exact timings
      specified in the ModeLine.

  -E FILE, --extract-edids-from-file=FILE
      Extract any raw EDID byte blocks contained in the specified X log file
      LOG; raw EDID bytes are printed by the NVIDIA X driver to the X log as
      hexidecimal when verbose logging is enabled with the "-logverbose 6" X
      server commandline option.  Any extracted EDIDs are then written as
      binary data to individual files.  These files can later be used by the
      NVIDIA X driver through the "CustomEDID" X configuration option.

  --extract-edids-output-file=FILENAME
      When the '--extract-edids-from-log' option is used, nvidia-xconfig writes
      any extracted EDID to a file, typically "edid.bin" in the current
      directory.  Use this option to specify an alternate filename.  Note that
      nvidia-xconfig, if necessary, will append a unique number to the EDID
      filename, to avoid overwriting existing files (e.g., "edid.bin.1" if
      "edid.bin" already exists).

  --flip, --no-flip
      Enable or disable OpenGL flipping

  --force-generate
      Force generation of a new X config file, ignoring any existing system X
      config file.  This is not typically recommended, as things like the mouse
      protocol, keyboard layout, font paths, etc, are setup by your Unix
      distribution.  While nvidia-xconfig can attempt to infer these values, it
      is best to use your Unix distribution's X config file for the basis of
      anything that nvidia-xconfig creates.

  --force-stereo-flipping, --no-force-stereo-flipping
      Normally, stereo flipping is only performed when a stereo drawable is
      visible. This option forces stereo flipping even when no stereo drawables
      are visible.

  --handle-special-keys=WHEN, --no-handle-special-keys
      Specify when the X server should use the builtin keyboard handler to
      process special key combinations (such as Ctrl+Alt+Backspace); see the X
      configuration man page for details.  The value of WHEN can be 'Always',
      'Never', or 'WhenNeeded'.

  --include-implicit-metamodes, --no-include-implicit-metamodes
      Enable or disable the "IncludeImplicitMetaModes" X configuration option.

  --keyboard=KEYBOARD
      When generating a new X configuration file (which happens when no system
      X configuration file can be found, or the '--force-generate' option is
      specified), use KEYBOARD as the keyboard type, rather than attempting to
      probe the system for the keyboard type.  For a list of possible keyboard
      types, see the '--keyboard-list' option.

  --keyboard-driver=DRIVER
      In most cases nvidia-xconfig can automatically determine the correct
      keyboard driver to use (either 'kbd' or 'keyboard'). Use this option to
      override what nvidia-xconfig detects. Typically, if you are using an
      X.Org X server, use 'kdb'; if you are using an XFree86 X server, use
      'keyboard'.

  --keyboard-list
      Print to stdout the available keyboard types recognized by the
      '--keyboard' option, and then exit.

  --layout=LAYOUT
      The nvidia-xconfig utility operates on a Server Layout within the X
      configuration file.  If this option is specified, the layout named LAYOUT
      in the X configuration file will be used.  If this option is not
      specified, the first Server Layout in the X configuration file is used.

  --logo, --no-logo
      Disable or enable the "NoLogo" X configuration option.

  --logo-path=PATH, --no-logo-path
      Set the path to the PNG file to be used as the logo splash screen at X
      server startup.

  --mode=MODE
      Add the specified mode to the mode list.

  --mode-debug, --no-mode-debug
      Enable or disable the "ModeDebug" X configuration option; when enabled,
      this option causes the X driver to print verbose details about mode
      validation to the X log file.

  --mode-list=MODELIST
      Remove all existing modes from the X configuration's modelist and add the
      one(s) specified in the MODELIST string.

  --remove-mode=MODE
      Remove the specified mode from the mode list.

  --mouse=MOUSE
      When generating a new X configuration file (which happens when no system
      X configuration file can be found, or the '--force-generate' option is
      specified), use MOUSE as the mouse type, rather than attempting to probe
      the system for the mouse type.  For a list of possible mouse types, see
      the '--mouse-list' option.

  --mouse-list
      Print to stdout the available mouse types recognized by the '--mouse'
      option, and then exit.

  --multigpu=MULTIGPU, --no-multigpu
      Enable or disable MultiGPU.  Valid values for MULTIGPU are 'Off', 'On',
      'Auto', 'AFR', 'SFR', 'AA'.

  --multisample-compatibility, --no-multisample-compatibility
      Enable or disable the use of separate front and back multisample
      buffers.

  --nvagp=NVAGP, --no-nvagp
      Set the NvAGP X config option value.  Possible values are 0 (no AGP), 1
      (NVIDIA's AGP), 2 (AGPGART), 3 (try AGPGART, then try NVIDIA's AGP);
      these values can also be specified as 'none', 'nvagp', 'agpgart', or
      'any'.

  --nvidia-cfg-path=PATH
      The nvidia-cfg library is used to communicate with the NVIDIA kernel
      module to query basic properties of every GPU in the system.  This
      library is typically only used by nvidia-xconfig when configuring
      multiple X screens.  This option tells nvidia-xconfig where to look for
      this library (in case it cannot find it on its own).  This option should
      normally not be needed.

  --only-one-x-screen
      Disable all but one X screen.

  --overlay, --no-overlay
      Enable or disable the "Overlay" X configuration option.

  --cioverlay, --no-cioverlay
      Enable or disable the color index overlay.

  --overlay-default-visual, --no-overlay-default-visual
      Enable or disable the "OverlayDefaultVisual" X configuration option.

  --transparent-index=INDEX, --no-transparent-index
      Pixel to use as transparent when using color index overlays.  Valid
      values for TRANSPARENT-INDEX are 0-255.

  -T, --post-tree
      Like the '--tree' option, but goes through the full process of applying
      any user requested updates to the X configuration, before printing the
      final configuration to stdout in a tree format.  Effectively, this option
      just causes the configuration to be printed to stdout as a tree instead
      of writing the results to file.

  --power-connector-check, --no-power-connector-check
      Disable or enable the "NoPowerConnectorCheck" X configuration option.

  --probe-all-gpus, --no-probe-all-gpus
      Disable or enable the "ProbeAllGpus" X configuration option.

  --query-gpu-info
      Print information about all recognized NVIDIA GPUs in the system.

  --randr-rotation, --no-randr-rotation
      Enable or disable the "RandRRotation" X configuration option.

  --render-accel, --no-render-accel
      Enable or disable the "RenderAccel" X configuration option.

  --render-extension, --no-render-extension
      Disable or enable the "NoRenderExtension" X configuration option.

  --rotate=ROTATE, --no-rotate
      Enable or disable the "Rotate" X configuration option.  Valid values for
      ROTATE are 'normal', 'left', 'CCW', 'inverted', 'right', and 'CW'. 
      Rotation can be disabled 

  --screen=SCREEN
      The nvidia-xconfig utility operates on one or more screens within a
      Server Layout in the X configuration file.  If this option is specified,
      the screen named SCREEN in the X configuration file will be used.  If
      this option is not specified, all screens within the selected Server
      Layout in the X configuration file will be used used.

  --separate-x-screens, --no-separate-x-screens
      A GPU that supports multiple simultaneous display devices can either
      drive these display devices in TwinView, or as separate X screens.  When
      the '--separate-x-screens' option is specified, each GPU on which an X
      screen is currently configured will be updated to have two X screens
      configured.  The '--no-separate-x-screens' option will remove the second
      configured X screen on each GPU.  Please see the NVIDIA README
      description of "Separate X Screens on One GPU" for further details.

  --sli=SLI, --no-sli
      Enable or disable SLI.  Valid values for SLI are 'Off', 'On', 'Auto',
      'AFR', 'SFR', 'AA', 'AFRofAA'.

  --stereo=STEREO, --no-stereo
      Enable or disable the stereo mode.  Valid values for STEREO are: 0
      (Disabled), 1 (DCC glasses), 2 (Blueline glasses), 3 (Onboard stereo), 4
      (TwinView clone mode stereo), 5 (SeeReal digital flat panel), 6 (Sharp3D
      digital flat panel), 7 (Arisawa/Hyundai/Zalman/Pavione/Miracube), 8 (3D
      DLP), 9 (3D DLP INV).

  --tv-standard=TV-STANDARD, --no-tv-standard
      Enable or disable the "TVStandard" X configuration option. Valid values
      for "TVStandard" are: "PAL-B", "PAL-D", "PAL-G", "PAL-H", "PAL-I",
      "PAL-K1", "PAL-M", "PAL-N", "PAL-NC", "NTSC-J", "NTSC-M", "HD480i",
      "HD480p", "HD720p", "HD1080i", "HD1080p", "HD576i", "HD576p".

  --tv-out-format=TV-OUT-FORMAT, --no-tv-out-format
      Enable or disable the "TVOutFormat" X configuration option. Valid values
      for "TVOutFormat" are: "SVIDEO" and "COMPOSITE".

  --tv-over-scan=TV-OVER-SCAN, --no-tv-over-scan
      Enable or disable the "TVOverScan" X configuration option. Valid values
      are decimal values in the range 1.0 and 0.0.

  --twinview, --no-twinview
      Enable or disable TwinView.

  --twinview-orientation=ORIENTATION, --no-twinview-orientation
      Specify the TwinViewOrientation.  Valid values for ORIENTATION are:
      "RightOf" (the default), "LeftOf", "Above", "Below", or "Clone".

  --twinview-xinerama-info, --no-twinview-xinerama-info
      Prohibits providing Xinerama information when in TwinView.

  --twinview-xinerama-info-order=TWINVIEW-XINERAMA-INFO-ORDER,
  --no-twinview-xinerama-info-order
      Enable or disable the "TwinViewXineramaInfoOrder" X configuration option.
      TWINVIEW-XINERAMA-INFO-ORDER is a comma-separated list of display device
      names that describe the order in which TwinViewXineramaInfo should be
      reported.  E.g., "CRT, DFP, TV".

  --ubb, --no-ubb
      Enable or disable the "UBB" X configuration option.

  --use-edid, --no-use-edid
      Enable or disable use of the EDID (Extended Display Identification Data)
      from your display device(s).  The EDID will be used for driver operations
      such as building lists of available modes, determining valid frequency
      ranges, and computing the DPI (Dots Per Inch).  This option defaults to
      TRUE (the NVIDIA X driver will use the EDID, when available).  It is NOT
      recommended that you use this option to globally disable use of the EDID;
      instead, use '--no-use-edid-freqs' or '--no-use-edid-dpi' to disable
      specific uses of the EDID.

  --use-edid-dpi, --no-use-edid-dpi
      Enable or disable use of the physical size information in the display
      device's EDID, if any, to compute the DPI (Dots Per Inch) of the X
      screen.  This option defaults to TRUE (the NVIDIA X driver uses the
      EDID's physical size, when available, to compute the DPI).

  --use-edid-freqs, --no-use-edid-freqs
      Enable or disable use of the HorizSync and VertRefresh ranges given in a
      display device's EDID, if any.  EDID provided range information will
      override the HorizSync and VertRefresh ranges specified in the Monitor
      section.  This option defaults to TRUE (the NVIDIA X driver will use
      frequency information from the EDID, when available).

  --use-int10-module, --no-use-int10-module
      Enable use of the X Int10 module to soft-boot all secondary cards, rather
      than POSTing the cards through the NVIDIA kernel module.

  --use-display-device=DISPLAY-DEVICE, --no-use-display-device
      Force the X driver to use the display device specified.

  --use-events, --no-use-events
      Enable or disable "UseEvents" X configuration option. Setting this option
      will enable the X driver to use the system events in some cases when it
      is waiting for the hardware. With this option X driver sets an event
      handler and waits for the hardware through the poll() system call. This
      option defaults to FALSE.

  --virtual=WIDTHxHEIGHT, --no-virtual
      Specify the virtual screen resolution.

  --x-prefix=X-PREFIX
      The X installation prefix; the default is /usr/X11R6/.  Only under rare
      circumstances should this option be needed.

  --xinerama, --no-xinerama
      Enable or disable Xinerama.

  --xvmc-uses-textures, --no-xvmc-uses-textures
      Forces XvMC to use the 3D engine for XvMCPutSurface requests rather than
      the video overlay.

Codice: Seleziona tutto

sudo nvidia-xconfig "opzione"

Codice: Seleziona tutto

sudo chmod u+x NVIDIA-Linux-*

Codice: Seleziona tutto

ERROR: You appear to be running an X server; please exit X before            
         installing.  For further details, please see the section INSTALLING   
         THE NVIDIA DRIVER in the README available on the Linux driver         
         download page at www.nvidia.com.

Codice: Seleziona tutto

 
  ERROR: Installation has failed.  Please see the file
         '/var/log/nvidia-installer.log' for details.  You may find            
         suggestions on fixing installation problems in the README available   
         on the Linux driver download page at www.nvidia.com.

Codice: Seleziona tutto

sudo sh NVIDIA-Linux-* -a -n -N

Codice: Seleziona tutto

NVIDIA.Linux- x86-195.36.08 - pkg1.run.1
NVIDIA.Linux- x86-195.36.08 - pkg1.run.2
NVIDIA.Linux- x86-195.36.08 - pkg1.run.3

Invalid commandline please run ./nvidia-installer --help for usage information

Codice: Seleziona tutto

Scritto il: 02 Marzo, 2010, 23:27:54
Inserito da: DragonEyes
Inserisci citazione
Disinstalla i driver attraverso driver hardware.

Poi da quel link clicca su "download immediato". Sul pulsante "accetta e scarica" fai click con il tasto dx e fai "salva destinazione con nome". Salva il driver nella tua home, poi rinominalo nvidia.run

A questo punto apri il terminale e digita
Codice:

sudo /etc/init.d/gdm stop

premi Alt+F1 e riafi il login (digita la password + invio anche se non te la visualizza).
A questo punto dai
Codice:

sudo sh nvidia.run

Segui le istruzioni. In sostanza puoi rispondere 'sì' a tutte le domande.
Al termine, per riavviare:
Codice:

sudo reboot

è tutto.

Codice: Seleziona tutto

Ricordati, che seguendo questa procedura, ogni volta che utilizzerai un kernel diverso da quello sul quale hai installato i driver, dovrai ripetere l'installazione.

Codice: Seleziona tutto

ai scaricato 50 volte il driver, con quel comando ne trova 3 e non sa quale lanciare, da quello che interpreto.
Scritto il: