Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Stefan

Pages: 1 ... 113 114 [115] 116 117 ... 212
The CUDA 5 Preview release is now available to you via the CUDA Registered Developer program, find out more about the CUDA 5 Toolkit,

To download, we recommend and invite you to use the new CUDA Registered Developer experience on

CUDA 5 introduces several new tools and features that make it easier than ever to add GPU acceleration to your applications.

New Nsight, Eclipse Edition helps you explore the power of GPU computing with the productivity of Eclipse on Linux and MacOS

    •  Develop, debug, and profile your GPU application all within a familiar Eclipse-based IDE
    •  Integrated expert analysis system provides automated performance analysis
       and step-by-step guidance to fix performance bottlenecks in the code
    •  Easily port CPU loops to CUDA kernels with automatic code refactoring
    •  Semantic highlighting of CUDA code makes it easy to differentiate GPU Code from
       CPU code
    •  Integrated CUDA code samples makes it quick and easy to get started
    •  Generate code faster with CUDA aware auto code completion and inline help

GPU callable libraries now possible with GPU Library Object Linking

    •  Compile independent sources to GPU object files and link together into a larger
    •  Design plug-in APIs that allow developers to extend the functionality of your kernels
    •  Efficient and familiar process for developing large GPU applications
    •  Enables 3rd party ecosystem for GPU callable libraries

GPUDirect RDMA provides fastest possible communication between GPUs and other PCI-E devices

    •  Direct memory access (DMA) supported between NIC and GPU without the need for
       CPU-side data buffering
    •  Significantly improved MPISendRecv efficiency between GPU and other nodes in a
    •  Eliminates CPU bandwidth and latency bottlenecks
    •  Works with variety of 3rd party network and storage devices

Dynamic Parallelism enables programmers to easily accelerate parallel nested loops on the new Kepler GK110 GPUs

    •  Developers can easily spawn new parallel work from within GPU code
    •  Minimizes the back and forth between the CPU and GPU
    •  Enables GPU acceleration for a broader set of popular algorithms, including adaptive
       mesh refinement used in aerospace and automotive computational fluid
       dynamics (CFD) simulations
    •  Supported natively on Kepler II architecture GPUs, preview programming guide and
       whitepaper available today

3D-Tech News Around The Web / AMD APP Profiler 2.5
« on: May 16, 2012, 10:32:48 PM »
What's New in Version 2.5

    Support for AMD APP SDK v2.7.
    Support for OpenCL™ 1.2.
    Support for collecting performance counters on APU devices.
    Full support for profiling with AMD Radeon™ HD7000 series GPUs based on Graphics Core Next/Southern Islands:
        Added support for kernel occupancy analysis.
        Added support for collecting performance counters for DirectCompute (DirectX 11) applications.
        Addition of SALUBusy counter.
        Fixed value reported for VALUBusy counter.
        The values reported for LDSFetchInsts and LDSWriteInsts counters were inaccurate on AMD Radeon™ HD7000 series GPUs; thus for those GPUs, those two counters have been replaced by a single LDSInsts counter.
        Fixed display of kernel ISA.
    Improved OpenCL™ analysis module:
        Added detection of deprecated OpenCL™ APIs.
    Added support for showing source and destination location, as well as zero-copy status for memory transfers initiated using clEnqueueMapBuffer or clEnqueueMapImage. This information is displayed in the API Trace view.
    Added support for Microsoft® Visual Studio® projects that use User-defined Macros in the project settings.
    Fixed the --workingdirectory (-w) command line switch (set current directory) on Linux.
    Fixed some problems with importing previously-generated profile results into Microsoft® Visual Studio®.
    Changed the default installation directory on Windows to %PROGRAMFILES(X86)%\AMD\AMD APP Profiler to make it more consistent with other AMD tools (i.e gDEBugger and CodeAnalyst)

Imagination makes the case for mobile OpenCL

Imagination was showing off GPU compute on a cell phone chip at GDC, physics in your pocket. If you think that upcoming GPUs will allow this to be reality, it was running on a current TI OMAP 4430 chip.

The idea is the same one that everyone is talking about, offload certain functions from a CPU to a GPU to speed them up and possibly save some energy while doing so. In a desktop, CPU power is in abundance, and power use is not really a big concern. While laptops are a little more energy conscious, anything that will tax a laptop CPU for long time periods will probably be done while plugged in. This means that GPU compute and OpenCL are used mainly to address problems that CPUs are really bad at, or just take a long time.

May 2012 UDK Beta Upgrade Notes
New Features and Functionality

    RealD Integration
        Incorporates best-of-breed stereoscopic 3D gaming technology in the Unreal Engine which allows for PC and console games to be rendered in stereo 3D
        3D experience improved by allowing management of camera separation and 3D depth effects in real time
        Depth Budget Allocation and AutoCAM available with the RealD Game Developer Toolkit
        Free demo showcasing the implementation available at
    Advanced Environment Color
        A simplified directional environment light where the environment color depends on direction
        It is blended between 2 values: ‘environment color’ and ‘environment sun color’
        For example, the side of the sky with the sun could have a slightly warmer and brighter hue compared to the opposite side

3D-Tech News Around The Web / (Android) ZiiLabs Mandelmark
« on: May 14, 2012, 06:14:37 PM »
MandelMark is an Android application that has been developed by ZiiLABS to measure the floating point performance of modern mobile processors by visualising the solutions to the Mandelbrot set using an OpenGL ES 2.0 shader program. ...
MandelMark Explained

    Two triangles are drawn on the screen and an OpenGL 2.0 shader is used to render the MandelBrot set
    For each pixel the shader program is called to calculate the colour value of the pixel
    The shader loops up to 255 times to calculate if the pixel is within the Mandlebrot set
    The colour of the pixel is based on the number of loops taken to confirm it lies outside the set, or black if it lies within it
    The shader defaults to 32-bit floating point for the rendering if available

11 minutes with ARMv7/Tegra2
5 minutes with ARMv6/Adreno200 

3D-Tech News Around The Web / MSI Afterburner 2.2.1 (2012-05-14)
« on: May 14, 2012, 04:42:15 PM »
Hi guys, the Afterburner 2.2.1 official version and Kombustor 2.3.0 are ready for you to download. this version supports the control of South Island and Kepler card, if you get the card, you may try to play it.

Changes list includes:

• Added reference design AMD RADEON 7850 series graphics cards support
• Added NVIDIA GeForce GTX 690 series graphic cards support
• Added big editions of default and legacy skins
• Updated localizations

Download link

We've updated the Dynamic Resolution Rendering sample with an optimisation to the temporal anti-aliasing (TAA) upscale pass and added a few new features for improved qualitative comparisons focused on TAA.

You can find the updated sample here.

The changes, listed below, are detailed in this pdf.

    Added optimization for temporal anti-aliasing (TAA) when using motion blur, which decreases the time this pass takes by approximately 40%.
    Corrected the velocity scale factor so that if the pixel in the last frame was 1 pixel away it is attenuated by 1⁄2. Previously a constant scale factor was used, which was chosen for the default resolution.
    Added a zoom box for viewing scaled portions of the screen.
    Added a basic TAA mode that does no velocity scaling so any ghosting issues can be found.
    Improved pause mode that correctly renders the previous and current frames to show TAA artifacts clearly in basic mode.
    Added a motion blur toggle so that the effect of motion blur and TAA can be properly considered.
    Added a symmetric TAA toggle.
    Changed pause key from Space to P.

OpenGL Extensions Viewer for Android displays the vendor name, the version, the renderer name and the extensions for OpenGL ES 2.0.

From the makers of the famous OpenGL Extensions Viewer on PC and Mac, the Android version can access each extension documentation, renderers capabilities.

Ideal if you are developing application for Android,

CPU information, Android OS revision are also reported.

Main key features:

- Reports Android version, model and CPU.

- Reports OpenGL ES information and gives access to OpenGL Extension extension specification online (requires internet connection).

- Submit your OpenGL EX information into to the Cloud storage and check the results on the Mac and PC version.

- Free of charge.

- Requires Android 2.2 minimum.

Follow us on Twitter:

What's New

What's in this version:

    Updated for Android 3.2

Screenshot from emulator (GPU emulation is still screwed in R20 preview 1)

3D-Tech News Around The Web / ASUS GPU Tweak v2.1.24
« on: May 10, 2012, 07:08:39 PM »
ASUS GPU Tweak v2.1.24
1. support GTX670 and GTX690.
2. OC range enhance support NVIDIA.

AMD Catalyst™ 12.4a Hotfix Driver 

Last Updated
Article Number

Issue Description:

We’ve recently identified an issue where some of our legacy graphics products are producing certain artifacts in Diablo III with our latest release of Catalyst™ 12.4. This issue only affects users of ATI Radeon™ HD 2400, 2600, 2900, 3400, 3600 and 4500 graphics cards under Windows® Vista® and Windows 7.

New in Release 270.00.00f06:

    Contains performance improvements and bug fixes for a wide range of applications.

    This driver update is for Mac Pro 5,1 (2010), Mac Pro 4,1 (2009) and Mac Pro 3,1 (2008) users only.

3D-Tech News Around The Web / TechPowerUp GPU-Z v0.6.2
« on: May 03, 2012, 02:27:58 AM »
TechPowerUp GPU-Z v0.6.2

Revision History

    Fixed boost clocks being displayed on Fermi

3D-Tech News Around The Web / TechPowerUp GPU-Z v0.6.2
« on: May 02, 2012, 05:15:35 PM »
Filename: GPU-Z.0.6.1.exe

GPU-Z is a lightweight utility designed to give you all information about your video card and GPU.

Revision History

    Added PowerColor giveaway
    Added support for NVIDIA GeForce GTX 690, GTX 670, GT 640, GT 630, GeForce 605, Quadro 3000M, GTX 675M, GTX 670M
    Added support for AMD Radeon HD 7970M, HD 7450
    Added full support for Intel Sandy Bridge and Ivy Bridge integrated graphics
    Fixed ROP count on GK107
    Improved NVIDIA Kepler Boost Clock detection
    Fixed crash on AMD Southern Islands when no graphics card driver installed
    Fixed installer not working properly on Windows XP
    Disabled broken BIOS save on NVIDIA Kepler
    BIOS files are saved as .rom now
    Fixed sensor graph redraw and improved its performance

Downloads at NVIDIA
    Fixed a bug that caused OpenGL programs to leak file descriptors when calling exec() on themselves.
    Fixed rendering corruption at the start of new X server generation.
    Added X driver support for RandR 1.2 and RandR 1.3. See "Support for the X Resize and Rotate Extension" in the README for details.
    Extended the MetaMode X configuration option syntax with the following new attributes:
        "Rotation" : specifies the display device's rotation
        "Reflection" : specifies the display device's reflection
        "Transform" : specifies a 3x3 transformation matrix to be applied to the display device
        "ViewPortOut" : specifies the region of the mode in which to display pixels
        "ViewPortIn" : specifies the size of the region in the X screen to display in the ViewPortOut
        For example, "DFP-0: nvidia-auto-select { Rotation=left }".
        See "Configuring Multiple Display Devices on One X Screen" in the
        README for details.
    Removed the "TwinView" X configuration option; the functionality it provided is now enabled by default.

    Previously, the NVIDIA X driver only enabled one display device for an X screen unless "TwinView" was enabled. Now, the NVIDIA X driver enables, by default, as many display devices as the GPU supports driving simultaneously. To limit how many display devices are driven by an X screen, use the "UseDisplayDevice" X configuration option.
    Added a "CurrentMetaMode" attribute to the nvidia-settings command line, to query and set the current MetaMode.

    As an example, these two commands are equivalent:

        xrandr --output DVI-I-2 --mode 1280x1024 --pos 0x0 --output DVI-I-3 --mode 1920x1200 --pos 1280x0

        nvidia-settings --assign CurrentMetaMode= "DVI-I-2: 1280x1024 +0+0, DVI-I-3: 1920x1200 +1280+0"
    Removed overscan compensation configurability from NV-CONTROL and nvidia-settings. This can be configured, with finer granularity, through the ViewPortIn and ViewPortOut MetaMode attributes. See "Configuring Multiple Display Devices on One X Screen" in the README for details.
    Altered mode validation such that for digital display devices the X driver only allows, by default, modes which are reported in the EDID. Previously, the NVIDIA X driver allowed other modes, but implicitly scaled those other modes to one of the EDID modes. Now, only the modes in the EDID are validated and the X driver does not do any implicit scaling. Any desired scaling can be configured explicitly through the new "ViewPortIn" and "ViewPortOut" MetaMode attributes. See "Configuring Multiple Display Devices on One X Screen" in the README for details.
    Removed Flat Panel Scaling configurability in nvidia-settings. Any desired scaling can be configured through the new "ViewPortIn" and "ViewPortOut" MetaMode attributes.
    Hotplug events (specifically, the RRNotify_OutputChange RandR event) are now generated when display devices are connected and disconnected from the GPU. Many desktop environments automatically resize the X desktop in response to these events.
    Added display device name aliases, such that X configuration options that use display device names can refer to a display device by one of several names, including the RandR Output name for the display device. The X log reports the list of aliases for each display device.
    Updated EDID parsing to include more complete support for EDID 1.4 and more recent versions of CEA-861.
    Removed the "Rotate" X configuration option. This was used to statically rotate the X screen. Its functionality is replaced by the "Rotation" MetaMode attribute and RandR 1.2 rotation support. See the README for details.
    Removed the "RandRRotation" X configuration option. This enabled configurability of X screen rotation via RandR 1.1. Its functionality is replaced by the "Rotation" MetaMode attribute and RandR 1.2 rotation support. See the README for details.
    Removed support for the following NV-CONTROL attributes:
    Improved rendering performance for RENDER bitmap text.
    Enabled the OpenGL "Sync to VBlank" option by default.
    Added a new option, --restore-original-backup, to nvidia-xconfig. nvidia-xconfig creates a backup of the original X configuration file when modifying an X configuration file that does not appear to have been previously modified by nvidia-xconfig. This option restores a backup of the original X configuration file, if such a backup is found.
    Expose the following additional FSAA modes via NV-CONTROL, nvidia-settings, and through X visuals and GLXFBConfigs:

        16X multisample FSAA on all GeForce GPUs

        Coverage sample FSAA on G80 and above GeForce GPUs

        32X multisample FSAA on G80 and above Quadro GPUs

        64X multisample FSAA on Fermi and above Quadro GPUs
    Enabled conformant texture clamping by default in OpenGL.
    Removed support for the GVO Clone mode NV-CONTROL attributes:
    Added a new, higher resolution icon for nvidia-settings.
    Updated the NVIDIA X driver's handling of X configuration options that affect an entire GPU, not just a particular X screen running on a GPU (e.g., "NoPowerConnectorCheck"): for such X configuration options, the X driver will now honor the option on any of the X screens configured on the GPU. Prior to this change, the NVIDIA X driver only honored such options on the first X screen configured on a GPU.
    Added a checkbox to nvidia-settings to control the texture clamping attribute. When the box is checked, OpenGL textures are clamped according to the OpenGL specification. When it is unchecked, GL_CLAMP is remapped to GL_CLAMP_TO_EDGE for borderless 2D textures.
    Removed the "Display" and "X Screen" tabs from the "X Server Display Configuration Page" of nvidia-settings, and added a new "Selection" dropdown menu for selecting X screens or display devices. This makes it easier to select X screens/Displays that are hidden.

Device fission is an addition to the OpenCL* specification that gives more power and control to OpenCL programmers over managing which computational units execute OpenCL commands. Fundamentally, device fission allows the sub-dividing of a device into one or more sub-devices, which, when used carefully, can provide a performance advantage, especially when executing on CPUs.

The newly released Intel® SDK for OpenCL* Applications 2012 is a comprehensive software development environment for OpenCL applications on 3rd generation Intel® Core™ processor family-based platforms. This SDK also provides developers with the ability to develop and target OpenCL applications on Intel CPUs of previous generations using both the Windows* and Linux* operating systems.

3D-Tech News Around The Web / The Escape Short movie
« on: April 30, 2012, 09:12:45 PM »
The Escape Short movie

Our first short movie.
(c) Enspire Studio, all rights reserved.

Just a nice short CG movie  :)

Here are some screenshots from the benchmark the NVIDIA boss talked about at NGF Shanghai, release is scheduled for May.

EDIT: someone already posted the video at Youtube

Cloth moving in the wind, electrical sparks

Falling leaves

Explosions with particles and shockwave (look at he tower in the background)

Dark Meadow: The Pact – Live The Nightmare on NVIDIA Tegra 3
Ever had one of those nightmares from which you couldn’t wake up? You twist and turn through narrow halls facing unimaginable horrors, hoping only to wake up from this terrifying phantasm. Only this is no nightmare, it’s all too real. That’s how it goes in the latest Tegra 3 gaming title, Dark Meadow.

Direct Playstore link

This is the second Android game powered with Unreal Engine in 2 years (as opposed to Unity3D where you see a dozen new games each week)

Briefly it's a horror game is like "Silent Hill", the most scary thing though is the german localization  :P

Pages: 1 ... 113 114 [115] 116 117 ... 212