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 ... 130 131 [132] 133 134 ... 212
3D-Tech News Around The Web / NVIDIA GPUDirect 2.0
« on: August 09, 2011, 05:50:50 PM »
First released in June 2010, GPUDirect v1.0 is already supported by InfiniBand solutions such as those available from Mellanox and QLogic, and other vendors are adding support for GPUDirect in their hardware and software products now.

Applications using GPUDirect v1.0 with network and storage devices can realize up to 30% performance improvement in data transfer times on NVIDIA Tesla™ and Quadro™ products. GPUDirect enables this significant speed-up by allowing network adapters, solid-state drives (SSDs) and other devices to directly access CUDA memory, eliminating unnecessary system memory copies and CPU overhead.

GPUDirect v2.0 adds support for peer-to-peer (P2P) communication between GPUs over PCIe in the same system, and lays the foundation for P2P communication between GPUs and other devices in a future release.

For more information, see the GPUDirect Technology Overview presentation.

3D-Tech News Around The Web / Blender 2.59 Release Candidate 1 Released
« on: August 09, 2011, 05:04:48 PM »
A first Release Candidate for Blender 2.59 has just been released. Check out the full changelog, or head straight for the download.

Rig used for screenshot: "Matt"

3D-Tech News Around The Web / [Demo Scene] Assembly Summer 2011
« on: August 08, 2011, 08:39:58 PM »
.___ _    _______  __ _   _                                                 _ .
|  _______\     /_________     _____     aSSEMBLY sUMMER 2o11 - 4.-7.8.2011 /(|
___\____   \______  \    /____/   _/____   ___  ___                           :
\    \_/    :  |_/   \______     _/    /._/   \/   \___________               .
.\    |     _        |   |_/     \      |   _       :   __    /______
|     |     |________|         :  \     |    \__/       |/   /|     |____.     
|_____|-----' H7/dS! |_________|---\____|_____||    :    ._   |     |    |_____
:  .  :                                        `----|    |/   |     |    |    |
|  |        -uggg  ggggu-ugggg -ugggg -ugggg        |_________|     |____     |
|  `------ -  XXX  XXXX   XXXX   XXXX   XXXX - -- ----------- |______/ \______|
|____ _ XXXX"""""  XXXX   XXXX   XXXX   XXXX _ ____ __________________________:
        XXXXu__u   XXXX___XXXX   XXXX   XXXX

ASSEMBLY Summer 2011 was held on 4.8.2011 - 7.8.2011 in Helsinki, Finland.

Downloads via Pouet
My favorites are the 4K demos daydream, cloudlight and invaders.

Note: your antivirus software probably complains about crinkler packed demos, MSE takes ages to check them  :P

Despite the fact that cube maps are defined on the spherical domain, standard cubemap mipchain generation techniques perform filtering independently on each cube face.  The main problem with this approach is that no information is propagated across edges, thus creating undesirable discontinuities along the cube face edges.  A limitation of nearly all cube mapping hardware which makes the seam problem substantially worse is the fact that the bilinear texel filtering is not able to fetch across cube faces thus producing a hard seam artifact in addition to introducing aliasing artifacts.  These two compounding problems limit the usefulness of cubemapping.

Our pre-processing approach alleviates this problem using two techniques.  The first, angular extent filtering, defines each tap’s filter kernel using an angular extent around the center tap as opposed to a fixed per-face pixel-based extent.  The advantages of angular extent filtering are that the filtering kernels used for all taps have a constant solid angle as well as have the ability to pull texels from multiple faces.   In addition to this, the filtering takes into account the solid angle subtended by each tap in the filter.  Also, by varying the angular extent of the filter used to generate the base mip-level, pre-convolved reflections for a variety of materials can be precomputed. 

The second technique, edge seam fixup, uses a per mip-level seam averaging and smoothing algorithm in order to effectively hide the seam artifacts.  This enables the lower-resolution cubemap miplevels (even 2x2 and 4x4) to be directly used for tarnished metal shaders, matte surfaces shaders, and even for diffuse lighting.   In addition to this, by providing good quality cubemaps throughout the mipchain, miplevel selection via the texCUBEbias, and texCUBELOD instructions can be used for selective blurring.

CubeMapGen is a tool for importing, filtering, visualizing, and exporting cubemaps, its list of features include:

Cubemap Filtering and MipChain Generation: Angular extent filtering and edge seam fixup provides new high quality filtering results for prefiltered environment mapping, and cubemap mipchain generation.

Interactive Cubemap Assembly: Cubemaps can be assembled from separate face images that can be loaded into each cubemap face and interactively flipped vertically, horizontally, and diagonally in order to align the faces.

Cubemap Import: Import entire cube maps from HDRShop vertical cross cubemap images, DirectX .dds cubemap files, and separate image files

Cubemap Export: Export cubemap with mipchain to single .dds file, cube cross images for each mip level, or a collection of separate images for each cubemap face for each mip level.  Also supported is a specialized DXT compressor which allows for the creation of seamless DXT compressed cubemaps.

Command Line Interface: Allows you to use CubeMapGen as a command line tool for processing cubemaps in scripts and batch files.

HDR Image Support: All filtering is performed in high dynamic range.

HDRShop v1.0 plugin functionality: CubeMapGen also functions as an HDRShop plugin that provides the filtering functionality of CubeMapGen to HDRShop.

Download here

3D-Tech News Around The Web / OpenGL 4.2 review @ g-truc
« on: August 08, 2011, 07:01:59 PM »
The Khronos Group has released the OpenGL 4.2 at Siggraph 2011.

To celebrate this event g-truc wrote an OpenGL 4.2 review which hopefully will help to get started quickly with all the goodness of OpenGL 4.2.

Quote from:
The Khronos™ Group today announced the immediate release of the OpenGL® 4.2 specification, bringing the very latest graphics functionality to the most advanced and widely adopted cross-platform 2D and 3D graphics API.  OpenGL 4.2 integrates developer feedback and continues the rapid evolution of this royalty-free specification while maintaining full backwards compatibility - enabling applications to incrementally use new features, while portably accessing state-of-the-art graphics processing unit (GPU) functionality across diverse operating systems and platforms.

The OpenGL 4.2 specification has been defined by the OpenGL ARB (Architecture Review Board) working group at Khronos, and includes the GLSL 4.20 update to the OpenGL Shading Language.  The OpenGL 4.2 specification contains new features that extend functionality available to developers and enables increased application performance.  The full specification is available for immediate download.

Answer ID 3021   |    Published 08/01/2011 09:18 PM   |    Updated 08/02/2011 08:32 AM

Changes in display driver 280.19 and higher may result in the GPU voltages reported by a graphics card diagnostic utility to differ from previous older drivers.  This is expected behavior.  Older NVIDIA display drivers may have reported incorrect voltages which is now corrected in newer drivers.  The changes in the new 280.19 display driver does not change the actual voltages of the GPU; only what is reported by a graphics card diagnostic utility to a user. 

Windows driver version 280.28 and Linux drivers version provide full support for OpenGL 4.2 and GLSL 4.20 on capable hardware. This driver also supports several new OpenGL extensions for both 4.2-capable GPUs and older GPUs. The driver download links are at the bottom of this page.

OpenGL 4.2 Driver Release Notes

You will need any ... Fermi based GPU to get access to the OpenGL 4.2 and GLSL 4.20 functionality

As it came unannounced, no GPU tool recognizes 4.2 yet properly.

The OpenGL ES 2.0 Emulator is a library that maps OpenGL ES 2.0 API calls to the OpenGL 2.0 API.

By running on a standard PC, the emulator helps software development and testing of OpenGL ES 2.0 applications since no embedded platform is required. The emulator requires a graphics card that supports at least OpenGL 2.0.


     Changes from versions 1.3.0 Milestone 1

        The OpenGL ES Emulator now includes OpenGL ES 1.1 and OpenGL ES 2.0 support in the package.
        Both can be used in a single application.

        The OpenGL ES Emulator now has full Mesa support.

        The OpenGL ES Emulator now includes a high performance mode; in this mode the OpenGL ES
        Emulator skips some of the checks on argument values.

Changes between versions 1.2.0 and 1.2.1 and version 1.3.0 Milestone 1

        The emulator now contains preliminary Mesa support on Windows.

        The emulator now supports the GL_OES_texture_npot and GL_EXT_texture_format_BGRA8888
        OpenGL ES extensions.

ZiiLabs has been offering an early access program for OpenCL SDK since last year...
Now they are planning to make their Android NDK a standard component, it’s a good time to ask them some questions.
VH: The last few months the number of searches on my site for ARM or Android and OpenCL has increased a lot. Do you see an increased interest too?
TL: Yes, we see a growing interest in finding ways to enhance the CPU performance of the ARM cores by taking advantage of the programmability and floating point performance of the media processor.

Full story at StreamComputing

3D-Tech News Around The Web / HWiNFO32/64 v3.83 Released
« on: August 03, 2011, 06:12:21 PM »
Changes in HWiNFO32 v3.83 - Released on: Aug-03-2011:

    Fixed fan speed reporting on some NCT6776 boards.
    Added support of Nuvoton NCT7511Y, NCT7802Y and NCT7904D hardware monitors.
    Enhanced sensor monitoring on GIGABYTE A75 series.
    Enhanced sensor monitoring on ASUS F1A75 series.
    Added reporting of CPU Microcode Update Revision for AMD Family 10h and later models.
    Added support of SMSC LPC47B397, SCH5307 and SCH5317 hardware monitoring.
    Enhanced sensor monitoring on ASUS Maximus IV Extreme-Z.
    Added reporting of Chassis Intrusion for Winbond/Nuvoton, ITE and Fintek sensors.
    Enhanced sensor monitoring on ASUS Maximus IV GENE-Z.
    Enhanced sensor monitoring on ASUS M4A78LT.
    Enhanced sensor monitoring on ASUS M5A87 and M5A88 series.
    Enhanced sensor monitoring on ECS H67H2.
    Improved Toshiba EC sensor monitoring.
    Added support of Shared Memory for publishing sensor information.

Note: all sensor data can be sent to Rivatuner display server, the number of lines seems to be limited though.

3D-Tech News Around The Web / S3 Chrome drivers WHQL
« on: August 03, 2011, 12:07:20 AM »
07/22/2011 Version:

Release Type: Microsoft Certified


- Bug Fix


- White window when launch ogl 3.x demo in GPU caps Viewer 1.8.2 (58217-0267)

- Black screen when click switch user with BD playback flying on PowerDVD10 v1516 on HDMI(59849-0267)

- Black screen when click switch user with DVD/BD playback flying on PowerDVD10 v1516 on HDMI((60244-0267)

> Chrome 500 Series: Windows 7 Driver (07/22/2011)
> Chrome 400 Series: Windows 7 Driver (07/22/2011)

3D-Tech News Around The Web / Intel Haswell will support DirectX 11.1
« on: August 02, 2011, 06:48:46 PM »
Sweclockers posted a roadmap that reveals capabilities of upcoming Intel hardware...

3D-Tech News Around The Web / AMD APP KernelAnalyzer 1.9
« on: August 02, 2011, 06:11:14 PM »
AMD APP KernelAnalyzer is a static analysis tool to compile, analyze and disassemble an OpenCL, Brook+ or IL kernel for AMD Radeon™ GPUs.

What's New in Version 1.9

    Support for Catalyst™ driver 11.4 - 11.7.
    Support for AMD APP SDK 2.5.
    Fixed a crash dumping x86 assembler.

3D-Tech News Around The Web / A Gentle Introduction to OpenCL
« on: August 02, 2011, 05:36:49 PM »
With great power comes great complexity. OpenCL, with its wealth of features, makes it possible to code routines capable of executing on devices ranging from graphics cards to supercomputers. But to take full advantage of OpenCL, you need to have a thorough understanding of host applications and kernels.

This article has presented two analogies intended to ease the learning process, but in the end, it's the code that matters. The OpenCL API takes time and effort to understand, but once you've ascended the learning curve, you'll be able to tap into computing performance that exceeds anything a regular C/C++ programmer could hope for.

Full story at Dr. Dobb's Journal

The major changes (2011-08-02):

1 – “monitor only” and close/reopen the server - it is performing in read only mode (access denied to OC functions) but the menu item is not checked as it should be
2 – Update the remote server to deal with this regional language issue.

Server is yet unstable under Vista; once you have it running, don't mess with it's menus  :P

3D-Tech News Around The Web / Triton Ocean SDK 1.0
« on: August 01, 2011, 05:13:04 PM »
We've just released version 1.0 of the Triton Ocean SDK at Triton integrates with OpenGL, DirectX9, or DirectX11 based engines to provide real-time simulations of oceans and large bodies of water. It maintains high framerates - hundreds per second on modern hardware - by automatically detecting the GPGPU capabilities of the system and adapting to it. Triton ships with the ability to take advantage of CUDA, OpenCL, DirectX11 Compute Shaders (DirectCompute,) as well as multi-core CPU technologies such as the Intel Integrated Performance Primitives and OpenMP. By integrating GPGPU API's with Triton's vertex and fragment shaders, the entire simulation of thousands of waves at once can take place on the GPU from setting up the waves in the frequency domain, conducting the inverse fast Fourier transforms, and applying this data to a projected grid and applying a full Fresnel refraction and reflection shader to the final output, with foam effects.

In addition to the FFT-based ocean simulation, Triton also offers a simpler sum-of-Gerstner-waves mode that's appropriate for shallower water simulations. It also allows the user to pass in an environmental cube map for more accurate water reflections.

Triton's tailored for serious game applications as well, and is able to take any given wind conditions or Beaufort scale and produce matching water visuals automatically.

Maintaining system compatibility was one of the biggest challenges during the development of Triton; GPGPU API's are still young, and much care had to be taken when selecting the API best suited to the user's system. Integrations between the GPGPU and OpenGL / DirectX API's were particularly challenging. We avoided runtime DLL errors by encapsulating our GPGPU code within DLL's of their own, and dynamically loading them only after verifying that their DLL dependencies are present on the system.

This is Triton's initial release, and we're definitely interested in constructive feedback! A demo, free evaluation SDK, and licensing information are available from our website.

Actually the ocean ain't rendered with Intel GPU.

3D-Tech News Around The Web / Fraps 3.4.6
« on: August 01, 2011, 04:41:29 PM »

Fraps 3.4.6 - 31st July 2011
- Fixed overlay color in Fear 3
- Fixed detection of IL2: Cliffs of Dover and Bloodline Champions
- Fixed counter appearing in Thunderbird and Pale Moon applications
- Other minor bug fixes

3D-Tech News Around The Web / PhysX 9.11.0621 System Software
« on: July 29, 2011, 05:26:55 PM »
Just 13 months after last release, here is a new one.

Download here: PhysX_9.11.0621_SystemSoftware.msi

NV Path Rendering
NVIDIA's Release 275 drivers for Windows, Linux, FreeBSD, and Solaris provide full GPU-acceleration of a style of 2D graphics known as path rendering.  Path rendering specifies a scene as a sequence of resolution-independent outlines, known as paths, that can be filled or stroked.  Such paths can be painted with constant colors, linear or radial gradients, or images.  Unlike bitmap images, apath rendering content can be arbitrarily zoomed and rescaled without pixelized results.  Path rendering contents are also easy to edit and animate because an artist can manipulate or edit the underlying paths that make up the scene instead of having to manipulate individual pixels

Pages: 1 ... 130 131 [132] 133 134 ... 212