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.

Topics - Stefan

Pages: 1 ... 104 105 [106] 107 108 ... 160
3D-Tech News Around The Web / OpenGL Extensions Viewer 3.33
« on: October 28, 2010, 04:55:39 PM »
OpenGL Extensions Viewer Updated to the version 3.33 :

Fixed compatibility with some graphics chipset, fixed rendering test selection, fixed Aero switch on/off.

Actually rendering test selection appears messed up with Forceware 260.99  ???

We managed to gain an extra 40MHz on the graphics memory with the aforementioned memory voltage, and a corresponding improvement in graphics performance...
As usual, shall not be responsible for any damages arising from the users of this tried-and-tested guide. If it befuddles you, forward it to someone else and forget about it.

Full story here

3D-Tech News Around The Web / NVIDIA driver 259.81 Quadro WHQL available
« on: October 28, 2010, 04:08:38 PM »
New in Release 259.81:

    * Support for newly announced Quadro 6000, Quadro 5000, Quadro 4000, Quadro 2000 and Quadro 600.
    * Adds support for CUDA 3.1 for improved performance in GPU Computing applications.  See for more details.
    * Adds support for OpenGL 4.0
    * Adds support for DirectX 11
    * ECC State
          o New page for GPUs that support ECC (error correction code). The page lets you
                + Change the Error Correction Code (ECC) state for GPUs.
                + View GPU memory details.
    * EDID Management
          o Accessible from the View System Topology page, this feature lets you:
                + View the connector information
                + Get an EDID from a connected display and save it to disk
                + Force an EDID at a connector
                + Remove a forced EDID from a connector
    * The SDI Capture page is now available with Windows Vista and Windows 7 drivers.

Supported devices:

NVIDIA_DEV.004E.01 = "NVIDIA Quadro FX 4000"
NVIDIA_DEV.009D.01 = "NVIDIA Quadro FX 4500"
NVIDIA_DEV.00CD.01 = "NVIDIA Quadro FX 3450/4000 SDI"
NVIDIA_DEV.00CE.01 = "NVIDIA Quadro FX 1400"
NVIDIA_DEV.00F8.01 = "NVIDIA Quadro FX 3400/4400"
NVIDIA_DEV.014A.01 = "NVIDIA Quadro NVS 440"
NVIDIA_DEV.014D.01 = "NVIDIA Quadro FX 550"
NVIDIA_DEV.014E.01 = "NVIDIA Quadro FX 540"
NVIDIA_DEV.0165.01 = "NVIDIA Quadro NVS 285"
NVIDIA_DEV.0197.01 = "NVIDIA Tesla C870"
NVIDIA_DEV.019D.01 = "NVIDIA Quadro FX 5600"
NVIDIA_DEV.019E.01 = "NVIDIA Quadro FX 4600"
NVIDIA_DEV.01DB.01 = "NVIDIA Quadro NVS 120M"
NVIDIA_DEV.01DE.01 = "NVIDIA Quadro FX 350"
NVIDIA_DEV.0245.01 = "NVIDIA Quadro NVS 210S / NVIDIA GeForce 6150LE"
NVIDIA_DEV.029C.01 = "NVIDIA Quadro FX 5500"
NVIDIA_DEV.029D.01 = "NVIDIA Quadro FX 3500"
NVIDIA_DEV.029E.01 = "NVIDIA Quadro FX 1500"
NVIDIA_DEV.029F.01 = "NVIDIA Quadro FX 4500 X2"
NVIDIA_DEV.039C.01 = "NVIDIA Quadro FX 560M"
NVIDIA_DEV.039E.01 = "NVIDIA Quadro FX 560"
NVIDIA_DEV.040A.01 = "NVIDIA Quadro FX 370"
NVIDIA_DEV.040D.01 = "NVIDIA Quadro FX 1600M"
NVIDIA_DEV.040E.01 = "NVIDIA Quadro FX 570"
NVIDIA_DEV.040F.01 = "NVIDIA Quadro FX 1700"
NVIDIA_DEV.042F.01 = "NVIDIA Quadro NVS 290"
NVIDIA_DEV.05E7.01 = "NVIDIA Tesla C1060"
NVIDIA_DEV.05E7.02 = "NVIDIA Tesla T10 Processor"
NVIDIA_DEV.05E7.03 = "NVIDIA Tesla C1060 "
NVIDIA_DEV.05E7.04 = "NVIDIA Tesla T10 Processor "
NVIDIA_DEV.05E7.05 = "NVIDIA Tesla M1060"
NVIDIA_DEV.05E7.06 = "NVIDIA Tesla M1060 "
NVIDIA_DEV.05E7.07 = "NVIDIA Tesla M1060  "
NVIDIA_DEV.05F9.01 = "NVIDIA Quadro CX"
NVIDIA_DEV.05FD.01 = "NVIDIA Quadro FX 5800"
NVIDIA_DEV.05FE.01 = "NVIDIA Quadro FX 4800"
NVIDIA_DEV.05FF.01 = "NVIDIA Quadro FX 3800"
NVIDIA_DEV.0619.01 = "NVIDIA Quadro FX 4700 X2"
NVIDIA_DEV.061A.01 = "NVIDIA Quadro FX 3700"
NVIDIA_DEV.061B.01 = "NVIDIA Quadro VX 200"
NVIDIA_DEV.061C.01 = "NVIDIA Quadro FX 3600M"
NVIDIA_DEV.061D.01 = "NVIDIA Quadro FX 2800M"
NVIDIA_DEV.0638.01 = "NVIDIA Quadro FX 1800"
NVIDIA_DEV.0658.01 = "NVIDIA Quadro FX 380"
NVIDIA_DEV.0659.01 = "NVIDIA Quadro FX 580"
NVIDIA_DEV.065C.01 = "NVIDIA Quadro FX 770M"
NVIDIA_DEV.06D1.01 = "NVIDIA Tesla C2050"
NVIDIA_DEV.06D1.02 = "NVIDIA Tesla C2050 "
NVIDIA_DEV.06D1.03 = "NVIDIA Tesla C2070"
NVIDIA_DEV.06D8.01 = "NVIDIA Quadro 6000"
NVIDIA_DEV.06D9.01 = "NVIDIA Quadro 5000"
NVIDIA_DEV.06DD.01 = "NVIDIA Quadro 4000"
NVIDIA_DEV.06F8.01 = "NVIDIA Quadro NVS 420"
NVIDIA_DEV.06F9.01 = "NVIDIA Quadro FX 370 LP"
NVIDIA_DEV.06FA.01 = "NVIDIA Quadro NVS 450"
NVIDIA_DEV.06FB.01 = "NVIDIA Quadro FX 370M"
NVIDIA_DEV.06FD.01 = "NVIDIA Quadro NVS 295"
NVIDIA_DEV.06FF.01 = "NVIDIA HICx16 + Graphics"
NVIDIA_DEV.06FF.02 = "NVIDIA HICx8 + Graphics"
NVIDIA_DEV.087A.01 = "NVIDIA GeForce 9400"
NVIDIA_DEV.0A3C.01 = "NVIDIA Quadro FX 880M"
NVIDIA_DEV.0A78.01 = "NVIDIA Quadro FX 380 LP"
NVIDIA_DEV.0DD8.01 = "NVIDIA Quadro 2000"
NVIDIA_DEV.0DF8.01 = "NVIDIA Quadro 600"

One of our areas of focus for the DirectX SDK (June 2010) release was to make sure our samples supported Direct3D 11 hardware from NVIDIA, as we had previously only had AMD/ATI Direct3D 11 hardware to work with. There were a few issues that arose right around the release with the SubD11 and PNTriangles11 samples. At the time we thought it was related to the driver, but in fact it turned out to be due to tessellation shader related optimization issues in the June 2010 version of the HLSL complier.

Full story here

3D-Tech News Around The Web / OpenCL Tutorial – Memory Spaces
« on: October 28, 2010, 03:51:37 PM »
In his second tutorial, GPGPU expert Rob Farber discusses OpenCL™ memory spaces and the OpenCL memory hierarchy, and how to start thinking in terms of work items and work groups. This tutorial also provides a general example to facilitate experimentation with a variety of OpenCL kernels.

3D-Tech News Around The Web / Rapunzel, let your 100,000 CGI hairs down
« on: October 27, 2010, 05:48:02 PM »
Rapunzel is getting a makeover—including a new hairstyle—thanks to some 21st-century technology.
Walt Disney Co. is set to release the movie "Tangled," a 3-D, computer-animated adaptation of the classic tale.
"We could have drawn the hair in 2-D but it would have been just a yellowish shape on screen," Mr. Howard says. "With CGI, we can render 100,000 strands of hair—it's almost a character in itself."

Full "furry tale" at WSJ

When you're on a tight deadline—working hard to make your application as fun as possible—you don't want to worry about the drudgery of standard operations. Three Intel® Graphics Media Accelerator (Intel® GMA) chipsets free you up to do what you do best—make great games.

•         Whether you are developing a new game, or have an existing game that you want to port to the netbook platform, optimizing the CPU and GPU on Intel® Atom™ processor-based netbooks will help increase your success. Here are some easy optimizations and quick performance gains you can achieve when developing games for this fast-growing market.

•         Want to quickly pinpoint that nasty glitch bottlenecking your game's performance? The Intel® GPA tools enable you to visualize, isolate, and stamp out graphics performance issues. Check out these tips and tricks so you can quickly bring up Microsoft* DirectX-based games and other graphics applications across multiple platforms.
Whether you are tweaking performance bottlenecks or optimizing for new platforms, we hope these articles make your life and job easier

3D-Tech News Around The Web / [Demo Scene] Parties round-up October
« on: October 26, 2010, 05:41:58 PM »
Misc. parties this month:

As usual some 4K demos are most impressing.

Graphics processing units (GPUs) were originally designed to make the massive calculations required for rendering 3D images to a display. Because of the nature of processing and creating images today, GPUs must have a large number of cores that work in parallel to render models in photo-realistic detail. The growth of the gaming market, both for PCs and for gaming consoles, has driven a rapid pace of technological improvement, while the commodity nature of the gaming market has helped reduce the price of GPUs.

The ability to deliver dramatic increases in compute performance at a reduced cost has positioned GPU computing at the forefront of the next wave of high-performance computing (HPC) architecture adoption. Now, Dell is helping make GPU processing power even more accessible through the Dell™ PowerEdge™ C410x PCI Express (PCIe) expansion chassis, which enables organizations to connect servers through the appropriate host interface card to up to 16 external GPU cards. On the measure of peak single-precision floating point performance, a PowerEdge C410x with 16 NVIDIA® Tesla™ M2050 GPU modules can deliver up to 16.5 TFLOPS of computing throughput.

In HPC environments, GPU computing offers one of today’s most powerful computational technologies on a price/performance basis. To help organizations extend their use of GPU computing, Dell offers IT consulting services, rack integration, on-site deployment, and support services for organizations deploying and using GPU-based Dell systems. Taking advantage of these services and systems like the PowerEdge C410x expansion chassis can help organizations dramatically increase performance while maximizing efficiency.

Also included: How the National Center for Supercomputing Applications (NCSA) is taking advantage of the cost economies and performance of GPUs to drive a 47 TFLOPS peak computing cluster.

Download PDF


Some spec sheets from AMD:
Dell PowerEdge R415
Dell PowerEdge R515

3D-Tech News Around The Web / CPU-Z 1.56 is out
« on: October 26, 2010, 04:37:13 PM »
News are :

- Intel Core i3/i5/i7 "2000" serie processors (Sandy Bridge) with new logos.
- Intel Core i7 970 "Gulftown".
- Intel P67 chipset.
- AMD Phenom II, Athlon II and Sempron II "Caspian" processors.

click here for full release notes and download link

3D-Tech News Around The Web / Mobile GPU Comparison Guide Rev. 10.0
« on: October 25, 2010, 09:38:13 PM »
The Mobile GPU Comparison Guide

Revision 10.0
Added the NVIDIA Quadro FX 3800M GPU.
Added the NVIDIA Quadro NVS 5100M, 3100M and 2100M GPUs.
Updated the specifications of the NVIDIA Quadro FX 3700M and 3600M GPUs.
Updated the specifications of the NVIDIA GeForce G210M GPU.
Updated the specifications of the NVIDIA Quadro FX 2700M GPU.
Updated the specifications of the NVIDIA GeForce 9700M GTS GPU.
Updated the specifications of the NVIDIA Quadro FX 1700M GPU.
Updated the specifications of the NVIDIA Quadro FX 770M GPU.
Updated the specifications of the NVIDIA Quadro NVS 510M GPU.
Updated the specifications of the NVIDIA Quadro NVS 320M and 300M GPUs.
Updated the specifications of the NVIDIA Quadro NVS 160M, 150M, 140M, 135M, 130M, 120M and 110M GPUs.
Removed the NVIDIA GeForce 8800M GS GPU.
Total GPUs : 240

The Desktop Graphics Card Comparison Guide Revision 20.6
Updated the specifications of the Intel GMA HD integrated graphics.
Updated the specifications of the Intel GMA X3500 and X3000 integrated graphics.
Updated the specifications of the Intel GMA 3100 integrated graphics.
Total cards : 428

Here's is another try at adapting MLAA on the GPU.

AMD Catalyst 10.10b Hotfix Features:

    * Support for AMD Radeon HD 6870 and AMD Radeon HD 6850 series graphics product installed in a Windows XP based system

This whitepaper provides an overview of the capabilities and display technologies integrated into the AMD Radeon™ HD 6800 series GPUs display engine.

The AMD  Radeon HD 6800 series GPUs marks the introduction of several key display technologies:

            * First GPU to support DisplayPort 1.2 Multi-Streaming
            * First GPU to support both DisplayPort and HDMI 1.4a for stereoscopic 3D
            * First GPU to support enhanced color gamut remapping for wide color gamut displays

3D-Tech News Around The Web / NVIDIA Quadro Dual Copy Engines
« on: October 24, 2010, 07:51:27 AM »
The copy engine featured in Quadro solutions provides real GPU-asynchronous texture
downloads. Texture data can be downloaded or uploaded in parallel with 3D rendering.
... supported Quadro solutions1  add an additional DMA engine
making it now possible to overlap download, processing, and readback. To take
advantage of this, one thread (channel) is used for rendering, one is used for download
and the third is used for upload, and all transfers are done via PBOs. When partitioned
this way, the render thread will run on the graphics engine and the transfer threads on
the copy engines in parallel and completely asynchronous. These are fully functional GL
contexts so that non-DMA commands can be issued in the transfer threads but will time
slice with the rendering thread. Copy engines can also handle format conversions and
swizzling for same data types without CPU intervention, in contrast to previous
hardware constraints where the input data formats had to be GPU native.

 download whitepaper

Supercomputing 2010 Technical Program Schedule is now available.

According to that Benedict Gaster from AMD and Tim Mattson from Intel join the same presentations.
So probably Intel will show off their previously announced OpenCL SDK?
I wonder if the compatibility will be as good as ATI vs. NVIDIA  ;D

...October 22, 2010, the sale of Windows XP comes to an end...
Video cards have changed dramatically, and Windows XP is barely aware of GPU except perhaps as a limited utility raster blit-engine or some weird exclusive bolt-on mode that Direct3D created for specific applications.
Unless the machine you are running is still a desktop machine from circa 2001, Windows XP is probably not a great fit for a modern PC…

Full story here

3D-Tech News Around The Web / Microsoft Windows Command Reference
« on: October 23, 2010, 03:58:54 PM »
This download describes the use of the command-line tools used to perform various tasks related to Windows Vista, Windows 7, Windows Server 2003, Windows Server 2008, and Windows Server 2008 R2.

This guide comes in handy as you rarely use the CLI nowadays and forget about the commands and its parameters.

AMD Core Math Library for Graphic Processors (ACML-GPU) provides an ATI Stream-accelerated version of ACML. ACML-GPU accelerates certain routines in ACML, such as SGEMM and DGEMM, by off-loading the computation to the compatible GPUs in the system. The library dynamically decides, based on the parameters passed to the routines, whether to run the computation on the CPU or GPU, depending on which processor will yield the best performance.

ACML-GPU automatically scales its computation across multiple GPUs, if available and can take advantage of the double precision floating point hardware in the GPU on products that contain hardware DPFP support.

New for v1.1:

    * Support for the latest AMD graphics cards
    * GPU acceleration of complex matrix multiply BLAS routines
    * libCALBLAS sample source

Pages: 1 ... 104 105 [106] 107 108 ... 160