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 ... 112 113 [114] 115 116 ... 160
2261
3D-Tech News Around The Web / ARM Mali GPU Device Driver open source
« on: August 12, 2010, 07:19:09 PM »
Quote
Recently we released a major update to the Linux drivers for the Mali-200 and Mali-400 MP  GPUs. Like many software projects we time-box our driver development, with two major releases each year. This release (r2p0) contains a bunch of exciting new features including Android  support, full SMP support, performance optimizations and some important EGL extensions. We'll talk about some of these in future blog posts, but for today I wanted to tell you about the other big change we made for r2p0: we've started to release parts of the driver stack under an open source license.

Full story at ARM

Check out their OpenGL ES 2.0 samples from 19th of October 2010  - "Lets take a look into the future, Conan"  ;D

Also ARM predicts: "OpenCL will be on Every Smartphone in 2014"

2262
Quote
Unlike existing browser-based 3D rendering solutions, WebVision provides a full-fledged game engine with high-end graphics power, extensive scalability, a comprehensive toolset, multiplatform support, all the deep integrations offered by Vision Engine 8, and interfaces that allow both development in native C++ code and the scripting language LUA.

inclusive SpeedTree and PhysX  8)


2263
3D-Tech News Around The Web / ATI Stream 2.2 samples tested with Fermi
« on: August 12, 2010, 05:22:28 PM »
I tested the samples from ATI Stream SDK 2.2 (OpenCL 1.1) with an NVIDIA GTX 465 and Forceware 258.19
You know a sample is incompatible if GPU usage = 0%  :P

Constant Bandwidth - 99% GPU usage
AccessType   : single(static index)
VectorElements   : 4
Bandwidth   : 1331.02 GB/s

AccessType   : single(dynamic index)
VectorElements   : 4
Bandwidth   : 847.839 GB/s

AccessType   : linear
VectorElements   : 4
Bandwidth   : 26.2818 GB/s

AccessType   : random
VectorElements   : 4
Bandwidth   : 19.087 GB/s

LDS bandwidth - 53% GPU usage
AccessType   : single
VectorElements   : 1
Bandwidth   : 820.885 GB/s

AccessType   : linear
VectorElements   : 1
Bandwidth   : 824.71 GB/s

PCIE bandwidth
Host to device : 1.8002 GB/s - 85% GPU usage
Device to host : 2.34058 GB/s - 70% GPU usage

Memory bandwidth 53-78% GPU usage
-----------------------------------------
Copy 1D FastPath   : 72.8925 GB/s
-----------------------------------------
Copy 1D CompletePath   : 72.3641 GB/s
-----------------------------------------
Copy 2D 32-bit (64x2)   : 67.5548 GB/s
Copy 2D 128-bit (64x2)   : 83.6624 GB/s
-----------------------------------------
Copy 2D 32-bit (64x4)   : 70.8011 GB/s
Copy 2D 128-bit (64x4)   : 81.9781 GB/s
-----------------------------------------
Copy 2D 32-bit (8x8)   : 37.8293 GB/s
Copy 2D 128-bit (8x8)   : 81.5374 GB/s
-----------------------------------------
Copy 2D 32-bit (256x1)   : 72.0669 GB/s
Copy 2D 128-bit (256x1)   : 82.2084 GB/s
-----------------------------------------
Copy 2D 32-bit (32x2)   : 46.1947 GB/s
Copy 2D 128-bit (32x2)   : 82.5491 GB/s
-----------------------------------------
Copy 2D 32-bit (64x1)   : 47.8214 GB/s
Copy 2D 128-bit (64x1)   : 81.668 GB/s
-----------------------------------------
Copy 2D 32-bit (16x16)   : 67.2074 GB/s
Copy 2D 128-bit (16x16)   : 81.3934 GB/s
-----------------------------------------
Copy 2D 32-bit (16x4)   : 42.3963 GB/s
Copy 2D 128-bit (16x4)   : 82.2428 GB/s
-----------------------------------------
Copy 2D 32-bit (1x64)   : 7.94972 GB/s
Copy 2D 128-bit (1x64)   : 35.0866 GB/s
-----------------------------------------
Copy 1D 128-bit    : 235.827 GB/s
-----------------------------------------
NoCoal Copy 1D 32-bit    : 99.3111 GB/s
-----------------------------------------
Split Copy 1D 32-bit    : 32.9356 GB/s

Quote from: CLINFO
Number of platforms:             1
  Platform Profile:             FULL_PROFILE
  Platform Version:             OpenCL 1.1 CUDA 3.2.1
  Platform Name:                NVIDIA CUDA
  Platform Vendor:             NVIDIA Corporation
  Platform Extensions:          cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll


  Platform Name:                NVIDIA CUDA
Number of devices:             1
  Device Type:                CL_DEVICE_TYPE_GPU
  Device ID:                4318
  Max compute units:             11
  Max work items dimensions:          3
    Max work items[0]:             1024
    Max work items[1]:             1024
    Max work items[2]:             64
  Max work group size:             1024
  Preferred vector width char:          1
  Preferred vector width short:          1
  Preferred vector width int:          1
  Preferred vector width long:          1
  Preferred vector width float:          1
  Preferred vector width double:       1
  Max clock frequency:             810Mhz
  Address bits:                32
  Max memory allocation:          260423680
  Image support:             Yes
  Max number of images read arguments:    128
  Max number of images write arguments:    8
  Max image 2D width:          8192
  Max image 2D height:          8192
  Max image 3D width:          2048
  Max image 3D height:    2048
  Max image 3D depth:          2048
  Max samplers within kernel:       16
  Max size of kernel argument:          4352
  Alignment (bits) of base address:       4096
  Minimum alignment (bytes) for any datatype:    128
  Single precision floating point capability
    Denorms:                Yes
    Quiet NaNs:                Yes
    Round to nearest even:          Yes
    Round to zero:             Yes
    Round to +ve and infinity:          Yes
    IEEE754-2008 fused multiply-add:       Yes
  Cache type:                Read/Write
  Cache line size:             128
  Cache size:                180224
  Global memory size:             1041694720
  Constant buffer size:             65536
  Max number of constant args:          9
  Local memory type:             Scratchpad
  Local memory size:             49152
  Profiling timer resolution:          1000
  Device endianess:             Little
  Available:                Yes
  Compiler available:             Yes
  Execution capabilities:            
    Execute OpenCL kernels:          Yes
    Execute native function:          No
  Queue properties:            
    Out-of-Order:             Yes
    Profiling :                Yes
  Platform ID:                0000000002C78F20
  Name:                   GeForce GTX 465
  Vendor:                NVIDIA Corporation
  Driver version:             258.19
  Profile:                FULL_PROFILE
  Version:                OpenCL 1.1 CUDA
  Extensions:                cl_khr_byte_addressable_store cl_khr_icd cl_khr_gl_sharing cl_nv_d3d9_sharing cl_nv_d3d10_sharing cl_khr_d3d10_sharing cl_nv_d3d11_sharing cl_nv_compiler_options cl_nv_device_attribute_query cl_nv_pragma_unroll  cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_fp64


Error : atomics mismatch!
Error : Bytes mismatch!
Error : d3d10Sharing mismatch!
Error : glSharing mismatch!
Error : images mismatch!
Error : printf mismatch!
Error : deviceAttributeQuery mismatch!
Failed!


   

2264
Want a faster way to extract full parallel performance on multicore processors?

Get three.

Power through common parallel programming headaches and maximize scalability and performance with step-by-step Evaluation Guides and a test-drive of the industry-leading Intel® Parallel Studio.

•         Locate a Hot Spot and Optimize It
—In this download, see how finding a single hot spot and recompiling one file yields a huge leap in performance.
 
•         Eliminate Memory Errors and Improve Program Stability
—Stamp out intermittent bugs with this quick guide to minimizing code defects that impact app reliability.
 
•         Optimize an Existing Program by Introducing Parallelism
—This six-step process makes a serial "for" loop parallel. You can tweak performance up to 1.99x in just 15 minutes and a few lines of code.
 
That's fast.

2265
3D-Tech News Around The Web / MSI Afterburner 2.0.0 Beta 6
« on: August 12, 2010, 03:57:47 PM »
Download Afterburner 2.0.0 Beta 6 from Guru3D
Archive also contains Kombustor 1.1.3

Quote
Revision history:

--------------------------------------------------------------------------------

Version 2.0.0

- Voltage control layer has been seriously revamped to give additional freedom
  to extreme overclockers with new custom design MSI graphics cards. Now MSI
  Afterburner is able to control up to 3 voltages on custom design MSI Fermi
  and other future custom design MSI graphics cards. New adjustable voltages
  include memory voltage and special multi-purpose auxiliary voltage feeding
  either memory bus (also known as VDDCI on AMD graphics cards) or PCIE bus
  and crystal (PEXVDD on NVIDIA graphics cards)
- Added core, memory and auxiliary PEXVDD voltage control for custom design MSI
  N480GTX Lighning series graphics cards with uP6225+uP6262 voltage regulators
- Added core, memory and auxiliary PEXVDD voltage control for custom design MSI
  N470GTX Twin Frozr II OV3 series graphics cards with uP6218+uP6262 voltage
  regulators
- Added core, memory and auxiliary PEXVDD voltage control for custom design MSI
  MSI N460GTX Hawk series graphics cards with uP6262 voltage regulators
- Added core voltage control for custom design MSI N460GTX Cyclone series
  graphics cards
- Added auxiliary VDDCI voltage control for reference design AMD RADEON HD 5870
  series graphics cards
- Added fan tachometer monitoring for NVIDIA graphics cards. Please take a note
  that not all graphics cards are tachometer reading capable. So depending on
  graphics card and cooling system fan tachometer monitoring can be unavailable
- Optimized NVIDIA driver-level clock frequency monitoring codepath
- Minimum clock limits for all graphics cards have been reduced from 75% to 50%
- Dynamic overclocking, voltage and fan speed limits. MSI Afterburner no longer
  uses static slider limits calibration and adjusts the limits dynamically when
  some external factors affect it (e.g. Overdrive clock limits on AMD cards or
  VGA BIOS fan speed and voltage limits on NVIDIA GeForce GTX 400 cards)
- New temperature hysteresis settings for software automatic fan control mode
  gives you additional way to improve cooling system thermal and acoustic
  parameters
- Now fan speed limits (i.e. minimum and maximum fan speeds accepted by VGA
  BIOS and display driver) are displayed in custom fan speed curve editor
  window
- Built-in skin sizes have been reduced due to optimized internal skin panels
  representation and optimized compiled bitmap cache
- Optional skin compression ability in the built-in skin compiler. Skin format
  reference documentation has been updated to document new compression options
- Minor built-in skins appearance tweaks
- Now MSI Afterburner uses previously undocumented power user oriented startup
  mode via the task scheduler under Windows Vista / Windows 7. MSI Afterburner
  launch no longer requires UAC confirmation at Windows startup. Please take a
  note that Microsoft Visual C++ 2008 runtime libraries must be installed to
  get new startup mode working
- Now MSI Afterburner automatically fixes startup link if <Start with Windows>
  or <Apply overclocking at system startup> is enabled but the registry or task
  scheduler startup entry is missing
- Startup profile is now displayed in <Apply at Windows startup> option floating
  tooltip instead of the main window
- Now power users can enable optional DirectInput based hotkeys handler via the
  configuration file. DirectInput based hotkeys processing can seriously reduce
  hotkey response time in the applications heavily loading CPU (mostly 3D games)
  and leaving not enough time for processing standard keyboard input message
  queues. Please take a note that enabling such sophisticated hotkeys handling
  mode can cause some system security applications (e.g. pro-active application
  behavior analysis module of KIS) to warn you about possible keylogging threat
- Improved skin engine, now skinned controls support horizontal and/or vertical
  centering. Skin format reference guide has been updated to document these new
  alignment modes
- MSI On-Screen Display server has been upgraded to version 3.7.2. New version
  gives you the following improvements:
  - Now screen capture events are identified visually by text message flashing
    in On-Screen Display during 0.25s
  - Built-in skin sizes have been reduced due to optimized compiled bitmap cache
  - Added On-Screen Display profile for Startcraft II : Wings of Liberty
- Added configuration file switch allowing sending MSI Afterburner to system
  tray instead of closing on <Close> button click
- <Start the task only when computer is running on AC power> option is no longer
  set in the startup task settings to allow automatically starting application
  via the task scheduler on laptops or on some UPS models
- Added Korean localization

2266
Quote
This quick start guide is designed to assist a new user in setting up and starting PerfHUD ES profiling. PerfHUD ES enables experimentation and in-depth analysis of OpenGL ES applications. Through data provided by the OpenGL ES driver and the NVIDIA GPU, developers are given access to GPU performance and bottleneck information in order to target optimizations where they are most needed. The debugging capabilities give full access to the state of the OpenGL ES pipeline, related textures and shaders, and all rendering states to help find the causes for improper setup and rendering anomalies.

For detailed information on the use of PerfHUD ES please refer to the PerfHUD ES User Guide that is installed with the host application.

Android PerfHUD ES quick start guide
Tegra Linux PerfHUD ES Quickstart Guide

2267
Quote
Phoenix Fluid Dynamics is the latest product released by Chaos Software. The software perfectly combines a grid based simulator with outstanding rendering capabilities. In addition to the usual uniform fluid behavior, Phoenix FD is capable of simulating a whole array of additional processes like pressure decay, thermal radiation cooling and mass-temperature dependence.

Check out their videos

2268
Quote
The new release of the Voreen 2.5 volume rendering engine contains several new features. It is now possible to create and edit animations, which can be exported as videos. Furthermore, Voreen now contains a GPU-based multi-volume renderer, allows aggregation of subnetworks and supports iterative processing by exploiting loops in the network graph, and contains support for OpenCL used for some processors.

[via]

2269
Created by Digimania, Muvizu is a free program that allows you to make animated videos without fancy equipment and years of training.

Download mirrored at GamersHell


2270
3D-Tech News Around The Web / Mafia II - PhysX APEX benchmark & demo
« on: August 10, 2010, 05:00:17 PM »
Download from Steam

Further infos at PhysXinfo


2271
Download 259.31 driver here, don't know what's different from 259.09  ???

If you can refuse the Mafia II demo that has been released today, download PhysX_10.05.12_9.10.0512_SystemSoftware.msi here.


2272
3D-Tech News Around The Web / Mobile GPU Comparison Guide Rev. 9.6
« on: August 09, 2010, 07:13:08 PM »
Quote
The Mobile GPU Comparison Guide

These days, there are so many mobile GPU models that it has become quite impossible to keep up with the different configurations. Therefore, we decided to compile this guide to provide an easy reference for those who are interested in comparing the specifications of the various mobile GPUs in the market as well as those already obsolescent or obsolete.

Currently covering 225 mobile GPUs, this comprehensive comparison will allow you to easily compare 17 different specifications for each and every GPU! We hope it will prove to be a useful reference. We will keep this guide updated regularly so do check back for the latest updates!

Revision 9.6 changelog:
Added the Pixel Fill Rate specifications for NVIDIA GPUs.
Added the NVIDIA GeForce GTX 480M GPU.
Updated the specifications of the NVIDIA GeForce 9800M GTX, 9800M GT, 9800M GTS and 9800M GS GPUs.
Updated the specifications of the NVIDIA GeForce 9700M GTS and 9700M GT GPUs.
Updated the specifications of the NVIDIA GeForce 9650M GT, 9650M GS, 9600M GT and 9600M GS GPUs.
Updated the specifications of the NVIDIA GeForce 9500M GS and 9500M G GPUs.
Renamed and updated the specifications of the NVIDIA GeForce 9400M G GPU.
Removed the NVIDIA GeForce 9700M GS GPU.

2273
3D-Tech News Around The Web / Computer Gamers Tackle Protein Folding
« on: August 09, 2010, 06:57:29 PM »
Quote
Biochemists and computer scientists at the University of Washington two years ago launched an ambitious project harnessing the brainpower of computer gamers to solve medical problems. The game, Foldit, turns one of the hardest problems in molecular biology into a game a bit reminiscent of Tetris. Thousands of people have now played a game that asks them to fold a protein rather than stack colored blocks or rescue a princess.

Full story at Dr. Dobb's

2274
Quote
The eventual creation of the APU was one of the key reasons for AMD’s acquisition of ATI in 2006. This acquisition brought to AMD industry-leading GPU and chipset technologies, and a software team to support them. As we sit on the precipice of launching the first generation of APUs, people are wondering if discrete GPUs will disappear altogether. Continue reading

2275
3D-Tech News Around The Web / The Misunderstanding - Presented by AMD
« on: August 09, 2010, 06:51:54 PM »
A humorous look at the risks of using anything but an ATI Radeon graphics card for gaming.
A short film presented by AMD.

2276
3D-Tech News Around The Web / More D3D v-table hacking
« on: August 09, 2010, 03:52:38 PM »
Quote
The trick in the previous post is as implied by its title and noted in the comments not going to give much return on the time invested in terms of performance, although it might give you a deeper understanding of the underlying mechanisms and costs involved in calling virtual functions. So it's more like an academic exercise, but nonetheless a cool thing to play around with.

Full article at Humus' site

2277
ATI Catalyst 10.7 Update Driver Feature:
The release of this software driver from AMD delivers support, when used in conjunction with the ATI Stream SDK v2.2, for OpenCL 1.1, an open and royalty-free programming standard for general-purpose computations on heterogeneous systems. For more information about OpenCL, please visit the OpenCL Zone.

2278
3D-Tech News Around The Web / [Demo Scene] Assembly Summer 2010
« on: August 08, 2010, 08:57:57 AM »
Assembly Summer 2010 demos have been uploaded at POUET.
"When I saw ASD's demo I seriously thought it was worth stopping the compo right there and just giving Navis every single damn prize in the building. Holy f.....g sh.."

Here are the demos i liked. Some of them take a long time to launch (because of building shader caches i assume).

Happiness is Around the Bend  8)

Atleeti - Physics and AI in 4K

Behind the Curtain - 64K GPU burner

Bubblin Under - makes a nice screensaver

Collideride - 4K GPU burner

Devolution - nice visuals except from the annoying flying text messages

Kaks Koiraa Savosta Redux - 1K GPU burner, system froze on pressing ESC :(


2279
3D-Tech News Around The Web / NVIDIA driver 198.19 HP
« on: August 07, 2010, 11:01:48 AM »
As you might have already noticed some programs don't run with driver branch 256+, e.g. NVIDIA's own "Raging Rapids Ride"
In this case roll back to 190+ drivers.
Modify NVHO.INF by yourself or wait until it appears at LaptopVideo2Go

Download here:
Vista/7 64 bit
Vista/7 32 bit


2280
Quote
A video demonstrating my GPU-based Smoothed Particle Hydrodynamics (SPH) implementation using CUDA to achieve extremely high performance.
The video shows the real-time performance of the implementation, more than 100 FPS (dt=0.002) is achieved on a NVIDIA Geforce GTX 470 (Fermi).
In this scene a water-like fluid is simulated using a SPH model described by Mueller in "Particle-Based Fluid Simulation for Interactive Applications".
Test scene 1, 128K particles and hue-based color gradient for particle velocity.
This work is part of my masters thesis in computer science at NTNU (Norwegian University of Science and Technology).

Announcement at Ogre3D forum
Videos at Youtube
The code and masters thesis - the binary files in the SVN trunk are incomplete  :(
NTNU HPC-lab website (including happy computer geeks in a lab coat unpacking a Tesla 2070 ;D )

[via]

Pages: 1 ... 112 113 [114] 115 116 ... 160