VizWorld.com

AMD has just released the newest version of ACML-GPU (AMD Core Math Library for GPU), which offers several popular math functions (like SGEMM and DGEMM) with GPU optimizations.  However, unlike other options these can automagically switch between the GPU and CPU execution modes depending on problem size and hardware available.

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.

Requires the STream SDK, but works in both PGI Fortran, Visual Studio C, and GCC in Windows and Linux.

via AMD Core Math Library for Graphic Processors (ACML-GPU) | AMD Developer Central.

Hardware, Science amd, gpgpu, library

The JavaScript InfoVis Toolkit has just rev’ed up to Version 2.0, with a whole slew of new features and visualizations.

With this version of the Toolkit the number of available visualizations has doubled. Some of the new visualizations are the AreaChart, BarChart and PieChart, which were described in more detail in this article. I’ve also added Sunburst and Force-Directed visualizations. I wrote about these visualizations before here and here. I also want to thank Pablo Flouret, who contributed most of the code for the Icicle visualization, also a new addition to the toolkit.

Be sure to go check out the new and improved Demos page to see the new visualization in action.

via The JavaScript InfoVis Toolkit 2.0 is out! at Overfloater. via FlowingData

Science javascript, library, software

An interesting project called the “Anti Grain Geometry Project”, or AGG Project, aims to develop a library for detailed development of graphics algorithms, without any of the constraints of libraries like OpenGL.

Anti-Grain Geometry is designed as a set of loosely coupled algorithms and class templates united with a common idea, so that all the components can be easily combined. Also, the template based design allows you to replace any part of the library without the necessity to modify a single byte in the existing code.

Also AGG is designed keeping in mind extensibility and flexibility. Basically I just wanted to create a toolkit that would allow me (and anyone else) to add new fancy algorithms very easily.

Aimed primarily at development of algorithms, the author concedes that working with it can be a bit difficult.  However, the results he shows in just basic antialiasing work show some great promise.

via Anti-Grain Geometry – Introduction.

Graphics, Science agg, library, sdk

While the individual components are nothing new, I hadn’t heard of NVidia calling the whole system “AXE”, for Application Acceleration Engines.

NVIDIA® application acceleration engines are highly optimized software modules enabling developers to take maximum advantage of the GPU with valuable, high performance capabilities that are license free to develop with and deploy.

NVIDIA is committed to ensuring engines continue advancing to take maximum advantage of the latest NVIDIA GPU innovations while also maintaining their leading-edge capabilities. Applications employing acceleration engines can rapidly exploit GPU advancements as consistent engine APIs evolve to leverage new methods. In doing so, NVIDIA is empowering developers to deliver the latest capabilities to their users in the shortest amount of time.

The complete suite consists of SceniX (scene graph), CompleX (GPU scaling system), PhysX (GPU accelerated Physics), and OptiX (ray-tracing), along with the usual CgFX and CUDA.

via NVIDIA® Application Acceleration Engines.

Science axe, complex, gpgpu, library, nvidia, optix, physx, scenix

Over at Insideria, Theresa Neil has gathered a list of 28 toolkits for visualization and charting that covers web languages, flash, silverlight, and more.

We’re currently working with a dozen different clients, all web application (re)designs. All of these clients have data rich applications and need equally rich data visualizations to help their end customers analyze data quickly and effectively.

What makes my job really interesting is that these clients are in different industries and are using different technologies. So we have pulled together a set of 28 tools for creating graphs, Gantt charts, diagrammers, calendars/schedulers, gauges, mapping, pivot tables, OLAP cubes, and sparklines, in Flash, Flex, Ajax or Silverlight.

via 28 Rich Data Visualization Tools – InsideRIA.

Science library, visualization, Website

Perlita Labs has an infographic online comparing the many topologies, visualizations, and data structures available in a wide variety of popular visualization tools.  It’s pretty comprehensive, including about 13 different libraries and almost 20 different features.

Comparing Data Visualization Tools.

Science infographic, library, perlita

CUDA and GPGPU developers can head on over to the NVidia Developer Zone and, if you’re a registered developer, download the beta of the new CUDA Toolkit 3.0 features early support for the upcoming Fermi architecture.  Also:

• CUDA Driver / Runtime Buffer Interoperability, which allows applications using the CUDA Driver API to also use libraries implemented using the CUDA C Runtime.
• A new, separate version of the CUDA C Runtime (CUDART) for debugging in emulation-mode.
• C++ Class Inheritance and Template Inheritance support for increased programmer productivity
• A new unified interoperability API for Direct3D and OpenGL, with support for:
  • OpenGL texture interop
  • Direct3D 11 interop support
• cuda-gdb hardware debugging support for applications that use the CUDA Driver API
• New CUDA Memory Checker reports misalignment and out of bounds errors, available as a debugging mode within cuda-gdb and also as a stand-alone utility.
• CUDA Toolkit libraries are now versioned, enabling applications to require a specific version, support multiple versions explicitly, etc.
• CUDA C/C++ kernels are now compiled to standard ELF format

CUDA Toolkit 3.0 beta released – NVIDIA Forums.

Hardware, Science api, cuda, library, nvidia

All of you young budding game developers have a new toolkit in your arsenal, the newly released Unreal Development Kit.  A new free version of the popular Unreal3 & Development tools, it brings the power of the Unreal Engine into hands of game developers worldwide.

“I’m excited about the possibilities the Unreal Development Kit opens to those who are looking to get into the game business but don’t otherwise have the means to acquire world-class technology and tools like ours,” said Mark Rein, vice president of Epic Games. “UDK is Unreal Engine 3, which has been used to create games in a wide range of genres, as well as military simulations, 3D architectural walkthroughs, animated movies and more. Users are only limited by their imaginations. Go ahead make something Unreal!”

The UDK comes with PhysX, LightMass, AnimTree, integration systems for SpeedTree, Bink, and FaceFX, and all of the usual Unreal Engine features like UnrealScript.  Licensing restricts it to noncommercial or educational use, so you’ll eventually need to purchase a commercial license if you plan to sell your game.

via News – Epic UDK.

Graphics, Science engine, library, sdk, unreal, videogame

GPU Ray-tracers, start your engines.. NVidia has just officially released the OptiX CUDA-based Real-Time Interactive Ray Tracer on their website, free.  Announced back at SIGGRAPH, there have been numerous demonstrations on various conferences and venues, but now you can try it for yourself.

Jeff Brown, NVIDIA’s GM for Professional Solutions, explains why OptiX is invaluable: “This opens the door to a new level of interactive realism. Ray tracing’s inherent parallelism makes it a perfect fit for GPU computing. The OptiX engine makes it easy for developers to exploit that power to create an exciting new class of applications. It enables critical design tasks — such as examining reflections, refractions and shadow – to be performed now in real-time.”

Of course, they claim it’s the “World’s First”, but I think the guys at Caustic, Intel‘s Larrabee , and Fryrender would argue that claim.

Go download it in the NVidia Developer’s Zone.

via nTersect Blog – NVIDIA.

Science cuda, gpgpu, library, nvidia, optix, raytracing, software

ITK, the popular image processing toolkit,  is about to turn 10 years old, having begun in November 1999.  To celebrate, Kitware is holding a few events to commemorate the occasion and help push ITK to the next level.  They’ve also got a bit of history about the project online.

After the release, the NLM awarded several Algorithms, Adapters, and Data Distribution (A2D2) contracts to test ITK’s architecture and API design. The NLM also extended the contract period for several of the original contractors. Today the NLM still supports ITK through maintenance contracts, but the bulk of the work is done by the growing, international ITK community. To help us celebrate the longevity of this toolkit, Kitware is hosting several community events.

There’s even prizes, like the largest number of builds in the CDash Dashboard wins a Kitware Polo Shirt!

via ITK – Segmentation & Registration Toolkit.

Science itk, kitware, library, software