In this episode:
MAQUET Cardiovascular LLC has announced a new endoscopic visualization system named “Vasovision” that uses heads-up displays with goggles to place the endocopie view directly in the operator’s field of view.
Worn as eyewear by the healthcare provider retrieving the vein, the VASOVISION system offers crisp, clear views of the harvesting tunnel (the area surrounding the vein to be harvested) and can eliminate the strain associated with poorly placed monitors or field of view challenges commonly found in today’s operating rooms. The system is lightweight, comfortable and easy to use with fingertip adjustments for brightness and contrast. This new product is part of the leading VASOVIEW Endoscopic Vessel Harvesting (EVH) portfolio of products from MAQUET Cardiovascular, and will be sold separately as well as in combination with other VASOVIEW products.
Always great to see technology making it into the OR, and this system not only reduces wound complicates and allows minimally invasive procedures, it reduces strain on the surgeon from cumbersome displays. See it on Maquet’s Site.
Tim Sweeny, the founder of Epic Games and architect of the Unreal Engine series, looks into the future and says that by 2020, developers will be performing 100% software rendering on the CPU instead of using graphics hardware. In all fairness, he notes that the CPU and GPU of today are getting progressively closer, therefore what we might call a CPU or GPU today may not be the CPU/GPU of the future.
He makes the assumption that in 2012, a 4 Teraflop processor would execute 16,000 operations per pixel at 1920×1080 @ 60 Hz, all the while assuming that 50% of the CPU is reserved for gameplay and 50% for software rendering. This will require massive memory bandwidth. Four Trillion Floating Point Operations per Second (4 TFLOPS) of computing power demands4 Trillion Bytes Per Second of effective memory bandwidth. It is interesting that he uses the 4 TFLOPS system as a benchmark throughout his presentation, for one can buy a 4 TFLOPS system now using four of the latest GPUs from NVidia. Of course, neither the memory bandwidth, nor the programming model, and most certainly not the price, is anywhere near the realm of where it needs to be.
The 74 page presentation includes his opinion of using raytracing, the REYES Rendering Model, and using volumetric rendering in games.
Over at 3dm3, Vaibhav Shah has a great making-of for an image called “Say Cheese”, built with Maya, zbrush, mental-ray, and Photoshop.
I was so inspired with water tanks in childhood. Whenever I’ve visit fisheries and aquariums, I always found some attraction towards one tank which was dirtier and poor a fish try to accommodate in that. I’ve always spend time to observe water movement and its whirl effects. So I approached to make cool dirty flooded water bathroom, and fish tries to survive there in between of leaked water pipes. Even leaked pipe blows air into water which makes sweetie bubbles. The idea clicked me once and started scribbles for making ahead.
Not only does he show the process used to conceptualize and create the image, but some interesting charts showing how long the various stages took.
Microsoft’s new ZuneHD, available currently for pre-order, has some NVidia in it’s DNA with the new “Tegra” mobile platform chip.
“Tegra provides the multimedia muscle in Zune HD,” said Michael Rayfield, general manager of NVIDIA’s mobile business. “Users will love the device’s new design, amazing multimedia features and HD video out capability. Zune HD is a must-have for anyone looking for the best portable digital media player on the market.”
The Tegra chip reminds me of the Cell architecture, combining an HD Video processor, an audio processor, a graphics processor, and two ARM cores all onto a single chip to maximize power efficiency.
Matt Artz over at t
he wonderful GIS & Science blog has posted a video of Southern California earthquakes between 1932 and 2008.
Produced with ArcGIS Explorer, this animation shows all earthquakes of magnitude 3 or greater in the Southern California area between 1932 and 2008. Data from USGS.
One highly anticipated piece of software amongst the HPC and GPGPU groups has been “CULA”, a port of the popular LAPack linear algebra library to CUDA systems allowing a near-painless performance boost by simply installing a high-end NVidia card and recompiling your code. That day is getting closer, as EM Photonics has released a beta of their upcoming version of this, ‘CULATools’.
EM Photonics’ CULAtools is a product family comprised of CULA Basic, Premium, and Commercial. The CULA library is a GPU-accelerated implementation of the most popular LAPACK routines. LAPACK is a collection of commonly used functions in linear algebra, used by millions of developers in the scientific and engineering community. The problems they tackle can often be approximated by linear models and can, therefore, be solved using linear algebra routines. CULA exploits the massively parallel CUDA architecture of NVIDIA’s GPUs to accelerate many of the common LAPACK routines.
The full production released is expected to coincide with the NVidia GPU Technology Conference September 30th-October 2nd.
via New CULA Linear Algebra Library from EM Photonics Brings GPU Computing to Millions of Developers.
The winners of the COLLADA contest at SIGGRAPH2009 have been announced, and the first-place winner is an impressive solution from Charles Han of San Francisco, Navicad.
Dan Halabe and Chuck Han have released NaviCAD, the only way to view the Google 3D Warehouse on the iPhone. You can search through thousands of 3D models to view and navigate in 3D. NaviCAD has been approved for sale by Apple and is available for download immediately here.
Check out the website for some impressive visuals, and you can download the app for a mere $0.99.
Winners Circle – COLLADA Contest – Champion the 3D Web using COLLADA.
As “Billion” becomes the new “Million” while the government continue to dump money into the economy in hopes of staving off disaster, David McCandless, an information designer in London, had combined several financial figures from government and corporations around the world to try and give some scale to the figures.
This image arose out of a frustration with the reporting of billion dollar amounts in the media. That is, they’re reported as self-evident facts, when, in fact, they’re mind-boggling and near incomprehensible without context. But they can start to be understood visually and relatively, IMHO.
(This is one of the first images I created for my book. So a lot of the figures are from 2006/07. I’ve also visually cheated slightly here and there to make everything fit)
Jack Pien has taken AMD’s new OpenCL driver for x86 CPU’s and put it through it’s paces with an Intel processor, using a modified version of the standard SimpleConvolution kernel.
I used the SDK’s SimpleConvolution sample as a testbench to get some performance measurements compared to the very straight forward reference CPU implementation also included the SDK (compiled w/ just the -O3 gcc flag in release mode). The regular CPU implementation is also used to validate the OpenCL results.
The results show a nice boost to performance at the higher parallelization levels, but not as nice as you would expect. He’s got a nice writeup, and he’s looking for comments on where to go next. Head on over and help him out.
via Jack Pien» Blog Archive » A look at AMD’s x86 OpenCL perf on an Intel system.
VizWorld.com Got an idea for a story? Drop us a line at [email protected] or other means.
Comments