Weihang Zhu's Blog

Chrome Web Store

Weihang Zhu — Wed, 12 Jan 2011 18:59:27 +0000

https://chrome.google.com/webstore

Marketing Web 2.0 applications online?

Airplay SDK

Weihang Zhu — Wed, 12 Jan 2011 18:57:53 +0000

Airplay SDK for cross-platform smartphone development.

http://airplaysdk.com/

Google Body

Weihang Zhu — Wed, 12 Jan 2011 18:48:42 +0000

Google Body for healthcare (only in Chrome now?):

http://bodybrowser.googlelabs.com/body.html#

http://www.youtube.com/watch?v=uQzAUaZ26co#t=12m12s

Human models are from Zygote.

http://www.3dscience.com/client_highlights/Google.php?utm_content=2008-05-21%2002%3A01%3A07&utm_source=VerticalResponse&utm_medium=Email&utm_term=Learn&utm_campaign=3DScience%2Ecom%20January%20Newslettercontent

GPU on smartphone and mobile devices?

Weihang Zhu — Wed, 12 Jan 2011 18:40:00 +0000

http://spectrum.ieee.org/tech-talk/semiconductors/processors/with-denver-project-nvidia-and-arm-join-cpugpu-integration-race

With Denver Project NVIDIA and ARM Join CPU-GPU Integration Race

cross platform/browser plugin development

Weihang Zhu — Sun, 28 Nov 2010 15:02:15 +0000

FireBreath works well.

http://www.firebreath.org

I tried nixysa from google. It seems to work but lacks documentation. I could not get it to work.

WebGL tools

Weihang Zhu — Mon, 01 Nov 2010 01:43:56 +0000

jQuery, Dojo, EXT
WebGL Libraries: GLGE, SpiderGL, X3DOM, CubicVR, CopperLicht, EnergizeGL, SceneJS
WebGLot = WebGL + plot

Shader toys

APE: AJAX Push Engine

Beat Heart Surgery enabled by GPU computing

Weihang Zhu — Sun, 31 Oct 2010 16:59:37 +0000

It is quite amazing to see this beat heart surgery enabled by GPU computing simulation. Also haptics …

http://www.brightsideofnews.com/news/2010/9/23/gpu-enables-beating-heart-surgery.aspx

CFD on GPU computing

Weihang Zhu — Tue, 22 Jun 2010 16:12:29 +0000

I saw this news. It makes me wonder why the speedup is only 2x compared to quad-core CPU.

http://www.marketwatch.com/story/nvidia-tesla-gpus-enable-shorter-design-cycles-improved-product-quality-using-acusolve-2010-05-27?reflink=MW_news_stmp

NVIDIA Tesla GPUs Enable Shorter Design Cycles, Improved Product Quality Using AcuSolve

Latest Release of the AcuSolve CFD Software Doubles Performance of Flow Simulations Using NVIDIA Tesla C2050 GPUs

SANTA CLARA, CA, May 27, 2010 (MARKETWIRE via COMTEX) — NVIDIA announced today that ACUSIM Software, a leading provider of computational fluid dynamics (CFD) solutions widely used by engineers and scientists involved in product design, has integrated support for NVIDIA(R) Tesla(TM) 20-series GPUs into the company’s latest AcuSolve 1.8 release.

Performance tests of the general-purpose finite-element-based CFD flow solver have demonstrated up to a 2x boost in performance with the Tesla C2050 GPU processor, compared with the latest quad-core CPU running the same simulation.

nVidia GTX 480 is released

Weihang Zhu — Wed, 31 Mar 2010 13:22:34 +0000

This has been long waited for: a consumer card that supports Fermi. CUDA 3.0 is released too.

http://money.cnn.com/news/newsfeeds/articles/marketwire/0601654.htm

OpenGL 4.0

Weihang Zhu — Sun, 14 Mar 2010 13:15:49 +0000

http://www.crunchgear.com/2010/03/11/opengl-4-0-comes-out-to-play/

…

OpenGL 4.0 was announced today at GDC, and clearly it has DirectX in its sights.

March 11, 2010 – San Francisco, GDC 2010 – The Khronos™ Group today announced the release of the OpenGL® 4.0 specification; a significant update to the most widely adopted 2D and 3D graphics API (application programming interface) that is used on all major desktop operating systems. OpenGL 4.0 brings the very latest in cross-platform graphics acceleration and functionality to personal computers and workstations and the OpenGL standard serves as the basis for OpenGL® ES, the graphics standard on virtually every shipping smart phone.

The big news for graphics is handhelds right now. …

The OpenCL thing is great as well. If you remember, OpenCL is a set of tools for offloading certain tasks from the CPU onto the GPU, when those tasks are better served by parallel processors. Loading web pages may work fine on your Snapdragon, but decoding video will have it at 100% and drain your battery — better to send it over to the GPU. The GPU computing thing hasn’t quite taken off yet, but it’s pretty much inevitable that it’ll start being implemented on a low level, since it can improve the user experience so dramatically.