Running Folding@Home on ATI Graphics CardsMon, Oct 09, 2006 - 12:00 AM
F@H: Doing Something Good For Everyone
When growing up I knew my Grandmother had Lymphoma, but for as long as I could remember her cancer was in remission thanks to a successful series chemotherapy treatments when she was first diagnosed. During the next decade my Grandmother was often praised by those around her as being a cancer survivor. The test of time proved that the word ‘survivor’ might have been superficial as Lymphoma would inevitably take her life. No one can hide from cancer. It doesn’t matter what color your skin is, how good your health is or how much money you have – Cancer doesn’t care. While I’d never want anyone to see one of their family members deteriorate before their eyes it’s more than likely that you have come in contact with someone that has had a battle with Cancer. Let’s be realistic here! Cancer is something we will all likely have to deal with in life whether we would like to admit it or not.
When my grandmother passed away deep down inside I wanted to do something. There are well documented stages of emotions based around dealing with grief, but the stage of anger is the one that had the greatest impact on me. How do you take anger out on Cancer? Cancer is the hidden killer. How do you fight an enemy when you don’t know exactly what causes it and how to destroy it without harming the person that habors these Cancer cells? The answer to that might not seem obvious to some people, but the answer is money and scientific research. Luckily for us we all live in a world where countries would rather spend more money blowing each other up rather than spending money on trying to figure one of the biggest killers known to man - Cancer.
Since the scientific community doesn’t have the finical backing required to fight they must find other means to get the work done and not spend mass amounts of money, which is nearly impossible. Stanford University decided to use distributed computing as a means of getting the research done and the project has been a success since day one. The Stanford Folding project has been in effect since September 2000 and uses the CPU in ones home computer to get the work load done. It was obvious that my way to ‘fight back’ was to fold and what a better way to do that by getting a team here on Legit Reviews to join forces for a great cause. For over two years Legit Folding Team # 38296 has been busy folding over 20,650 work units on thier CPU’s to help research against a number of medical issues.
That all changed as two weeks ago in San Francisco, Stanford University released a beta version of its Folding@Home client that runs on ATI’s Radeon X1900 and X1950 series graphics cards (Radeon X1900 GT, Radeon X1900 XT, Radeon X1900 XTX, Radeon X1900 CrossFire Edition, Radeon X1950 XTX). The GPU client is available for download in either console, graphical or a graphical with “fancy graphics” version for Windows XP. By being able to use high performance Graphics Processing Units (GPUs) from ATI they have been able to achieve performance previously only possible on supercomputers. With this new technology (as well as the new Cell processor in Sony?s PlayStation 3), the Folding project will soon be able to attain performance on the 100 gigaflop scale per computer!
The goal of Stanford is to apply this new technology to dramatically advance the capabilities of the project, applying their simulations to further study of protein folding and related diseases, including Alzheimer?s Disease, Huntington’s Disease, and certain forms of cancer. With these computational advances, coupled with new simulation methodologies to harness the new techniques, Stanford believes they will be able to address questions previously considered impossible to tackle computationally, and make even greater impacts on our knowledge of folding and folding related diseases. It seems that Folding on CPU’s will remain, but eventually Folding on GPU’s will be a faster means of completing work units.
Let’s take a look at how to get folding on the GPU going.