Adequacy front page
Stories Diaries Polls Users
Google

Web Adequacy.org
Home About Topics Rejects Abortions
Poll
Quantum chemists should get their thrills...
...by buying bigger cars 0%
...by getting out more and meeting more hot chicks 10%
...by mountain climbing 0%
...torturing helpless lab animals 80%
...chemists are just fine getting their thrills from bragging about their CPUs 10%

Votes: 10

 CPU Cycle Theft

 Author:  Topic:  Posted:
Nov 15, 2001
 Comments:
There is a group of people in our society that is a drain on the world's resources. They consume CPU cycles in the biggest waste of computational power the world has ever seen. They are running one of the largest computing projects ever to see the light of day and to do it they are stealing CPU cycles from millions of innocent people around the world. I'm talking about computational chemists and their flagship CPU burner Folding@Home
Firstly I must point out that I was once one of them. I worked in the computational chemistry labs of one of the world's largest pharmaceutical companies.

Here's the important thing to know: to simulate chemical processes you need to do quantum mechanics. You may have heard of the 3 body problem. Finding the orbits of 3 particles bound in a gravitational or electrostatic field is called the 3-body problem. And it's hard. Well, basically it has no solution apart from throwing CPU power at it. Well that's nothing. It's the most trivial little problem compared to the quantum 3-body problem. It's orders of magnitude harder. But that's nothing. A protein has thousands of particles in it. And get this: each time you add one more particle to the quanum n-body problem it gets an order of magnitude harder. Solving to find how a protein folds isn't just an order of magnitude harder than figuring out how 3 planets orbit. It's many many orders of magnitude harder.

Well, what can we simulate if proteins are so hard? We can do a good job on single particles like a free electron. We can even do a stupendously good job of figuring out what an electron and a nucleus do and so we have great theoretical predictions for atomic energy levels. What about the simplest molecule: 2 nuclei and some shared electrons. Well guess what - the simulations are poor. Very poor. We can barely predict the existence of chemical bonds (in fact I don't think we can at all - I'm just not sure because I'm giving the chemists who claim we can the benefit of the doubt). What about a real commonplace molecule like water. It turns out that we understand next to nothing about water. Just about every physical prediction you'd like to make about water from its molecular structure is wrong (unless you do the standard computational chemist trick of tweaking your parameters after running the failed simulation).

Protein folding isn't going to be solved by getting innocent kids all over the world to donate their CPUs. It'll be solved by some chemist pulling their finger out and figuring out a better way to do it.

So switch your PCs over to seti@home now! It's no less likely to succeed and it's a lot more fun.


Wow. (none / 0) (#1)
by TheReverand on Thu Nov 15th, 2001 at 05:51:12 PM PST
I've never seen anyone so angry about computational chemistry.

Then again, I've never heard of computational chemistry before today.


 
who is this guy? (none / 0) (#2)
by nathan on Thu Nov 15th, 2001 at 06:08:28 PM PST
And how did he get so cool? This could be an article...

Nathan
--
Li'l Sis: Yo, that's a real grey area. Even by my lax standards.

 
fascinating... (none / 0) (#3)
by error27 on Mon Nov 19th, 2001 at 03:51:31 PM PST
thanks for the article. I'd never thought about it that way before.


 

All trademarks and copyrights on this page are owned by their respective companies. Comments are owned by the Poster. The Rest ® 2001, 2002, 2003 Adequacy.org. The Adequacy.org name, logo, symbol, and taglines "News for Grown-Ups", "Most Controversial Site on the Internet", "Linux Zealot", and "He just loves Open Source Software", and the RGB color value: D7D7D7 are trademarks of Adequacy.org. No part of this site may be republished or reproduced in whatever form without prior written permission by Adequacy.org and, if and when applicable, prior written permission by the contributing author(s), artist(s), or user(s). Any inquiries are directed to legal@adequacy.org.