[Beowulf] commercial clusters

Robert G. Brown rgb at phy.duke.edu
Fri Sep 29 14:22:47 PDT 2006


On Fri, 29 Sep 2006, Vincent Diepeveen wrote:

> The basic flaw i see in some assumptions taken here is that you as a human 
> want
> to do a running contest with a car.
>
> In reality that isn't the case.
>
> People just want to see whose car is better, not whether they can outrun the 
> car.

No, no, no.  Most NORMAL human beings wouldn't give a rodent's furry
behind for whether or not their computer chess program was faster than
my computer chess program.  Most normal human beings DO view computer
chess as running a footrace against a car, and insist on being able to
set the car's speed control to a point where they have a chance of
winning.  Very, very few people -- and you probably know just about all
of them personally, at a guess -- care about building a computer chess
program that can beat other computer chess programs or grand masters.

Even if you don't know them ALL, consider the number compared to your
original billion.  Ooo, small small, small, and the buy-in required to
be a serious contender is large, large, large.  Not the sort of market
to make a venture capitalist go "Wow, if I do this I can be the next
Microsoft", or even the next Blizzard.

You want to get rich on computer games, invent the nextgen successor to
World of Warcraft, or figure out how to turn Google Earth into a
realtime game.  There is a mass market for games, but chess ain't it.

No disrespect intended, by the way -- it's just that most people don't
really like COMPUTERS the way you and I do -- as you yourself often
point out.  All they want to do is run applications, and don't even know
HOW to judge those applications save that they do or don't do what they
"expect" them to do.

>
> Additional they want to analyze their own running (most important feature) in 
> the most objective manner possible, or they want to compete with it in a 300 
> KM/h contest.
>
> To look like a winner, people are prepared to pay any price.

If by "people" you mean the few hundred, or maybe few thousand, people
who are computer AND chess fanatics and spend their time on computer
chess engines and software, well, you'd be one who would know;-) About
the "any price" bit.

The other 5,999,997,000 people on the planet are busy earning their
daily bread, doing research with a somewhat more direct payoff, playing
Go or World of Warcraft or Soccer instead of Chess, or being rich and
jetting all over the place snorting cocaine.  Most of them don't really
know that "computer chess" as a human endeavor exists.  Of those who do,
most don't care about it beyond the use of maybe Gnuchess/Crafty or a
$10 chess game for Windows purchase over the counter at Best Buy.  This
is reasonable, because those programs can currently beat even masters.

    rgb

>
> Vincent
>

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu





More information about the Beowulf mailing list