Deeper Blue Power - When Will PC's Match it?

[Sean Evans]

Hello,

When Deeper Blue vs. Kasparov happened Don believed we would all be processing on PCs at the same strength as the Deeper Blue hardware. As I recall DB was chugging along at 200,000,000 nodes per second, I don't believe a typical one CPU PC can match that yet.

So when will I have a PC that will match or better the DB hardware ??

Sean

[Terry McCracken]

http://www.cray.com/Home.aspx?404;http: ... index.html

If you're rich, I'm sure Cray can build it for you for $100,000.00 USD.

But on a single processor? Not for a while yet, maybe with spintronics when that technology is feasible.

[bob]

Hello,

When Deeper Blue vs. Kasparov happened Don believed we would all be processing on PCs at the same strength as the Deeper Blue hardware. As I recall DB was chugging along at 200,000,000 nodes per second, I don't believe a typical one CPU PC can match that yet.

So when will I have a PC that will match or better the DB hardware ??

Sean

The comparison is more complex than that. The DB processors were not as efficient with respect to move ordering as today's software programs are. The 1997 deep blue didn't have hash tables in the hardware. The processors were capable of using one, but the multi-ported memory was never designed. So while it could peak at 1 billion nodes per second, and average 200M nodes per second, a program today running at 200M would be significantly better. And we can actually get there with a parallel architecture. I run at 20M on a dual-cpu quad-core box (8 cores total). There are 16 x 4-way boxes already out and available...

[Ovyron]

As I recall DB was chugging along at 200,000,000 nodes per second

Most of them garbage. Rybka showed that it's more important to look at the right nodes than to look at as much nodes as possible (Even though you have to multiply Rybka's nodes by about 17 to get a more reliable figure.)

Take out all the evaluation from some program and the node count will for sure skyrocket, but the program with evaluation will also for sure beat it badly.

[Terry McCracken]

Hello,

When Deeper Blue vs. Kasparov happened Don believed we would all be processing on PCs at the same strength as the Deeper Blue hardware. As I recall DB was chugging along at 200,000,000 nodes per second, I don't believe a typical one CPU PC can match that yet.

So when will I have a PC that will match or better the DB hardware ??

Sean

The comparison is more complex than that. The DB processors were not as efficient with respect to move ordering as today's software programs are. The 1997 deep blue didn't have hash tables in the hardware. The processors were capable of using one, but the multi-ported memory was never designed. So while it could peak at 1 billion nodes per second, and average 200M nodes per second, a program today running at 200M would be significantly better. And we can actually get there with a parallel architecture. I run at 20M on a dual-cpu quad-core box (8 cores total). There are 16 x 4-way boxes already out and available...

How much money would you be looking at to build a beast that could sustain 200M nodes per sec. ?

I would think quite a bit.

Can you grab some of that Bail Out funding?

P.S. Check next door, they have your country pushing daisies

[Sean Evans]

You fellows seem to be confusing the issue:

The question is when will a single CPU be able to do 200M nodes?

Don said this year, looks like he was wrong !

Cordially,

Sean

[Terry McCracken]

You fellows seem to be confusing the issue:

The question is when will a single CPU be able to do 200M nodes?

Don said this year, looks like he was wrong !

Cordially,

Sean

I got it and gave you an answer...it's a ways off, but with multi-core processors it could be done today if the money was there and the will.

So Don Daily isn't that far wrong.

[M ANSARI]

I would think that a dual socket 8 core per CPU Nehalem based (total 16 cores) could reach the 200 M node count. As for DB strength chesswise ... an overclocked Quadcore running Rybka 3 would probably spank it badly. It is too bad that the system was dismantled ... what were IBM thinking. This would have been a great piece to put in a museum or even test the abilities of it compared to new hardware and software. I cannot for the life of me understand why the system was dismantled and allowed to "evaporate" into thin air.

[sje]

I cannot for the life of me understand why the system was dismantled and allowed to "evaporate" into thin air.

Much like Fischer quitting after winning the 1972 World Championship, after the match DB had no where to go but down.

The alternative would be for IBM to have continued funding for constant improvements to stay ahead of the upcoming pack of machine challengers. Apparently IBM saw little marketing value to be had from this and so retired DB permanently.

[Dann Corbit]

I would think that a dual socket 8 core per CPU Nehalem based (total 16 cores) could reach the 200 M node count. As for DB strength chesswise ... an overclocked Quadcore running Rybka 3 would probably spank it badly. It is too bad that the system was dismantled ... what were IBM thinking. This would have been a great piece to put in a museum or even test the abilities of it compared to new hardware and software. I cannot for the life of me understand why the system was dismantled and allowed to "evaporate" into thin air.

If this assessment were true:

"As for DB strength chesswise ... an overclocked Quadcore running Rybka 3 would probably spank it badly"
(I do not know if it is correct or not)

Then wouldn't that provide the reason that they dismantled it?
Where is the myth of invincibility if the hundred million dollar monster can be dashed to bits by a $5000 box?

Like a boxer who retires undefeated, nobody can say what would have happened if he faced some new opponent at some time in the future.

I guess that all the scientists were crushed at the dismantling and all the marketing execs were clicking their heels in glee.