Could a current top player beat the same DeepBlue that beat Kasparov?

@RamblinDave said in #10:
> I've actually played against it on basically equivalent hardware to the setup that beat Kasparov (ie the same model of IBM mainframe). Got trounced, obviously - it played a Kan Sicilian and somehow ended up rolling its center pawns forward and pushing me off the board.

That would make for an interesting "Ocean's Eleven" quest. Go into these museums, get both halves and then connect them up and play it.

@RamblinDave said in #10:
> I've actually played against it on basically equivalent hardware to the setup that beat Kasparov (ie the same model of IBM mainframe).

Not true.

Only very few people (aside from the IBM insiders) got to play the Deep Blue hardware. This project was practically on the edge of failure of their custom chips and nobody from their R&D department would risk frying (electro-migrating to failure) their precious, hand selected chips that worked correctly.

If you are an IBM ex-insider, you would've known the difference between z/Arch mainframe and POWER supermicro and wouldn't mix those two.

Okay, I've rechecked my old photo and it's a "similar" Type 7025 RS/6000 SP. But no, it didn't have the VLSI chips and so it was presumably weaker than the original. Thanks for ruining my fun story.

@WorldRenownPatzer
Chess is expanding and theory is developed as time passes but modern top players who have achieved great heights miss significant factor which is quite seen.
Due to engines, since their arrival players have become dependent on technology (engine) and become mentally weak and lost creativity in games.
Players of older era would thrive over current players as I have explained before on forums.

@Akbar2thegreat said in #14:
> @WorldRenownPatzer
> Chess is expanding and theory is developed as time passes but modern top players who have achieved great heights miss significant factor which is quite seen.
> Due to engines, since their arrival players have become dependent on technology (engine) and become mentally weak and lost creativity in games.
> Players of older era would thrive over current players as I have explained before on forums.

Or we would realize that "creativity" doesn't equal sound moves. The obsession with Mikail Tal would then pop like a balloon, because we would realize he wasn't that great.

@WorldRenownPatzer said in #15:
> Or we would realize that "creativity" doesn't equal sound moves. The obsession with Mikail Tal would then pop like a balloon, because we would realize he wasn't that great.
You are wrong.
Creativity is mostly positional. And it's not easy for any player even at top level to be regarded as creative.
There were great creative players: Morphy, Lasker, Capablanca, Fisher, Tal, Nezhmetdinov, Spassky, Kasparov.
And playing unsound is different. That too for very long period even risking World Championship and getting longest unbeaten streak of 96 games (47 wins) with greater win percentage with out of mind and complex moves is nearly impossible. And Tal actually did that.

@Akbar2thegreat said in #16:
> You are wrong.
> Creativity is mostly positional. And it's not easy for any player even at top level to be regarded as creative.
> There were great creative players: Morphy, Lasker, Capablanca, Fisher, Tal, Nezhmetdinov, Spassky, Kasparov.
> And playing unsound is different. That too for very long period even risking World Championship and getting longest unbeaten streak of 96 games (47 wins) with greater win percentage with out of mind and complex moves is nearly impossible. And Tal actually did that.

I see one person's creation go unattested as being very non-creative.

I would need to see a "brilliancy" which the computer engines show as the best move in a decisive position to really call it creative. Until then, it's just players throwing darts. Eventually one will hit the bull's eye.

This is no different that sawing off a rifle and shooting. You are eventually going to hit your target. Let's look for sniper one shot brilliancies supported by engines.

It BEAT Kasparov ,,, So only Carlsen would stand a chance but be would be out rated by it as it was rated higher than Kasparov at the time ... It was ahead of it's time at the time it BEAT Kasparov

@RamblinDave said in #13:
> Okay, I've rechecked my old photo and it's a "similar" Type 7025 RS/6000 SP. But no, it didn't have the VLSI chips and so it was presumably weaker than the original. Thanks for ruining my fun story.

Oh, well...

What you played wasn't even close to the true Deep Blue as it implemented major portion of the search algorithm in the VLSI using the design flow that IBM couldn't properly simulate. That's why the project was so close to failing.

My best guess is that you might have played some version of Crafty. It was a very portable program written in C language that was at one time portion of the SPEC benchmark, thus many people used it for hardware testing. But its architecture is completely different from the Deep Blue and actually fits quite well SMP supermicros like the large models of RS/6000, nowadays called System p.

at least it could serve as external milestone. perhaps having it in SF dev pools would also be interesting.
relaxing time constraints perhaps (like not symmetric). If there was some in-game self-accuracy measure that could tell that no variation was truncated and hasted to spit some answer. perhaps competing on number of node evaluated. or number of legal branching points.. i don't know how to compare when time is a factor. that might be true for humans too.

saying same hardware might not be enough. unless strength is really a time included notion. (I guess it is, but not sure).