View Single Post
Old March 16th 17, 05:56 PM posted to
external usenet poster
Posts: 3,536
Default 100 million FICS games

On Monday, March 13, 2017 at 8:35:16 PM UTC-4, Rainer wrote:
On 13/03/2017 19:46, raylopez99 wrote:
On Monday, March 13, 2017 at 6:07:22 AM UTC-4, Rainer wrote:
On 12/03/2017 23:32, raylopez99 wrote:

This week's Economist has a special section on how quantum
computers have gone mainstream. Chess tree to be solved soon,
to xyz moves, depending on what kind of quantum computer you
have (the more qubits you have--which is hardware related--the
more moves you can search; I can see in the future the entire
chess tree to say 1000 moves deep being completed exhaustively
searched with a top-of-the-line quantum computer).

Science fiction says Andy Walker, but the poor chap, being
retired and retarded, doesn't keep up with the literature!


Rubbish. Quantum computers are able to solve a limited subset of
computational problems efficiently. Playing chess is none of these

Quantum computers will contribute as much to the solution of chess
as you--nothing.

Cheers, Rainer

Stupidity rains down on us. You, not a programmer like me, know
nothing --NOTHING--about what you speak of. If you know how RLC
analog computers work, and how they can solve second-order
differential equations if you set them up correctly, you'd
understand that programming a general purpose quantum computer is the
same. You program the quantum computer to assume the state of the
chess tree, maybe with one or more qubits representing legal moves
(how many legal moves are there in chess per position? That would be
your maximum number of qubits). They already have a dedicated
quantum computer to solve the Traveling Salesman problem, which is an
O(n!) complex problem (do you even know what Big-Oh notation means?
Doubtful), and the chess tree is no more difficult than that.

And oh--big Oh--I have solved such problems and even more difficult
problems in my professional capacity. As part of a team, but I have
done it.


More rubbish. You fail to understand the very fundamentals of quantum
computing. The following simple question will make it clear:

How do you verify the output of a (hypothetical) quantum chess solver?

You didn't think of it, right?


Wow you're denser than water Rainer. How you say? That's easy. If the output of a quantum computer assumes the state of the chess board, the value of the pieces at any given position would determine the potential of the position, and the results would flow naturally like water flows downhill. Example: the board position in a quantum computer shows all legal moves simultaneously. The best move to make is the one with the highest non-equilibrium in material for either Black or White. Example: equal position but mate in one means the 'mate in one' is the best move unless the other side can refute it, then, if not, it would be some other move. Possibly there's no best move that wins material so you pick a 'drawish' move (randomly, or using some trivial evaluation function, like centralize knights or rooks to open files). Unlike today's chess engine, there's NO NEED for an evaluation function, except a trivial one in the event there's no win of material move. No need because you're examining all moves in the chess tree, and depending on how much hardware you have, you can examine all the moves in the chess tree up to say 200 plies (if your quantum computer is powerful enough)..

Nobel prize in physics follows for me.