Magic Bitboards -- What's the current state of the art?

Code, algorithms, languages, construction...
Post Reply
QueenDrop
Posts: 1
Joined: Wed Jan 20, 2021 1:13 am

Magic Bitboards -- What's the current state of the art?

Post by QueenDrop » Wed Jan 20, 2021 1:38 am

I've been reading about these on chessprogramming.org and it seems like a really intriguing problem to me. I'd love to dive deeper into it, but first I want to figure out if what I'm missing by only reading that page:

- Are the values listed on https://www.chessprogramming.org/Best_Magics_so_far accurate?
- The page describes a "stunning idea" of sharing attacks by Lasse Hansen. Has this been picked up by engines? The page says it's possible to share tables by 2 rooks, it seems to be that you could do even better than that, like up to 8 rooks for certain configurations or maybe more. Has any further work been done on this?
- Are there other advancements not listed on the wiki page?

Michael Sherwin
Posts: 10
Joined: Tue Dec 19, 2017 6:05 am
Real Name: Michael Sherwin

Re: Magic Bitboards -- What's the current state of the art?

Post by Michael Sherwin » Thu Mar 16, 2023 11:16 pm

Is this place so dead that a post from January 2021 is still on the first page. TalkChess died today. Maybe this place will liven up a bit.

The current state of the art may have just changed from Black Magic bitboards to Kindergarten Supper SISSY bitboards.

A recent modification and test by martinn at TalkChess has shown a 13% increase in performance of KSS over BM. Of course compiler and cpu can make a huge difference.

Here is working code for KSS if someone would like to investigate further.

Code: Select all

#pragma once
// Datum  : 21.01.2023; updated code as of 16.03.2023 (martinn)
// Author : Michael J Sherwin 2023
// Content: The name is Kindergarten Super SISSY Bitboards - https://www.talkchess.com/forum3/viewtopic.php?f=7&t=81234&start=30
//          The code can be further "minimized" according to the author
//			C++20 constexpr port by Daniel Infuehr

#include <stdint.h>

namespace Chess_Lookup::KGSSB
{
	static uint64_t vMask[64];
	static uint64_t hMask[64];
	static uint64_t dMask[64];
	static uint64_t aMask[64];
	static uint64_t vSubset[64][64];
	static uint64_t hSubset[64][64];
	static uint64_t dSubset[64][64];
	static uint64_t aSubset[64][64];

	// new lookup table for shifts in calculation of hIndex - martinn
	static unsigned int shift_horizontal_table[64]; // new lookup table for shifts in calculation of hIndex


	static constexpr uint64_t Size = (64 * 64 * 4 + 64 * 4) * sizeof(uint64_t);

	enum { FILEa, FILEb, FILEc, FILEd, FILEe, FILEf, FILEg, FILEh };

	enum { RANK1, RANK2, RANK3, RANK4, RANK5, RANK6, RANK7, RANK8 };


	static void InitSquare(int sq)
	{
		int ts, dx, dy;
		uint64_t occ, index;
		int x = sq % 8;
		int y = sq / 8;

		// Initialize Kindergarten Super SISSY Bitboards
		// diagonals
		for (ts = sq + 9, dx = x + 1, dy = y + 1; dx < FILEh && dy < RANK8; dMask[sq] |= 1ull << ts, ts += 9, dx++, dy++);
		for (ts = sq - 9, dx = x - 1, dy = y - 1; dx > FILEa && dy > RANK1; dMask[sq] |= 1ull << ts, ts -= 9, dx--, dy--);

		// anti diagonals
		for (ts = sq + 7, dx = x - 1, dy = y + 1; dx > FILEa && dy < RANK8; aMask[sq] |= 1ull << ts, ts += 7, dx--, dy++);
		for (ts = sq - 7, dx = x + 1, dy = y - 1; dx < FILEh && dy > RANK1; aMask[sq] |= 1ull << ts, ts -= 7, dx++, dy--);

		// diagonal indexes
		for (index = 0; index < 64; index++) {
			dSubset[sq][index] = 0;
			occ = index << 1;
			if ((sq & 7) != FILEh && (sq >> 3) != RANK8) {
				for (ts = sq + 9; ; ts += 9) {
					dSubset[sq][index] |= (1ull << ts);
					if (occ & (1ull << (ts & 7))) break;
					if ((ts & 7) == FILEh || (ts >> 3) == RANK8) break;
				}
			}
			if ((sq & 7) != FILEa && (sq >> 3) != RANK1) {
				for (ts = sq - 9; ; ts -= 9) {
					dSubset[sq][index] |= (1ull << ts);
					if (occ & (1ull << (ts & 7))) break;
					if ((ts & 7) == FILEa || (ts >> 3) == RANK1) break;
				}
			}
		}

		// anti diagonal indexes
		for (index = 0; index < 64; index++) {
			aSubset[sq][index] = 0;
			occ = index << 1;
			if ((sq & 7) != FILEa && (sq >> 3) != RANK8) {
				for (ts = sq + 7; ; ts += 7) {
					aSubset[sq][index] |= (1ull << ts);
					if (occ & (1ull << (ts & 7))) break;
					if ((ts & 7) == FILEa || (ts >> 3) == RANK8) break;
				}
			}
			if ((sq & 7) != FILEh && (sq >> 3) != RANK1) {
				for (ts = sq - 7; ; ts -= 7) {
					aSubset[sq][index] |= (1ull << ts);
					if (occ & (1ull << (ts & 7))) break;
					if ((ts & 7) == FILEh || (ts >> 3) == RANK1) break;
				}
			}
		}

		// horizontal lines
		for (ts = sq + 1, dx = x + 1; dx < FILEh; hMask[sq] |= 1ull << ts, ts += 1, dx++);
		for (ts = sq - 1, dx = x - 1; dx > FILEa; hMask[sq] |= 1ull << ts, ts -= 1, dx--);

		// vertical indexes
		for (index = 0; index < 64; index++) {
			vSubset[sq][index] = 0;
			uint64_t blockers = 0;
			for (int i = 0; i <= 5; i++) {
				if (index & (1ull << i)) {
					blockers |= (1ull << (((5 - i) << 3) + 8));
				}
			}
			if ((sq >> 3) != RANK8) {
				for (ts = sq + 8; ; ts += 8) {
					vSubset[sq][index] |= (1ull << ts);
					if (blockers & (1ull << (ts - (ts & 7)))) break;
					if ((ts >> 3) == RANK8) break;
				}
			}
			if ((sq >> 3) != RANK1) {
				for (ts = sq - 8; ; ts -= 8) {
					vSubset[sq][index] |= (1ull << ts);
					if (blockers & (1ull << (ts - (ts & 7)))) break;
					if ((ts >> 3) == RANK1) break;
				}
			}
		}

		// horizontal indexes
		for (index = 0; index < 64; index++) {
			hSubset[sq][index] = 0;
			occ = index << 1;
			if ((sq & 7) != FILEh) {
				for (ts = sq + 1; ; ts += 1) {
					hSubset[sq][index] |= (1ull << ts);
					if (occ & (1ull << (ts & 7))) break;
					if ((ts & 7) == FILEh) break;
				}
			}
			if ((sq & 7) != FILEa) {
				for (ts = sq - 1; ; ts -= 1) {
					hSubset[sq][index] |= (1ull << ts);
					if (occ & (1ull << (ts & 7))) break;
					if ((ts & 7) == FILEa) break;
				}
			}
		}
	}

	static void Init()
	{
		for (int i = 0; i < 64; i++) {
			InitSquare(i);
		}

		for (int i = 0; i < 64; i++) {
		  shift_horizontal_table[i] = (i & 56) + 1;
		}
	}

	static constexpr uint64_t file_b2_b7 = 0x0002020202020200ull;
	static constexpr uint64_t file_a2_a7 = 0x0001010101010100ull;
	static constexpr uint64_t diag_c2h7 = 0x0080402010080400ull;

	static uint64_t Bishop(int sq, uint64_t occ)
	{
	  return
		dSubset[sq][(((occ & dMask[sq]) * file_b2_b7) >> 58)] |
		aSubset[sq][(((occ & aMask[sq]) * file_b2_b7) >> 58)];
	}

	static uint64_t Rook(int sq, uint64_t occ)
	{
	  return
		hSubset[sq][(occ >> shift_horizontal_table[sq]) & 63] |
		vSubset[sq][((((occ >> (sq & 7)) & file_a2_a7) * diag_c2h7) >> 58)];
	}


	static uint64_t Queen(int sq, uint64_t occ)
	{
		return Rook(sq, occ) | Bishop(sq, occ);
	}
	
}
It appears all that can be done has been done to optimize this code. So for better or worse this ends an ~16 year quest to outperform Magic Bitboards.

JoAnnP38
Posts: 1
Joined: Fri Mar 17, 2023 1:45 pm
Real Name: JoAnn Peeler

Re: Magic Bitboards -- What's the current state of the art?

Post by JoAnnP38 » Fri Mar 17, 2023 2:46 pm

Michael Sherwin wrote:
Thu Mar 16, 2023 11:16 pm
TalkChess died today. Maybe this place will liven up a bit.
Has that ever happened before? I was just assuming someone forgot to renew the domain name subscription and that it would be back up as soon as that is resolved. Do you have any other info?

It's good to see your post!

Michael Sherwin
Posts: 10
Joined: Tue Dec 19, 2017 6:05 am
Real Name: Michael Sherwin

Re: Magic Bitboards -- What's the current state of the art?

Post by Michael Sherwin » Fri Mar 17, 2023 3:43 pm

JoAnnP38 wrote:
Fri Mar 17, 2023 2:46 pm
Michael Sherwin wrote:
Thu Mar 16, 2023 11:16 pm
TalkChess died today. Maybe this place will liven up a bit.
Has that ever happened before? I was just assuming someone forgot to renew the domain name subscription and that it would be back up as soon as that is resolved. Do you have any other info?

It's good to see your post!
Thanks! It's back up.

ZamChess
Posts: 196
Joined: Sun Dec 19, 2021 10:24 am

Re: Magic Bitboards -- What's the current state of the art?

Post by ZamChess » Wed Mar 22, 2023 9:01 am

Hi Michael,

I have been reading about RomiChess for more than an hour on talkchess.com.
I find all your work with RomiChess very interesting. I don’t know if you’re still working on it or have given up. I congratulate you with all my heart for your work.
I wanted to download RomiChess from the net but I can only find expired links...
Please post a link here so I can download it.
Thanks in advance.

PS: I don’t know why OpenChess won’t let me send you a message, that’s why I wrote here.

Michael Sherwin
Posts: 10
Joined: Tue Dec 19, 2017 6:05 am
Real Name: Michael Sherwin

Re: Magic Bitboards -- What's the current state of the art?

Post by Michael Sherwin » Thu Mar 23, 2023 9:21 am

ZamChess wrote:
Wed Mar 22, 2023 9:01 am
Hi Michael,

I have been reading about RomiChess for more than an hour on talkchess.com.
I find all your work with RomiChess very interesting. I don’t know if you’re still working on it or have given up. I congratulate you with all my heart for your work.
I wanted to download RomiChess from the net but I can only find expired links...
Please post a link here so I can download it.
Thanks in advance.

PS: I don’t know why OpenChess won’t let me send you a message, that’s why I wrote here.
Hi, thanks for the kind words. I uploaded RomiChess to MediaFire.
https://www.mediafire.com/file/ju22evcm ... n.zip/file

ZamChess
Posts: 196
Joined: Sun Dec 19, 2021 10:24 am

Re: Magic Bitboards -- What's the current state of the art?

Post by ZamChess » Fri Mar 24, 2023 7:10 am

Thank you very much !

With respect and friendship.

Post Reply