Stockfish (20 kn/s) vs Fritz in Bahrain(3600 kn/s)

[MikeB]

I made another test. Now time control is 10 minutes/game

Stockfish (10 kn /s) vs Junior 6 (5000 kn/s)

Program Elo + - Games Score Av.Op. Draws

1 Stockfish 271119 64 BMI2 : 2916 61 59 150 77.3 % 2702 14.7 %
2 Junior 6.0 : 2702 59 61 150 22.7 % 2916 14.7 %

(Fritz 11 Gui, 10 minutes/game, Core i7 9750h, Book2014 opening book, ponder off)

Individual statistics:

1 Stockfish 271119 64 BMI2 : 2916 150 (+105,= 22,- 23), 77.3 %

Junior 6.0 : 150 (+105,= 22,- 23), 77.3 %

2 Junior 6.0 : 2702 150 (+ 23,= 22,-105), 22.7 %

Stockfish 271119 64 BMI2 : 150 (+ 23,= 22,-105), 22.7 %

http://www.mediafire.com/file/hcd3kv4ki ... 5.pgn/file



The performance of Deep Junior 6 is 2702 at Dortmund 2000 tournament. (Hardware: 8 x Pentium 700 Mh)

https://theweekinchess.com/html/twic297.html#2

In my test Junior 6 % 60 faster than Deep Junior 6 (Dortmund 2000)


The human elo of Stockfish with 10 kn/s is ~2950 according to this test

Just keep in mind that Deep Junior performance rating of 2702 was based on 9 games and has error bars of about ~150 Elo , plus or minus - so that means your little test is plus or minus 150 Elo , plus the Elo error bar from your test or about 25 Elov( assuming your test was 150 games or so) - so Stockfish rating at 10/kns is somewhere between ~2775 to ~3125 - according to your test.

[Vinvin]

Thanks for this 2 matches, Mehmet !
Very good memory and it's very interesting to see how engines improved since then !!

[mehmet1921]

My third test is Stockfish (20 kn/s) vs Shredder 9

Program Elo + - Games Score Av.Op. Draws

1 Stockfish 101219 64 BMI2 : 3108 66 63 123 80.9 % 2858 20.3 %
2 Shredder 9 : 2858 63 66 123 19.1 % 3108 20.3 %

(Fritz 11 Gui, 5 minutes/game, Core i7 9750h, Book2014 opening book, ponder off)


Individual statistics:

1 Stockfish 101219 64 BMI2 : 3108 123 (+ 87,= 25,- 11), 80.9 %

Shredder 9 : 123 (+ 87,= 25,- 11), 80.9 %

2 Shredder 9 : 2858 123 (+ 11,= 25,- 87), 19.1 %

Stockfish 101219 64 BMI2 : 123 (+ 11,= 25,- 87), 19.1 %

(In 124th game Stockfish is ahead with +8 pawn but the game failed)

http://www.mediafire.com/file/q18wfcowx ... 2.pgn/file


The performance of Shredder 9 is 2758 elo at 2005 Mercosur Cup (8.5/10 points)

The performance of Shredder 9 is 2807 elo at 2005 Vıcente Lopez ITT 2005 (9.5/10 points)

https://en.chessbase.com/post/shredder- ... a-mercosur

The average rating of Shredder 9 is 2783. The hardware of Shredder is Pentium 3.5 Ghz.
In my test Shredder 9 is % 200 faster than Shredder 9 at Mercosur Cup. So I set the elo of Shredder 9 to 2858 ( 50 elo for doubling)

[mehmet1921]

In my 3 tests againts Fritz, Junior, Shredder the average human elo of Stockfish (20kn/s) is 3035.

In my test against Junior the elo of Stockfish (10 kN/s) is 2950. If we give doubling 50 elo than Stockfish (20 kN/s) has ~3000 elo.

First of all I was expecting a higher elo. Because the performance of Pocket Fritz 4 is 2938 elo at 2009 Mercosur Cup
Pocket Fritz uses Hiarcs 13.1 engine. And Hiarcs 13.1 searches ~20 kn/s ( HTC Touch HD mobile phone ,1 core, 528 mHz)

The elo of Stockfish 10 is 513 elo higher than Hiarcs 13.1. The elo of Stockfish dev. version is ~550 elo higher than Hiarcs 13.1

In my test the human elo of Stockfish (20 kn/s)is only 100 elo higher than Hiarcs 13.1 (20 kn/s)

https://en.chessbase.com/post/breakthro ... enos-aires

[Vinvin]

First of all I was expecting a higher elo. Because the performance of Pocket Fritz 4 is 2938 elo at 2009 Mercosur Cup
Pocket Fritz uses Hiarcs 13.1 engine. And Hiarcs 13.1 searches ~20 kn/s ( HTC Touch HD mobile phone ,1 core, 528 mHz)

Don't forget HIARCS was a slow searcher (or it counts nodes differently).
To compare to Fritz, Junior or Shredder, you have to multiply by some factor (from my bad memory, I should say around x5).

[mehmet1921]

First of all I was expecting a higher elo. Because the performance of Pocket Fritz 4 is 2938 elo at 2009 Mercosur Cup
Pocket Fritz uses Hiarcs 13.1 engine. And Hiarcs 13.1 searches ~20 kn/s ( HTC Touch HD mobile phone ,1 core, 528 mHz)

Don't forget HIARCS was a slow searcher (or it counts nodes differently).
To compare to Fritz, Junior or Shredder, you have to multiply by some factor (from my bad memory, I should say around x5).

Yes you are right.
In my test Fritz 10 calculated 3525 kn/s at starting position (1 minute). Hiarcs 10 calculated 707 kn/s at starting position (1 minute). (Core i7 9750h)

Fritz calculates 5x position according to Hiarcs in same time.

[mehmet123]

I made another test to measure the strenght of Stockfish 11 at 1 kn/s.

Elo performance of Chess Tiger 14 at Republica Argentina Chess Master is 2788 (9.5/ 11, Pentium 3 866 Mhz)

Elo performance of Chess Tiger 14 is 2805 elo at Najdorf Memorial Open 2001 (5.5/ 6, Pentium 3 866 Mhz)
https://en.chessbase.com/post/the-revenge-of-the-humans

The average elo of Chess Tiger 14 is 2794

In my laptop (Core i7- 9750h, 16 Gb Ram) the speed of Chess Tiger is 8x faster than Chess Tiger at Argentina 2001.

The speed of Stockfish is 8 kn/s in my test (8 x 1 kn/s)

Program Elo + - Games Score Av.Op. Draws

1 Chess Tiger 14.0 : 2794 47 47 174 64.4 % 2692 23.0 %
2 Stockfish 11 64 BMI2 : 2692 47 47 174 35.6 % 2794 23.0 %

Individual statistics:

1 Chess Tiger 14.0 : 2794 174 (+ 92,= 40,- 42), 64.4 %

Stockfish 11 64 BMI2 : 174 (+ 92,= 40,- 42), 64.4 %

2 Stockfish 11 64 BMI2 : 2692 174 (+ 42,= 40,- 92), 35.6 %

Chess Tiger 14.0 : 174 (+ 42,= 40,- 92), 35.6 %

(Fritz 11 Gui, i7 9750h, 5 minutes/game, Vitamin Book/ 8 move)
http://www.mediafire.com/file/k2xj56ji4 ... 1.pgn/file

Stockfish at 1 kn/s has 2692 human elo



ELo performance of Hiarcs 8 at Republica Argentina Chess Master is 2578 https://en.chessbase.com/post/human-tri ... rcs-8-just

In my laptop (Core i7- 9750h, 16 Gb Ram) the speed of Hiarcs 8 is 8x faster than Chess Tiger at Argentina 2002.

The speed of Stockfish is 8 kn/s in my test (8 x 1 kn/s)


Program Elo + - Games Score Av.Op. Draws

1 Hiarcs 8 : 2578 59 58 110 56.8 % 2530 20.9 %
2 Stockfish 11 64 BMI2 : 2530 58 59 110 43.2 % 2578 20.9 %

Individual statistics:

1 Hiarcs 8 : 2578 110 (+ 51,= 23,- 36), 56.8 %

Stockfish 11 64 BMI2 : 110 (+ 51,= 23,- 36), 56.8 %

2 Stockfish 11 64 BMI2 : 2530 110 (+ 36,= 23,- 51), 43.2 %

Hiarcs 8 : 110 (+ 36,= 23,- 51), 43.2 %

(Fritz 11 Gui, i7 9750h, 5 minutes/game, Vitamin Book/ 8 move)
http://www.mediafire.com/file/74akd6gwt ... 2.pgn/file

Stockfish at 1 kn/s has 2530 human elo


According to these tests Stockfish at 1 kn/s has 2612 human elo (2694 + 2530) /2)

[mehmet123]

According to the my tests.

The human elo of Stockfish with 1 kn/s is 2612.

The human elo of Stockfish with 10 kn/s is 2950.

The human elo of Stockfish with 20 kn/s is 3035.

[Vinvin]

Thanks !
Very interesting !

[Alayan]

For such comparisons, the TC is critical. If Chess Tiger 14 is 8x faster on your hardware, then dividing the TC by 8 and giving SF 8knps is the thing to do, but I don't think 40 minutes/game was the original TC at the events where Chess Tiger beat those human GMs.

if it was something like classical with 2h for the first 40 moves then some additional time ; this could change the result by a fair amount.

Where do the 10knps and 20knps results come from ?