Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games)
Author Message mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sat Jul 03, 2021 3:03 pm VIDEO Hardware Threadripper 2950x, RTX 2080 TI, 64 Gb Ram, SSD 2 TB Evo 970 Plus.
Book Perfect Book 2021
7 Man Table bases.
4 GB Hash
32 Threads
TC = 10m+5s
Stockfish 14 Information.
Code: Today, we have the pleasure to announce Stockfish 14. As usual, downloads will be freely available at https://stockfishchess.org The engine is now significantly stronger than just a few months ago, and wins four times more game pairs than it loses against the previous release version [0]. Stockfish 14 is now at least 400 Elo ahead of Stockfish 7, a top engine in 2016 [1]. During the last five years, Stockfish has thus gained about 80 Elo per year. Stockfish 14 evaluates positions more accurately than Stockfish 13 as a result of two major steps forward in defining and training the efficiently updatable neural network (NNUE) that provides the evaluation for positions. First, the collaboration with the Leela Chess Zero team - announced previously [2] - has come to fruition. The LCZero team has provided a collection of billions of positions evaluated by Leela that we have combined with billions of positions evaluated by Stockfish to train the NNUE net that powers Stockfish 14. The fact that we could use and combine these datasets freely was essential for the progress made and demonstrates the power of open source and open data [3]. Second, the architecture of the NNUE network was significantly updated: the new network is not only larger, but more importantly, it deals better with large material imbalances and can specialize for multiple phases of the game [4]. A new project, kick-started by Gary Linscott and Tomasz Sobczyk, led to a GPU accelerated net trainer written in pytorch.[5] This tool allows for training high-quality nets in a couple of hours. Finally, this release features some search refinements, minor bug fixes and additional improvements. For example, Stockfish is now about 90 Elo stronger for chess960 (Fischer random chess) at short time control. The Stockfish project builds on a thriving community of enthusiasts (thanks everybody!) that contribute their expertise, time, and resources to build a free and open-source chess engine that is robust, widely available, and very strong. We invite our chess fans to join the fishtest testing framework and programmers to contribute to the project on github [6]. Stay safe and enjoy chess! The Stockfish team [0] https://tests.stockfishchess.org/test... [1] https://nextchessmove.com/dev-builds [2] https://stockfishchess.org/blog/2021/... [3] https://lczero.org/blog/2021/06/the-i... [4] https://github.com/official-stockfish... [5] https://github.com/glinscott/nnue-pyt... [6] https://stockfishchess.org/get-involved/
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 4:42 am Score of Stockfish 14 vs Stockfish 13 : 1 - 0 - 35 [0.514] ... Stockfish 14 playing White: 1 - 0 - 17 [0.528] 18 ... Stockfish 14 playing Black: 0 - 0 - 18 [0.500] 18 ... White vs Black: 1 - 0 - 35 [0.514] 36 Elo difference: 9.7 +/- 18.7, LOS: 84.1 %, DrawRatio: 97.2 % 36 of 1000 games finished.
Admin Admin
Posts : 2608Join date : 2020-11-17Location : Netherlands Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 10:16 am We are not used to that, a new Stockfish that at first glance only looks slightly better.
Admin Admin
Posts : 2608Join date : 2020-11-17Location : Netherlands Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 11:24 am mwyoung likes this post
Like 1
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 5:58 pm Admin wrote: We are not used to that, a new Stockfish that at first glance only looks slightly better. Yes, and some are even losing their minds over it on my channel. Proclaiming Fishtesting infallibility+30 Elo.
My testing has been pretty consistent. Showing little to no progress, and sometimes regression on the Stockfish DEV moving to SF 14. Depending on the time controls and cores used.
What is clear and I think other testings is showing the same. As you move away from faster time controls and 1 thread testing. There is little difference between SF 13 and SF 14.
The latest results.
Score of Stockfish 14 vs Stockfish 13 : 1 - 0 - 65 [0.508]
... Stockfish 14 playing White: 1 - 0 - 32 [0.515] 33
... Stockfish 14 playing Black: 0 - 0 - 33 [0.500] 33
... White vs Black: 1 - 0 - 65 [0.508] 66
Elo difference: 5.3 +/- 10.2, LOS: 84.1 %, DrawRatio: 98.5 %
66 of 1000 games finished.
Admin Admin
Posts : 2608Join date : 2020-11-17Location : Netherlands Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 7:11 pm
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 7:12 pm Admin wrote: CEGT got +40 for SF14.https://cegt.forumieren.com/t1527-testing-stockfish-14-0 I saw that result. Again fast time controls, and 1 thread testing only.
Sf 14 will have 1000 games to show its true form, with 32 threads and 10m+5s time controls. When playing SF 13. The results will be interesting.
Mclane
Posts : 3022Join date : 2020-11-17Age : 57Location : United States of Europe, Germany, Ruhr area Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 8:47 pm I find it boring. Engine x versus x+1 Testing of engines makes only sense if there is a variance of opponents
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 11:35 pm Mclane wrote: I find it boring. Engine x versus x+1 Testing of engines makes only sense if there is a variance of opponents Sorry you find it borning. But many people do not find it boring.
Mclane
Posts : 3022Join date : 2020-11-17Age : 57Location : United States of Europe, Germany, Ruhr area Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 11:50 pm So you do it for the people who want to see it ?!
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 04, 2021 11:59 pm Mclane wrote: So you do it for the people who want to see it ?! This match was a match request from my viewers. Players, time controls, and number of games.
And yes I do match request, along with my other engine tests. Under various conditions, and time controls. From Head to Head matches, and multi-player tournaments.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Mon Jul 05, 2021 4:14 pm Score of Stockfish 14 vs Stockfish 13 : 1 - 0 - 117 [0.504] ... Stockfish 14 playing White: 1 - 0 - 58 [0.508] 59 ... Stockfish 14 playing Black: 0 - 0 - 59 [0.500] 59 ... White vs Black: 1 - 0 - 117 [0.504] 118 Elo difference: 2.9 +/- 5.7, LOS: 84.1 %, DrawRatio: 99.2 % 118 of 1000 games finished.
TheSelfImprover likes this post
Like 1
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Thu Jul 08, 2021 8:42 am Some in the chess community do not like my results.
dsvc saas
The forum is right. Broadcaster Mark Young has received a sum of money from Komodo to deliberately make SF14 look a lot worse than reality.
dsvc saas
This act is to intentionally tarnish the image of SF14 while boosting sales of Komodo.
Score of Stockfish 14 vs Stockfish 13 : 1 - 0 - 260 [0.502]
... Stockfish 14 playing White: 1 - 0 - 130 [0.504] 131
... Stockfish 14 playing Black: 0 - 0 - 130 [0.500] 130
... White vs Black: 1 - 0 - 260 [0.502] 261
Elo difference: 1.3 +/- 2.6, LOS: 84.1 %, DrawRatio: 99.6 %
261 of 1000 games finished.
Admin likes this post
Like 1
Admin Admin
Posts : 2608Join date : 2020-11-17Location : Netherlands Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Thu Jul 08, 2021 8:58 am If you don't like what is reported attack the reporter.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Thu Jul 08, 2021 9:04 am Admin wrote: If you don't like what is reported attack the reporter. dsvc saas
The forum is right. Broadcaster Mark Young has received a sum of money from Komodo to deliberately make SF14 look a lot worse than reality.
I just want to know one thing. Where is my money at Larry Kaufman!
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 11, 2021 1:35 am Score of Stockfish 14 vs Stockfish 13 : 1 - 0 - 406 [0.501] ... Stockfish 14 playing White: 1 - 0 - 203 [0.502] 204 ... Stockfish 14 playing Black: 0 - 0 - 203 [0.500] 203 ... White vs Black: 1 - 0 - 406 [0.501] 407 Elo difference: 0.9 +/- 1.7, LOS: 84.1 %, DrawRatio: 99.8 % 407 of 1000 games finished.
Mclane
Posts : 3022Join date : 2020-11-17Age : 57Location : United States of Europe, Germany, Ruhr area Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Tue Jul 13, 2021 6:31 pm is stockfish 14 really stronger then 13
P>
Ryzen9 3950x
Platz Motor Punkte Lc Lc St St Dr Ko Et Ko St St Dr Bo Sl Wa Ch Xi Se De Se De De St En Pr Pr Za Di De De Co Gr Ma Ch Lo Ge Ts S-B 01 Lc0 0.27.0 Nvidia RTX 2070 [j92-100] 244,0/283 · ·· ·· ·· ·· ====1==== ==0==01== 11===== =1===1=1= ==1==1=== ==1111= 1===1==== 1=1==11 =1==1=1 1===11111= =1=1=1111 1111=== 111=111=1 =111==1 11=1==1 111111=1= 1111111111 =111111 1=1111111 11111=1 1111111 111=111111 1111111 111111111 11=111111 1111111 11111111 11111111 111111111 11111111 1111111 11111111 11111111 1111111 1111111 29350,00 02 Lc0 0.27.0 Nvidia RTX 2070 [j94-130] 242,5/282 ====0==== · ·· ·· ·· ·· ========== 10===== ===1===1= 1=11==11= 1===1== 11=11==== 11==1=0 0===111 1==111=1= 111=11=== 1=1=1=1 1111111=1 1111111 =1==1111 11=11111= 1111111111 11=1111 11111=111 11111111 1111111 1=1111111 11111111 111=11111 111111111 1111111 1111111 1111111 1111111111 1111111 1111111 1111111 1111111 11111111 1111111 29225,00 03 Stockfish_13_win_x64_avx2 [t8] 232,5/281 ==1==10== ========== · ·· ·· ·· ·· ======0= 1===1==== =11=1=11= 1=====1 01==1==1= 11===== 111111=0 11=11=1==1 1=1111=== 111=011 101101011 11=1001 11=1110 111111110 111=1==111 1=111=1 1111=1111 1=1=1=1 1111111 1=111111= 1111111 111111111 =11111111 1111111 1111111 1111111 1111111111 1111111 1111111 1111111 1111111 1111111 1111111 28162,25 04 Stockfish14_x64_avx2 [t8] 212,5/250 00===== 01===== ======1= · ·· ·· ·· ·· ===1=== =1111=1 1===11= 1=1=1== =110==1 11010110 11=01=1 110=111 1111111 111111= 111=1=0 1111111 1011111 11111111 ==111111 =1111=11 1111=11 111111= 1111=11 1111111 1111111 1111111 1111111 1111011 1111111 1111111 1111111 1111111 1111111 1111111 1011111 1111111 25487,25 05 Dragon-64bit-avx2 [t8] 207,5/282 =0===0=0= ===0===0= 0===0==== ===0=== · ·· ·· ·· ·· 11==1===== 1=====1 ======1== 0001=0= ===01==1 1=====11= 1=====011 ===111= 1=011110=1 111==10 1=11001 111111111 110011101 10111=11 111010=1= 11====1 1111111 11111=100 11=11=1 1111111110 111111011 1111110 1011111 0111111 111111111 11011111 1111111 1111111 11111111 1111111 1111111 24125,50 06 Komodo-14.1-64bit-bmi2 [t8, mcts] 203,5/283 ==0==0=== 0=00==00= =00=0=00= =0000=0 00==0===== · ·· ·· ·· ·· =10=0=00 0====0=000 1====== =00=01= =1=====1= =====1=== =0===1= 11=1===11 =1=10=1 ===11001 1==11=1111 =1=111111 111=1=1 =11=1=111 1=1=111 111111= 111=111=1 1111111 111111111 111111111 1111111 1111111 1111=11 111111111 1111111 11111111 1111111 1111111 11111111 11111111 22024,00 07 Ethereal-13.00-pext-avx2 [t8, e13standard] 196,5/250 ==0000= 0===0== 0=====0 0===00= 0=====0 =01=1=11 · ·· ·· ·· ·· ======= 00=====1 ==0=100 =1===01= 11111=1 =====01 =1=1111 11111==1 1=11111 1=1=111 111111= 11=1=11 =1=1111 1=1=111 1111111 11111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 21946,25 08 Komodo-14.1-64bit-bmi2 [t8] 195,0/282 0===0==== 00=00==== 10==0==0= 0=0=0== ======0== 1====1=111 ======= · ·· ·· ·· ·· ====1=1 ===010= 1=1====== ====10=0= =0=0=== 1011101=1= 1=10101 ===0=11 =11111101 11111=110 11=1==0= 11==1=1=1 =00111= 1111011 1=011=110= 11101=1 1=0101110 110111111 1111111 1111110 1111111 1=1110111 11111111 1111101 11111111 1111=111 1111111 1111111 22314,00 09 Stockfish14_x64_avx2 [t8, no NNUE] 194,0/250 0=0==00 00==0=1 00===== =001==0 1110=1= 0====== 11=====0 ====0=0 · ·· ·· ·· ·· ===0=== 0=1=01= 111111= 010011= ===1=11 =1===1=1 1==1=11 1111111 11111111 1111=11 1111110 1101111 1111=11 =1=1111 1111111 11111=1 0111111 1111111 1111111 1111111 11111111 1111111 11111111 1111111 1111111 1111111 1111111 21586,00 10 Stockfish_13_win_x64_avx2 [t8, ohne NNUE] 188,5/250 =0==0=0 1===000 000000=1 00101001 ===10==0 =11=10= ==1=011 ===101= ===1=== · ·· ·· ·· ·· =1=0==1 11=1=11 ==101=1 ==1=1== =111=== 11=0111 1011111 111110=1 ==11111 =1011111 1111111 1111111 11=1101 1101111 1110111 0111111 1111111 1111111 1101111 1111111 1111110 1111111 1111111 1111111 1110111 1111111 21497,50 11 Dragon-64bit-avx2 [t8,mcts] 188,0/282 0===00000= 0==000=0= 00=00=0==0 00=10=0 0=====00= =0=====0= =0===10= 0=0====== 1=0=10= =0=1==0 · ·· ·· ·· ·· 1=00010== =01==00 11=0011== =1====== =1=11== 1111===111 11==11=11 11====1 ==1==1110 111=1== 1==1111 1=1110=01 11=11=1 =11111111 1111111=1 111=111 11111=1 1111111 111111111 111=1=1 11111111 111=111 1111111 1111111 11111111 20180,00 12 Booot6.5 [t8] 183,5/282 =0=0=0000 000=00=== 0=0000=== 001=000 0=====100 =====0=== 00000=0 ====01=1= 000000= 00=0=00 0=11101== · ·· ·· ·· ·· =10==10 ==11110=1 ==11=== ======0 111=1==1= 1=1111=1= =0==1== 111==1111 1=1=11= 01==111= ==1=1=11= 1=1=111 1111111111 1=11111=1= 101=111 11111111 101=11=1 1111=11111 111=111 1111111 11111111 111=111 1111111 1111111 19300,00 13 Slow2.5 64-avx [t8] 183,0/249 0000=== 0=0=0=0 000=100 0000000 ===000= =1===0= =====10 =1=1=== 101100= ==010=0 =10==11 =01==01 · ·· ·· ·· ·· 1=11=11 ==1==1= =1=1=11 1=11==1 111=11= ==1=1== 1111111 1111110 1=1=111= 1111011 11111=1 1111111 1111111 11111111 11111111 11111111 1111111 1111111 1111111 1111111 1111111 1111111 1111111 19332,00 14 Wasp450-x64-modern [t8] 168,0/283 000=000=0 0000000=0 010010100 000000= 0=100001=0 00=0===00 =0=0000 0100010=0= ===0=00 ==0=0== 00=1100== ==00001=0 0=00=00 · ·· ·· ·· ·· ==0==== =0=0==== ==110=0== ====1==11 0==1==1 111==0=100 ==111=11 1111=11 111=1===1 11111=1 1111111=1 1==111=11 111=111 1111111 1111111 111111111 1111111 1=111111 1111111 1111111 11111111 11111111 16239,25 15 Chiron-5-64bit [t8] 163,0/250 =000==0 0000000 00=0110 000=0=1 000==01 =0=01=0 00000==0 0=01010 =0===0=0 =000=== =0====== ==00=== ==0==0= ==1==== · ·· ·· ·· ·· 1=01==0 =11111= 0==1==== 0==1=== 11===11 1111111 11==111 1=1=1=1 1111111 1111111 1111=11 1111111 1111111 1111111 11111=11 1111111 1111111 1111111 1111111 1111111 1111111 16245,25 16 Xiphos-0.6.1-w64-bmi2 [t8] 159,0/250 00=0==0 =0==0000 00=0001 0000000 0=00110 ===00110 0=00000 ===1=00 0==0=00 00=1000 =0=00== ======1 =0=0=00 =1=1==== 0=10==1 · ·· ·· ·· ·· 1==11== 1===1=1 111=010 1111110 1===111 =11==1= 11=1==== 111=111 1111=111 11=1111 1111111 1111111 1111=11 1111111 1111111 1111111 1111111 11=1111 1111111 1111111 15793,50 17 Senpai_20-bmi2-x64 [t8] 150,0/283 000000=0= 00=00000= 000000001 0100000 000000000 0==00=0000 0=0=000 =00000010 0000000 0100000 0000===000 000=0==0= 0=00==0 ==001=1== =00000= 0==00== · ·· ·· ·· ·· =01==101= 11010=0= 0=1==011= =01=0=1 111=111 1100111=1= 1111111 111111111 1=1111110 1111=11 11111111 11=1111 111=11111 11111111 1111111 1111=111 11111111 1111111 1111111 13099,25 18 Deep Rybka 4.1 x64 [t8] 146,5/282 0000000000 0000000000 000=0==000 00000000 001100010 =0=000000 000000= 00000=001 00000000 000001=0 00==00=00 0=0000=0= 000=00= ====0==00 1==0==== 0===0=0 =10==010= · ·· ·· ·· ·· 01=1=00 ==1111=1= 1101=11 10=0111 11===1111 111=11= 111111111 1111=1111 1111111 1111=11 1=11111 1111=1111 1111111 1101111 1111111 1111111 1111111 1111111 12937,00 19 Seer_znver2.0.1 [t8] 140,0/250 =000000 00=0000 0=000=0 ==000000 01000=00 000=0=0 00=0=00 00=0==1= 0000=00 ==00000 00====0 =1==0== ==0=0== 1==0==0 1==0=== 000=101 00101=1= 10=0=11 · ·· ·· ·· ·· 10=01==0 101=1=1 1111111 =1=0101 1111110 1111011 101=111 1111111 1111111 1=111=1 111=111 =11111= 1111111 1111111 111=111 1111111 1111111 13036,50 20 DeepHiarcs14WCSC [t8] 139,0/282 0=0000000 00000=000 0000=0000 =0000=00 000101=0= =00=0=000 =0=0000 00==0=0=0 0000001 =0100000 ==0==0001 000==0000 0000000 000==1=011 00===00 0000001 1=0==100= ==0000=0= 01=10==1 · ·· ·· ·· ·· =1===== 1011=1= 0=00110== ===1=11 1=11=11111 1=11101111 =1111== 1111=11 1=1111= 111111111 11111111 1111111 1111111 1111111 1111111 1111111 12256,75 21 DeepHiarcs14WCSC [t8, agg/hyp/comb off/sel7] 123,5/250 00000=0 00000000 0=0=0=0 0000=00 00====0 0=0=000 0=0=000 =11000= 0010000 0000000 000=0== 0=0=00= 0000001 ==000=00 0000000 0===000 =10=1=0 0010=00 010=0=0 =0===== · ·· ·· ·· ·· ==111=== 00=1=1= 11111=1 111=111= 110==1= 1111=111 111111= 1111111 111111= 1111111 1=11111 1111111 1111111 1111111 1111111 10314,25 22 Sting-sf-28 [t8] 109,5/249 0000000 0000000 0000000 000000= 0000000 000000= 0000000 0000100 0000=00 0000000 0==0000 10==000= 0=0=000= 0000=00 00==000 =00==0= 000=000 01=1000 0000000 0100=0= ==000=== · ·· ·· ·· ·· 0001011 11111=11 1111111 =011111 111111= 1111111 1111111 1111111 1111=11 1111111 1111111 0111111 1111111 1111111 7937,00 23 EngineDeepShredder13UCIx64 [t8] 103,0/282 000=000000 0=0000000 0=000000= 0000=00 00000=011 000=000=0 00000000 0=100=001= =0=0000 00=0010 0=0001=10 ==0=0=00= 0000100 000=0===0 0=0=0=0 00=0==== 0011000=0= 00===0000 =0=1010 1=11001== 11=0=0= 1110100 · ·· ·· ·· ·· 0010=01 010000001 011101010 111010= 1100110 0010111 000101110 1100000 =101=111 0111111 1111101 0111011 11110111 10110,50 24 Prodeo 3.1 79,5/249 0000000 00000000 0000000 0000000 00=00=0 0000000 0000000 00010=0 0000000 0010000 00=00=0 0=0=000 00000=0 00000=0 0000000 000=000 0000000 000=00= 0000001 ===0=00 00000=0 00000=00 1101=10 · ·· ·· ·· ·· ==0=1== 1001111 ==110=1= 10=1011 ==0=1111 ===1110 11=1111 111=0=0 111==== =111110 1111111 1111111 5416,25 25 Prodeo YAT 78,0/282 000000000 000=00000 000000000 0000000 0000000001 000000000 0000000 0=1010001 00000=0 0001000 =00000000 0000000000 0000000 0000000=0 0000000 0000=000 000000000 000000000 0000100 0=00=00000 000=000= 0000000 101111110 ==1=0== · ·· ·· ·· ·· 011010011 0=0=1=1 0000==1 0111111 =0===1=1= 1=1==11 1=11=10 111=010 0011011 11111111 11111111 5377,75 26 Zappa Mexico II x64 [t8] 70,5/282 00=000000 000000000 =00000000 0000000 000000100 000000000 0000000 001000000 1000000 1000000 0000000=0 0=00000=0= 0000000 0==000=00 0000=00 00=0000 0=0000001 0000=0000 010=000 0=00010000 001==0= =100000 100010101 0110000 100101100 · ·· ·· ·· ·· 0101111 11000001 00110010 1100110100 0100011 1101100 10010100 11111011 0111011 0011110 5726,25 27 Diepm 2.10 [t8] 66,0/249 0000000 0000000 0000000 0000000 0000001 0000000 0000000 0000000 0000000 0000000 000=000 010=000 00000000 000=000 0000000 0000000 0000=00 0000000 0000000 =0000== 0000=000 000000= 000101= ==001=0= 1=1=0=0 1010000 · ·· ·· ·· ·· 1=11111 11=01=0 =1=1=0= 10=011= 0101101 ==11=== 1=11111 11111111 1111111 3873,00 28 DeepGandalf7-64 [t2] 64,0/250 00000000 0000000 0000000 0000100 0100000 0000000 0000000 0000001 0000000 0000000 00000=0 00000000 00000000 0000000 0000000 0000000 00000000 0000=00 0000000 0000=00 000000= 0000000 0011001 01=0100 1111==0 00111110 0=00000 · ·· ·· ·· ·· 101=001 1=111=1 11===01 111=100 11==111 11=1010 1111111 1111111 3798,50 29 Delfi5 61,0/250 00000000 0000000 0000000 0000000 1000000 0000=00 0000000 0000000 0000000 0010000 0000000 010=00=0 00000000 0000000 0000000 0000=00 00=0000 0=00000 0=000=0 0=0000= 0000000 0000000 1101000 ==1=0000 1000000 11001101 00=10=1 010=110 · ·· ·· ·· ·· 0=1=00= 11=1010 1=1==10 11=1=10 =1=10=1 1111111 1111111 3868,75 30 Colossus2008b 59,5/281 000000000 0000000000 0000000000 0000000 000000000 000000000 0000000 0=0001000 00000000 0000000 000000000 0000=00000 0000000 000000000 00000=00 0000000 000=00000 0000=0000 000=000 000000000 000000= 0000000 111010001 ===0001 =1===0=0= 0011001011 =0=0=1= 0=000=0 1=0=11= · ·· ·· ·· ·· 10=0=01 0111==0 111=1=1 11==00= 1=11111 1111111 3540,50 31 GreKo1.1 psq 54,5/250 00000000 0000000 0000000 0000000 00100000 0000000 0000000 00000000 0000000 0000001 000=0=0 000=000 0000000 0000000 0000000 0000000 00000000 0000000 =00000= 00000000 0000000 0000=00 0011111 00=0000 0=0==00 1011100 01=100= 00===10 00=0101 01=1=10 · ·· ·· ·· ·· 1==01== 0=011=0 =1=1=11 1111111 1111111 3242,50 32 MadChess.Engine3 49,5/250 0000000 0000000 0000000 0000000 0000000 00000000 0000000 0000010 00000000 0000000 00000000 0000000 0000000 0=000000 0000000 0000000 0000000 0010000 0000000 0000000 0=00000 0000000 =010=000 000=1=1 0=00=01 0010011 1010010 000=011 0=0==01 1000==1 0==10== · ·· ·· ·· ·· 1=101=1 1110010 1111111 1=11111 2650,75 33 Chinito 3.5 43,0/250 00000000 0000000 0000000 0000000 0000000 0000000 0000000 00000000 0000000 0000000 000=000 00000000 0000000 0000000 0000000 0000000 0000=000 0000000 0000000 0000000 0000000 0000000 1000000 000==== 000=101 01101011 ==00=== 00==000 00=0=01 000=0=0 1=100=1 0=010=0 · ·· ·· ·· ·· 11=1101 1111111 11=1111 2046,25 34 Loki3.0.0-x64 [t8] 37,5/250 00000000 0000000 0000000 0000000 00000000 0000000 0000000 0000=000 0000000 0000000 0000000 000=000 0000000 0000000 0000000 00=0000 00000000 0000000 000=000 0000000 0000000 1000000 0000010 =000001 1100100 00000100 0=00000 00=0101 =0=01=0 00==11= =0=0=00 0001101 00=0010 · ·· ·· ·· ·· 1=1=111 1111111 1945,50 35 Gerbil2 14,5/250 0000000 00000000 0000000 0100000 0000000 00000000 0000000 0000000 0000000 0001000 0000000 0000000 0000000 00000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 1000100 0000000 00000000 1000100 00000000 0000000 0000000 0=00000 0000000 0000000 0000000 0=0=000 · ·· ·· ·· ·· 1111111 850,25 36 Tscp181 5,0/250 0000000 0000000 0000000 0000000 0000000 00000000 0000000 0000000 0000000 0000000 00000000 0000000 0000000 00000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 0000000 00001000 0000000 00000000 1100001 0000000 0000000 0000000 0000000 0000000 0=00000 00=0000 0000000 0000000 · ·· ·· ·· ·· 360,75
4755 Partien von 6300 gespielt Beginn des Turniers: 2021.03.12, 16:40:14
Letzte Aktualisierung: 2021.07.13, 18:22:59
Ort/ Land: DESKTOP-D269HUC, Deutschland
Spielstufe: Turnier 40/5
Hardware: AMD Ryzen 9 3950X mit Nvidia Geforce RTX 2070 mit 8 GB Speicher
Betriebssystem: Windows 10 Enterprise Professional (Build 9200)
PGN-Datei: Ryzen9 3950x .pgn Tabelle erstellt mit: Arena 2.0.1 ok some games to go. but it will need to win.
btw good results of ethereal
Mclane
Posts : 3022Join date : 2020-11-17Age : 57Location : United States of Europe, Germany, Ruhr area Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Tue Jul 13, 2021 6:45 pm On my ryzen 5 3600 with 6 cores each it is even worse. Stockfish 13 has 122/148 While stockfish 14 only 116,5 from 148. Maybe the release candidate is not the big progress. Or the nets integrated is not helping much.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Tue Jul 13, 2021 11:05 pm Thanks Mclane for your results.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Tue Jul 13, 2021 11:21 pm I believe it is the flawed testing of fishtest that is to blame for this debacle!
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Fri Jul 16, 2021 9:57 pm Score of Stockfish 14 vs Stockfish 13 : 2 - 0 - 710 [0.501] ... Stockfish 14 playing White: 1 - 0 - 355 [0.501] 356 ... Stockfish 14 playing Black: 1 - 0 - 355 [0.501] 356 ... White vs Black: 1 - 1 - 710 [0.500] 712 Elo difference: 1.0 +/- 1.3, LOS: 92.1 %, DrawRatio: 99.7 % 712 of 1000 games finished.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Fri Jul 16, 2021 10:12 pm The stream and match have now been playing for 332 hours nonstop. No computer issues to report, and my new 4K internet connecting has been stable without any loss of connection for over 332 hours.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Sun Jul 18, 2021 7:44 pm Score of Stockfish 14 vs Stockfish 13 : 2 - 0 - 812 [0.501] ... Stockfish 14 playing White: 1 - 0 - 406 [0.501] 407 ... Stockfish 14 playing Black: 1 - 0 - 406 [0.501] 407 ... White vs Black: 1 - 1 - 812 [0.500] 814 Elo difference: 0.9 +/- 1.2, LOS: 92.1 %, DrawRatio: 99.8 % 814 of 1000 games finished.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Thu Jul 22, 2021 8:53 pm End of Match. Score of Stockfish 14 vs Stockfish 13 : 3 - 0 - 997 [0.501] ... Stockfish 14 playing White: 2 - 0 - 498 [0.502] 500 ... Stockfish 14 playing Black: 1 - 0 - 499 [0.501] 500 ... White vs Black: 2 - 1 - 997 [0.500] 1000 Elo difference: 1.0 +/- 1.2, LOS: 95.8 %, DrawRatio: 99.7 % 1000 of 1000 games finished.
mwyoung
Posts : 880Join date : 2020-11-25Location : USA Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games) Thu Jul 22, 2021 9:52 pm Thank you all for watching. This test had 30,369 views
Sponsored content
Subject: Re: Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games)
Stockfish 14 vs. Stockfish 13 (32 Threads) (TC=10m+5s) (1000 Games)