Chess2u
Would you like to react to this message? Create an account in a few clicks or log in to continue.

Chess2uLog in

Proteus Chess SF v004

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:


Then thanks Chessman for publishing SF-PB for the whole chess community.

Go on altervista, open your eyes and look who introduced SF-PB and who compiled the engines.

AlexChess wrote:

Your are right, your it's not even humor more likely grudge against the World. Be constructive, not destructive.

Grudge against the World? LOL.. Oh boy, everything that i said has to do ONLY with you. Did you forget take your pills again? You are not even understand what are you talking about, DK effect?
AlexChess wrote:

If you think that being a simple moderator  without other skills is interesting, go on.

How do you came to this conclusion and what do you even know about my abilities? Asylum waiting for you, go there immediately. lol
AlexChess wrote:

And YOU are a baby, I'm 57.

clap Typical answer from the baby..

AlexChess wrote:

...luckily not Skynet from Third Dimension or Chessfan1 from Saturn  lol  lol  lol  lol
Are you ashamed of your real name or about your behavior permanently blocking members you offend, to be sure that they cannot reply??

Why i'm not telling my real name? Are you from f*cking IRS?  lol Ok i'll give you a tip, why do you you think other members using nicknames and not real names? Use your little brain and I hope you understand.
P.s. I'm not ashamed about my behavior or actions, the whole point of the ban is to get rid of idiots like you, and make the forum calm and user-friendly. Unfortunately, you will never understand this.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:
@ChessFan1

You are offending even @Oliprog :

Read my text again, then again, then again one more time.. Where exactly did you see that I offended Oliprog? What I wrote is 100% true, and people (including Oliprog) know this.. Or you really think that SF team (with 60.000 games, for each patch) or Stefan Pohl (with more than 7000 games, for engine) etc are stupid and you are a smart guy? I don't think so..
AlexChess wrote:


"Yes, i have see all those tests, but there is a huge difference between "Tests for fun" (like Oliprog's) and professional tests. Tests for fun with small number of games can't be taken as serious, due to huge number of error bars. We can't claim that, engine A is stronger than engine B after 200, 300 or 600 games (where each engine will play 100, 150, 300 games - switching colors)"


What kind of bad guy are you, SKYNET?

Al Capone.. twisted

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
[quote="ChessFan1"]
AlexChess wrote:
@ChessFan1


What kind of bad guy are you, SKYNET?

Al Capone.. twisted[/quote

Please return to OCF, go to play your ridiculous role thinking that you are important. Don't bore anymore people in other forums. I will answer again to you only when you reveal you real name, I can even verify if you have minimal programming skills due to your silly nick.

Probably you are only a or a childish Terminator's fanboy.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
He spent the match from ten batches between the Proteus Chess SF v 004 and Kayra 1.1 engines.
NNUE did not connect - the engines played according to the classic scheme. The victory with the score of 3: 2 won Proteus Chess and was 5 draws.
The battle of the engines was carried out in the Chess - Tactics and Strategy v 2021.2.3 program.
AlexChess we are waiting for version 005. Good luck !

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Aleksey52 wrote:
He spent the match from ten batches between the Proteus Chess SF v  004 and Kayra 1.1 engines.
NNUE did not connect - the engines played according to the classic scheme. The victory with the score of 3: 2 won Proteus Chess and was 5 draws.
The battle of the engines was carried out in the Chess - Tactics and Strategy v 2021.2.3 program.
AlexChess we are waiting for version 005. Good luck !

Thank you, YES, also on mytest it plays very well without NNUE

You must import this NNUE for Proteus included in the folder: https://tests.stockfishchess.org/api/nn/nn-6877cd24400e.nnue

Chessfan1-Skynet test demostrates that after 4600+ games is has only 1 point less than Stockfish. Wonderful! lol

Both work fine on Chess for Android 6.2.1

I'm already working on v005, thank you for your support, continue testing and enjoy it!

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:
Aleksey52 wrote:
He spent the match from ten batches between the Proteus Chess SF v  004 and Kayra 1.1 engines.
NNUE did not connect - the engines played according to the classic scheme. The victory with the score of 3: 2 won Proteus Chess and was 5 draws.
The battle of the engines was carried out in the Chess - Tactics and Strategy v 2021.2.3 program.
AlexChess we are waiting for version 005. Good luck !

Thank you, YES, also on mytest it plays very well without NNUE

You must import this NNUE for Proteus included in the folder: https://tests.stockfishchess.org/api/nn/nn-6877cd24400e.nnue

Chessfan1-Skynet test demostrates that after 4000+ games is has only 1% less than Stockfish. Wonderful!  lol

Both work fine on Chess for Android 6.2.1

I'm already working on v005, thank you for your support, continue testing and enjoy it!

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
deeds wrote:
Smarter than sf team :
- nnue authors
- alphazero team
....


Smarter? I would not say that. I think both teams are brilliant.

The Japanese NNUE authors used ideas from Stockfish for their Shogi engine. They also adopted Stockfish's search. This was all done prior to them gifting NNUE to the Stockfish community in appreciation. It would seem logical to me that the Shogi authors original search was deemed inferior after the Stockfish search was implemented into their Shogi engine.

deeds wrote:
The A0 team has demonstrated that by self-training from scratch the chess engine was much stronger in few hours/days than sf team while years.


The AlphaZero team had the resources to employ 5,000 state-of-the-art first generation Tensor Processing Units as well as 64 second generation TPUs for training.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
deeds wrote:
Katabatic wrote:
It's a specialized tool specifically for endgame situations. Of course a tablebase specifically meant for resolving endgames is going to be better than an engine that calculates through the lines. It's supposed to be.

Tablebases are not only used at the end of games ! Several private chess engines fully analyze (by Brute Force or not) openings to store the best moves. From what I've seen, we're already at +80 elo from SF14.1, this system is 10 years ahead of the SF team.

Again, completely different kinds of tools. TBs are stored transpositions. It takes known positions and resolves them as optimally as current chess theory can.

It doesn't do the same thing as a chess engine. Take your TB and try to play entire matches (1000+ games) with it against other chess engines. It won't work, because it's not a chess engine.

Trying to argue that you're "+80 from SF14.1" is pointless. You're adding your tool ON TOP of SF14.1. In other words, SF14.1 + TB is 80 elo better than SF14.1 without any kind of tablebase. Take just your TB vs SF14.1 and see how well you do. Unless your TB is like 90 zetabytes large you'll run out of positions and then lose because it's not an engine.

SF developers are working on the engine. An engine, by definition, can't look at a stored position and say, "this is how you play it." An engine is the program that calculates optimal next move scenarios based on any given chess position, not specifically referencing known positions.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Katabatic wrote:
Again, completely different kinds of tools. TBs are stored transpositions. It takes known positions and resolves them as optimally as current chess theory can.

So it's the same idea as the system used by private engines : to take known positions and to resolve them as optimally as current chess theory can.

Katabatic wrote:
Take your TB and try to play entire matches (1000+ games) with it against other chess engines.

it's even easier and faster because there is even a tool that allows you to extract games from the data recorded by this system :
Proteus Chess SF v004 - Page 2 OKbCQmN

Katabatic wrote:
Trying to argue that you're "+80 from SF14.1" is pointless. You're adding your tool ON TOP of SF14.1. In other words, SF14.1 + TB is 80 elo better than SF14.1 without any kind of tablebase. Take just your TB vs SF14.1 and see how well you do.

Arguing that private engines are "+80 from SF14.1" was successful weeks ago :
Proteus Chess SF v004 - Page 2 Omci0Dl

Katabatic wrote:
Unless your TB is like 90 zetabytes large you'll run out of positions and then lose because it's not an engine.

No need for 90 zetabytes because the private engine system intelligently stores :
Proteus Chess SF v004 - Page 2 Bj4UC4V

Katabatic wrote:
An engine, by definition, can't look at a stored position and say, "this is how you play it."

It's been years that engines can play the moves of bin books or tablebases or experience bin/exp files without losing time to analyse them. There is no interest with engines that can't learn.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Proteus Chess SF v004 - Page 2 Proteu10Proteus Chess SF v004 - Page 2 Proteu10Proteus Chess SF v004 - Page 2 Proteu10Proteus Chess SF v004 - Page 2 Proteu10


Dear Chess2U friends,

I've just compiled the first ProteusChessSFv005dev build based on latest Stockfish code, but it will take a lot of time to release it, because I'm planning a lot of changes. Please send me your tournament games with ProteusChessSFv004 (especially lost games, studying theses positions is the better way to improve it smile)

Kind regards, Alessandro Morales

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:

Chessfan1-Skynet test demostrates that after 4600+ games is has only 1 point less than Stockfish.

LOL...Once again, you yourself proved that your knowledge about programming and testing engines is ZERO..clap  You can't even read simple results correctly.lol lol lol  In your place i wouldn't speak at all.
total 4016 games (with bigger numbers of games result for your engine would be much worse), 3202 - 3198 = 4 elo, 1 ERROR for each engine..


Code:

# PLAYER                     :  RATING  ERROR  PLAYED  (%)    W    D    L  D(%)  CFS(%)
1 SF-PB-250222_x64_AVX2      :    3202      1    4016  50.5  153  3752  111  93.4    100
2 ProteusChessSFv004-avx2    :    3198      1    4016  49.5  111  3752  153  93.4    ---

This means that your engine is worse by 4 elo - with confidence of superiority 100%. Simple explanation for you, dumb ass. Your engine is 100% sucks, with performance -4 elo..
Keep playing with random changes in evaluation, the only thing you can do..

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:

Please return to OCF, go to play your ridiculous role thinking that you are important. Don't bore anymore people in other forums. I will answer again to you only when you reveal you real name, I can even verify if you have minimal programming skills due to your silly nick.

Probably you are only a or a childish Terminator's fanboy.

Look who's talking.. Once again, (note that no one forced you to say this nonsense), you prove how stupid you are, there is nothing more to be expected from you. You either like to pretend to be a victim or a moron, you are pathetic. I feel sorry for the people that involved with you, but don't worry soon they will realize who you really are.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:

Please send me your tournament games with ProteusChessSFv004 (especially lost games, studying theses positions is the better way to improve it smile)

You got to be kidding, are you looking for fools that will waste their time on another nonsense? Go hide in your hole, mole.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:
Proteus Chess SF v004 - Page 2 Proteu10[Proteus Chess SF v004 - Page 2 Proteu10


Dear Chess2U friends,

I've just compiled the first ProteusChessSFv005dev build based on latest Stockfish code, but it will take a lot of time to release it, because I'm planning a lot of changes. Please send me your tournament games with ProteusChessSFv004 (especially lost games, studying theses positions is the better way to improve it smile)

Kind regards, Alessandro Morales


Dear im48,

sorry for the repeated image, but my intention was to publish 4 different screenshots to show the SS3, SSE41, AVX2 and BMI2 build speed
smile

Kind regards, Alex

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Katabatic wrote:
An engine, by definition, can't look at a stored position and say, "this is how you play it."

It's been years that engines can play the moves of bin books or tablebases or experience bin/exp files without losing time to analyse them. There is no interest with engines that can't learn.[/quote]

OK for tablebases (I'd like if all engines could use online 7 men TB even during blitz games) but for for the openings I prefer that each engine finds his way, sometimes also discovering interesting opening novelties. On next tournaments I'll set opening books to 3 full moves only to randomize a little the games smile

Best regards, Alex

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Katabatic wrote:
An engine, by definition, can't look at a stored position and say, "this is how you play it."
It's been years that engines can play the moves of bin books or tablebases or experience bin/exp files without losing time to analyse them. There is no interest with engines that can't learn.


I don't like also customized openings books for each engine up to 50 moves. redface

Best regards, Alex

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Thank you to all participating to this "brainstorming", I'm very interested to opinions different from mine to improve ProteusChessSF smile

Ps: I have added to zoe the troll-spammer-hater infesting this topic, so I will not read anymore his silly comments . bigsmile

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:
Thank you to all participating to this "brainstorming", I'm very interested to opinions different from mine to improve ProteusChessSF smile

Ps: I have added to zoe the troll-spammer-hater infesting this topic, so I will not read anymore his silly comments . bigsmile


...Zoe instead of foe is my personal ZOO. A collection of boring members than do not deserve my attention lol

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Katabatic wrote:
An engine, by definition, can't look at a stored position and say, "this is how you play it."

It's been years that engines can play the moves of bin books or tablebases or experience bin/exp files without losing time to analyse them. There is no interest with engines that can't learn.[/quote]

If experience data could be accessible online and shared between all running instances of the same engine it would be wonderful, but if you use them locally you cannot understand the real strength of an engine...

Regards, Alex

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
AlexChess wrote:
AlexChess wrote:
Thank you to all participating to this "brainstorming", I'm very interested to opinions different from mine to improve ProteusChessSF smile

Ps: I have added to zoe the troll-spammer-hater infesting this topic, so I will not read anymore his silly comments . bigsmile


...Zoe instead of foe is my personal ZOO. A collection of boring members than do not deserve my attention lol

After embarrassing yourself repeatedly, are you still talking?? Damn, you have not a single drop of self respect..

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
He held a match from ten batches between the Proteus Chess SF v 004 engines (ARM64) and StockFish 13.03.2022 (ARM64). NNUE did not connect, the engines played according to the classic scheme without debut books, each was given for 5 seconds.
Victory with a score of 1: 0 won Stockfish 13.03.2022 Winning the ninth party, the rest of the parties ended in a draw.
For the match, the application of Chess – tactics and strategy v 2021.2.3 was used.
Record the ninth party:

[Event "pbchess match"]
[Site "pbchess"]
[Date "2022.03.17"]
[Round "-"]
[White "StockFish 13-03-2022 arm64"]
[WhiteId "11"]
[WhiteControls "MT||5000"]
[Black "Proteus Chess SFv004 arm64"]
[BlackId "9"]
[BlackControls "MT||5000"]
[Result "1-0"]
[Flags "0"]

1.e4 e5 2.Nf3 Nc6 3.Bb5 Nf6 4.O-O Nxe4 5.Re1 Nd6 6.Nxe5 Nxe5 7.Rxe5+ Be7 8.Bf1 O-O 9.d4 Bf6 10.Re1 Nf5 11.c3 d5 12.Nd2 Nh4 13.g3 Ng6 14.h4 Bf5 15.h5 Ne7 16.g4 Bc8 17.Qf3 h6 18.Qg3 Be6 19.Nf3 Qd7 20.Bf4 Rae8 21.Bh3 Qb5 22.Rab1 c5 23.dxc5 Qxc5 24.Bd6 Qb6 25.Re3 Kh8 26.Rbe1 Rg8 27.g5 Bxh3 28.Bxe7 Rxe7 29.Rxe7 Be6 30.R1xe6 fxe6 31.Ne5 Bxe5 32.Qxe5 Qxb2 33.Kg2 Qd2 34.g6 d4 35.cxd4 Qxa2 36.Rxb7 a5 37.Ra7 Qc4 38.Rxa5 Qc6+ 39.Kh2 Qf3 40.Qe3 Qg4 41.f3 Qh4+ 42.Kg2 Rb8 43.Qc3 Qf4 44.Rc5 Rf8 45.Rc8 Kg8 46.Qb3 Qg5+ 47.Kf1 Qf5 48.Rxf8+ Kxf8 49.Qb8+ Ke7 50.Qc7+ Ke8 51.Kf2 Qf6 52.Qc8+ Ke7 53.Qc5+ Ke8 54.Kg3 Kd7 55.Kg4 Kd8 56.Qd6+ Kc8 57.Qf4 Qe7 58.Qe5 Qd7 59.Kf4 Kb7 60.Ke4 Qc6+ 61.d5 Qc4+ 62.Ke3 Qc5+ 63.Qd4 Qa3+ 64.Kf2 Qe7 65.dxe6 Ka6 66.Qd5 Kb6 67.Kg3 Qc7+ 68.Kg4 Qc8 69.Qd6+ Ka5 70.f4 Qc4 71.Qe5+ Kb4 72.e7 Qc8+ 73.Kg3 Qe8 74.Qxg7 Qa8 75.Qd4+ Kb3 76.Qe3+ Ka2 77.e8/Q Qxe8 78.Qxe8 Kb2 79.Qe5+ Ka3 80.g7 Ka2 81.g8/Q+ Ka3 82.Qc3+ Ka4 83.Qgc4# $500

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Katabatic wrote:
Trying to argue that you're "+80 from SF14.1" is pointless.




ENJOY !

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
deeds wrote:

ENJOY !

You still are missing the point. Adding things on top of SF doesn't prove that the creators of those things are "smarter than stockfish." Like I said, it's like saying the person who added airbags to the car is smarter than the engineers who made the engine for it, because the engine doesn't have any tools that protect you from a crash.

It's well known that adding a tablebase can improve Stockfish's play. That's why SF includes TB options in their native releases.

Also the test you used does not prove anything. Proof is done through tens of thousands of games across a variety of hardware, not single party test matches.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
For whose in denial of reality :

https://www.chess2u.com/t18033p25-proteus-chess-sf-v004#114181

An engine that learns it's smarter than an engine that searches in vain.

A book that improves the WDL ratio it's smarter than tens of thousands games to reduce the error margin.

Tablebases that solve a fraction of chess it's smarter than engines that analyze the same positions over and over again.

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
deeds wrote:
For whose in denial of reality :

https://www.chess2u.com/t18033p25-proteus-chess-sf-v004#114181

An engine that learns it's smarter than an engine that searches in vain.

A book that improves the WDL ratio it's smarter than tens of thousands games to reduce the error margin.

Tablebases that solve a fraction of chess it's smarter than engines that analyze the same positions over and over again.



They play ten thousand games to reduce the error margin on ELO calculation, not to improve the WDL ratio.

But I prefer to play a fraction of games at a decent depth using all 8 CPUs instead of playing 8 games using only 1 CPU simply to say "I have played 4000 games (full of blunders) . I'm very professional" lol

descriptionProteus Chess SF v004 - Page 2 EmptyRe: Proteus Chess SF v004

more_horiz
Permissions in this forum:
You cannot reply to topics in this forum