Wasp and a question

[fern]

Please do not read this comment as a critic of any kind to wasp and less to John.
It is just a candid, real question.
what Wasp offer that is different to what other engines of very high level offer?
I mean, what a thing offer that could be detected by a player like me that is curretly defeted by any plus 2300 or 2400 Elo engine?
Which is the difference between the defeat I suffer with Stokfish, Amyan or genius3 with being mauled by Wasp?

I ask this because there are over there some engines that DO make a difference, clearly. By example, CSTAL. You can feel it. Or any Schroeder engine.
What wasp pretend beyond the strenght?
I have just played a game and of course I was defeated, but I did not feel anything special in that defeat than what I feel when I am kicked in the ass by any other engine.
Again, no my intention to say anything derogatory. Just curiosity.
And of course it could be that this kind of enquiere is out of the question these days.
sorry sorry sorry

Fern

[jstanback]

Hi Fern,

I don't make any claims for why Wasp should be preferred over any other engine. Wasp uses similar search and evaluation techniques as other engines. But it is an original engine, so all the routines are coded and tuned differently from any other engine which may give it a slightly different style of play.

John

[fern]

I am sure you did great job. I am a louzy player and so any and every mountain seems to me same in height.
I will continue playing wasp to see IF I can detect the difference. If I do not, its my fault, not yours.
In music you need a very refined ear to understand Bach. Same with chess. it is necesary a really strong player to get the nuances. I am just around 1900-2000, so you can understand that i can perhaps detect tactical differences of strenght, but not more soubtil things
a hug

Fern

[Frank Quisinsky]

Hi Fernando,

indeed, Wasp do the same in games stronger humans do.

Opening: Very own ideas, aggressive and a lot of positionally understandings. Won more games fast vs. similar engines in strength as to lost. Please replay the fast won games from my beta test on Sunday with a bottle of wine on your big chess board.

Middlegame: Engine hold the style and search the way for open chess.

Late middlegame: Here the same problems maybe best human players have ... (before endgame started) if we compare it with very strong engines . Wasp lost the great strength from the beginning up to the middlegame. Sure here John can make Wasp 100-150 Elo stronger ... on the other hand ... nothing to do (I like it).

Endgame: Again, very strong!

So, if I looking in Wasp games, I am thinking a super GM is playing. Perfect simulation to humans.

John hold all the years the style Zarkov have for many years. Thats what I like from him.

Manfred Mailer (German computer chess friend) is working many years on Excel tables to BT-Test. At this time I got from John the first WB engines and we are looking on BT results.

I try to write a little story in my bad English ...

So at first we are disappointed because Zarkov can't solve many problems other engines solved. We are looking more in detail. Sensation ... Zarkov solved all the complicated problems but need clearly more time as the others. And after all patience we saw that most of the problems solved by Zarkov (more problems solved as all other engines in that test). Of course we gave the others more time too.

The search of Zarkov works better as in best available commercial chess programs at this time. So what John do is right and his ideas in programming are great.

And my love to Zarkov was born again!
DOS Zarkov 2.51 I like a lot too ... and all the games vs. Rexchess 2.1 / 2.3 we are playing on UCI PCs ... for around 32 years or so.

What John do here in chess programming is very special in times today. John is a chess programmer from the first hour and goes all the time his own way. Can be see if you know many of his versions in detail.

Never Zarkov / Wasp lost his own face with a complete new style. Elo isn't all ... Wasp is fantastic for self playing, my absolutly first class number 1 program for self playing.

Special / particular and an event for computer chess today!
I hope that many others can see that and I am sure we will have many years many fun with programs by John Stanback.

Really a legend in chess programming!
I am very happy that John is working again for the computer chess community, means is working on his new engines Wasp.

Best
Frank

For John ...
Sorry John, loud thinking only to a late hour.
Time for sleeping!

Fact is:
Zarkov / Wasp is 100% computer chess humans like.

[Frank Quisinsky]

UCI PCs ... bullshit

I try to talk from UNI PCs.
To expensive fast PCs at this time.
So we are searching other ways to test engines.

UCI PCs ... to late for me.
Sorry!

UNI PCs is right.

Best
Frank

[BrendanJNorman]

Please do not read this comment as a critic of any kind to wasp and less to John.
It is just a candid, real question.
what Wasp offer that is different to what other engines of very high level offer?
I mean, what a thing offer that could be detected by a player like me that is curretly defeted by any plus 2300 or 2400 Elo engine?
Which is the difference between the defeat I suffer with Stokfish, Amyan or genius3 with being mauled by Wasp?

I ask this because there are over there some engines that DO make a difference, clearly. By example, CSTAL. You can feel it. Or any Schroeder engine.
What wasp pretend beyond the strenght?
I have just played a game and of course I was defeated, but I did not feel anything special in that defeat than what I feel when I am kicked in the ass by any other engine.
Again, no my intention to say anything derogatory. Just curiosity.
And of course it could be that this kind of enquiere is out of the question these days.
sorry sorry sorry

Fern

Don't be disheartened Fern.

Look at it this way.

For pure chess pleasure (not truth seeking, scientific exploration), use Wasp as your primary ANALYSIS engine for strategical positions/games - he will suggest moves much like the legend of strategy....Vladimir Kramnik.

Sharp when he needs to be, but strategic virtuoso most of the time.

Gaviota is also excellent for strategic positions.

For tactical positions, use the Rodent III Tal personality (which we are working on, but is amazing in strength and style, and will come in December)

In fact, in Wasp 2.6's gauntlet last night, Rodent III Tal got 2/2 against him!

[pgn][Event "Wasp"]
[Site "BRENDANNORMD8A2"]
[Date "2017.11.23"]
[Round "1"]
[White "Wasp 2.60"]
[Black "Rodent III Tal (Pawel)"]
[Result "0-1"]
[ECO "A70"]
[WhiteElo "2200"]
[BlackElo "2200"]
[PlyCount "95"]
[EventDate "2017.??.??"]

1. d4 Nf6 2. c4 e6 3. Nf3 c5 4. d5 exd5 5. cxd5 g6 6. Nc3 Bg7 7. e4 d6 8. Bd3
O-O 9. O-O a6 10. a4 Bg4 11. Be2 Nbd7 12. Be3 Qa5 13. Qc2 Rac8 14. h3 Bxf3 15.
Bxf3 Rfe8 16. b3 c4 17. b4 Qxb4 18. Rab1 Qa5 19. Rxb7 Nc5 20. Bxc5 Qxc5 21.
Rfb1 h5 22. R7b6 Nh7 23. Rxa6 Ng5 24. Nb5 Be5 25. Ra7 c3 26. a5 f5 27. exf5 Rf8
28. Bd1 Rxf5 29. Re7 Nxh3+ 30. gxh3 Rg5+ 31. Kf1 Qc4+ 32. Be2 Qh4 33. Bf3 Rf8
34. Qe4 c2 35. Rc1 Rg1+ 36. Kxg1 Qg5+ 37. Bg4 Qxc1+ 38. Kg2 Qg5 39. Rxe5 dxe5
40. Qxc2 hxg4 41. h4 Qf6 42. d6 Rf7 43. Qe2 Qxh4 44. a6 Qh3+ 45. Kg1 g3 46. f3
Rh7 47. Qg2 Qh6 48. Qxg3 0-1[/pgn]

On the other hand, witness Wasp's beautiful dismantling of Chessmaster 8000 (King 3.12) in the following game.

Wasp's 11.b6! pawn sac is a novelty and has never been played before (previously tried has been 11.Re1). It ties black up on the queenside, giving white time to switch his attack to the kingside. Maybe someone might like to try this in an OTB game?

Very Kramnik-like KID demolition.

[pgn][Event "Wasp"]
[Site "BRENDANNORMD8A2"]
[Date "2017.11.23"]
[Round "1"]
[White "Wasp 2.60"]
[Black "ChessMaster 8000"]
[Result "1-0"]
[ECO "E97"]
[WhiteElo "2200"]
[BlackElo "2350"]
[PlyCount "78"]
[EventDate "2017.??.??"]

1. d4 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. Nf3 O-O 6. Be2 e5 7. O-O Nc6 8. d5
Ne7 9. b4 a5 10. b5 Nh5 11. b6 cxb6 12. Rb1 Bd7 13. g3 Nf6 14. Qc2 h6 15. Be3
Nc8 16. Kg2 Ng4 17. Bd2 Qe7 18. h3 Nf6 19. Rh1 Rd8 20. Nh2 Kh7 21. Qc1 Ne8 22.
Ng4 f5 23. exf5 gxf5 24. Bxh6 fxg4 25. hxg4 Kg8 26. f3 b5 27. Bd3 bxc4 28. Bxc4
Rb8 29. Bd3 Nf6 30. Bxg7 Qxg7 31. g5 e4 32. gxf6 exf3+ 33. Kf2 Qxf6 34. Bh7+
Kf7 35. Ne4 Qd4+ 36. Kxf3 Rf8 37. Kg2 Ke8 38. Qc7 Ke7 39. Bg6 Rf7 1-0[/pgn]

And as for opponents.

If looking for positional opponents.

Amyan 1.72

Zarkov (Frank has early versions as winboard on his site)

Hamsters

RomiChess

Rodent III Karpov or Petrosian (coming December)

Or for attacking opponents.

Rodent III Tal, Shirov, Henny, Topalov etc (December also)

Baron 1.6.1

Comet A90

Dimitri 3.81 (with aggressive setting) - not well known, but amazing!

LG Evolution

ProDeo 1.2

Rebel Century Tal

Natwarlal 0.14 (not well known, but AGGRESSIVE engine from India)

Thinker 4.7 or 5.1e Passive

WChess 1.06

etc.

I think it's important to separate engines into opponents and analysis. Anything above 2900 gets relegated to analysis, while 2800 I can generally manage (although I usually lose).

I have held games against Frenzee 3.5.19 (excellent human-like with balanced style IMO...maybe similar to Carlsen, but with more logical moves haha), Thinker 5.5 (years ago when i HAD it), and others and beaten various u2400 engines like RomiChess and others.

All without resorting to Anti-Computer tricks.

So give some of those a try, and imo, keep Wasp for analyzing the resulting games.

Throw any engines over 3150 in the trash bin.

[fern]

Happy to do that PROVIDED you offer 32 bits products. I have a couple of 64 bits comps, but the one I generally use is an old but fast 32 bits thing...

Indulge me..

Fern

[Michael Sherwin]

Happy to do that PROVIDED you offer 32 bits products. I have a couple of 64 bits comps, but the one I generally use is an old but fast 32 bits thing...

Indulge me..

Fern

Hi Fern, RomiChess has compiles for 64 bit and 32 bit.

http://kirill-kryukov.com/chess/discuss ... p?id=40457

[fern]

Thanks Michel

[fern]

could not install it in Fritzes, but i did in arena.