It was established we have to wait for Tomasz's specification. All programs will use exactly the same tables. Hello all! I have prepared a test version of...
Tak jak Tomek je opublikuje. To nawet nie chodzi tylko o tablice. Tomek powiedział, że poda specyfikacje. Więc spokojnie. czekamy na Tomka tak jak to było...
Sidorowicz Grzegorz napisał: SG> It was established we have to wait for Tomasz's specification. All programs will SG> use exactly the same tables. Yes, I...
Piotr Cichy
pic@...
Dec 1, 2006 9:53 am
6380
No problem. We could use different tables. The only requirement is that we all have to use the same tables. My personal opinion is that once we do something,...
Tomasz Michniewski
tmichniewski@...
Dec 1, 2006 1:32 pm
6381
Basically I am going to repeat the results of previous discussion about ultra-simple evaluation function, originated by Mariusz ("przezroczystek"). we agreed...
... At least in part, it is because there is no universally correct weight for any single evaluation term. It depends on the rest of the evaluation function....
Hi Pawel! ... I don't think this is necessary. On modern hardware, it is possible to find such mates by search alone. I just made a test with my new program...
I wanted to say (or I thought I had said) that I don't say that my proposal would be the best. If we decide to use Gerbil's or Piotr's tables - that's fine. ...
Tomasz Michniewski
tmichniewski@...
Dec 1, 2006 7:39 pm
6386
OK. It doesn't matter for me what kind of tables or other modifications we will be using. I just don't want wait too long for final decision :-) ... From:...
Tomasz Michniewski wrote: TM> My personal opinion is that once we do something, it should be TM> preceeded by some analysis. And since everybody agrees that ...
Piotr Cichy
pic@...
Dec 2, 2006 1:18 am
6388
Tomasz Michniewski napisał: TM> However since some time I wanted to play with equations like: TM> B + N = R + 2P TM> or TM> Q + P = 2R TM> and several others...
Piotr Cichy
pic@...
Dec 2, 2006 9:39 am
6389
Oh Gosh, what is wf, bf and why the subtraction gives bonus/penalty for bad trades? I know that you tend to create as short source code as possible, but it...
Tomasz Michniewski
tmichniewski@...
Dec 2, 2006 7:01 pm
6390
... Could you check mating with two bishops and then with a bishop and a knight. The latter might be the most difficult. ... OK, but if we have two versions of...
Tomasz Michniewski
tmichniewski@...
Dec 2, 2006 7:03 pm
6391
... I would like to have piece values and piece/square tables designed this way that they would be the simple in both: light and heavy evaluation function. I...
Tomasz Michniewski
tmichniewski@...
Dec 2, 2006 7:55 pm
6392
Tomasz Michniewski wrote: TM> Oh Gosh, what is wf, bf These are temporary variables, computed 2 lines earlier, look at the whole function again: int Material()...
Piotr Cichy
pic@...
Dec 2, 2006 10:26 pm
6393
Hallo, ... You meant 10th rule (so the second 3rd), didn't you? I've just read them. I don't understand rules 2 and 7 (I mean I don't understand the idea...
Tomasz Michniewski
tmichniewski@...
Dec 3, 2006 3:28 pm
6394
Hallo, There are two parts of this email, the first is about simple piece values, the second is about piece/square tables. PIECE VALUES Setting piece values I...
Tomasz Michniewski
tmichniewski@...
Dec 4, 2006 8:16 am
6395
I would propose slightly different piece/square table for Rook. It gives additional bonus for an attacking position on the 8th rank. Moreover, the...
Hello Tomasz, it is interesting to know your point of view, because you are probably the best chessplayer among us. TM> You meant 10th rule (so the second...
Piotr Cichy
pic@...
Dec 4, 2006 12:27 pm
6397
Hello Paweł, ... With this I'm not sure. Sometimes it may simply waste time and encourage the opponents king to approach the centre. I'd rather left the...
Tomasz Michniewski
tmichniewski@...
Dec 4, 2006 2:41 pm
6398
Hi Tomasz, Bonus for a Rook on the 8th rank is rather defensive in nature, as it encurages the program to avoid positions where the opponent has the ...
pewnie to wyjdzie przy okazji turnieju testowego, ale pytanie chcialbym zadac juz teraz: w jakie glupie na pierwszy rzut oka pozycje pakuja sie Wasze (lub...
hopelesschess napisał: h> 2) g3 grane zamiast h3 przy krotkiej roszadzie Przyczyną jest prawdopodobnie to, że nagradzam pionka za wsparcie od innego pionka....
Piotr Cichy
pic@...
Dec 4, 2006 4:44 pm
6401
... I don't understand. For me bonus for 1st rank is defensive, while bonus for 8th rank is something different. I to be honest I would move rook to 8th rank...
Tomasz Michniewski
tmichniewski@...
Dec 4, 2006 4:45 pm
6402
Witam wszystkich, Zmieniam wynik partii [Number "60"] czyli 109 na liscie partii do przegl±dania [White "Robin 0.983"] [Black "Belzebub v0.65"] na zwycięstwo...
Hi Tomasz, Let's move the discussion about changes in piece/square tables until the end of the test tournament. I promise to look at the games carefully in...
Witam wszystkich, Zakończył sie turniej programów biorących udział w grupie 3. Wygrał program pikoSzachy. Gratulacje dla Piotra Cichego. Wyniki i partie...
... OK, now I see. Firstly, there were several inaccuracies. From my point of view h5 was too optimistic. Then you allowed to exchange your Be6 for Ng5...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
