Winboard Forum

You can click the X button at the top right corner of the window, I just bought Fritz 8, and it allows any engine to show the score of all moves of a given position with any engine compatible with Fritz 8 So sorry to ruin your fun if my Grand son discovers mate from the starting position in year 2112, when people won't die anymore

This topic is about all winboard engines only showing the 'best move they found and hiding everything else', and finding a way to show that information for opening analyzis (Nobody told me that Fritz does that, but in another topic, someone gave a link to a site with Fritz 5.32 for download, and I figured out Fritz 8 was my best option); most information of this topic its somewhat X (Which means, a loss of time for the reader).

Hello, I am trying to do what Nalimov did with the tablebases but with the opening, there are good moves and bad moves, and I'm computing all the important ones, give them a score, and maybe discover that the starting position is Mate in 216 for white... (In Futurama, year 3000, a robot says it's 187 moves).

I'm doing it with Fritz 8, I'm going to run a tourney with all the important commercial engines (Shredder 8/9, Fritz 8, Junior 8, Ruffian 2, Hiarcs 9, Hydra and so... Tell me if I'm missing one! please ) and the best free ones (Prodeo, Pharaon, Aristarch, List, GLC300.32, LittleGolitath and even some non premium ones, like Movei, my favorite ).

I remmember when I tryed something similar with Rebel 12.00.02 a while ago with "one move at a time", but the score got biased towards the enemy's score (White gets Black's score and viceversa) and the time required went up exponentially (Now I get the score of 8 moves in Three minutes, when it took, in the past 24 minutes with one move at a time, and 576 minutes [OMG] with one move at a time for the next ply against 27 minutes now that I can make the engine to already show the best 8 moves at a given time)

I have wanted to do this for a long time, but now I have the time to do it, I need to know what it's the strongest computer in my P4 2.5Ghz 256 RAM, to have the best results.

You can suggest anything you want... I may do it with several engines with the time I save, and be finnished within months to a very deep ply ala Chess Book Browser http://catalin.francu.com/nilatac/book.php After this, Opening books will never be the same! (I hope)

Thanks in advance.

Ulysses Omycron wrote:...

skoony wrote:r1bqkb1r/4pppp/ppnp1n2/8/3NP1P1/2N2P2/PPP4P/R1BQKB1R w KQkq -
...

...

Where did skoony wrote that?

Volker

http://volker-pittlik.name/wbforum/viewtopic.php?t=886

skoony had a wrong PGN and thought Kasparov actually made super duper Blunder.

So, huh, oh, anybody reading this, what about clicking the RED X up there?

I'm pretty patient, though (Chorus: ALT+F4 will close me!)

Ulysses Omycron wrote:http://volker-pittlik.name/wbforum/viewtopic.php?t=886
....

Strange, the search engine doesn't find it...

Volker

Ulysses Omycron wrote:Hello, I am trying to do what Nalimov did with the tablebases but with the opening, there are good moves and bad moves, and I'm trying to compute all the important ones, give them a score, and maybe discover that the starting position is Mate in 216 for white... :P (I'm serious about it!).

Don't be.

I don't even know where to start discouraging you. Even _if_ you'd manage to get a couple of super computers to help you and a couple of decades time, I'd really rather you didn't make the terrabyte harddisk (or even if it's a win or draw) publicly available.

Chess wouldn't nearly be as much fun if I knew it was solved.

Have fun though, Laurens.

Hello, I am trying to do what Nalimov did with the tablebases but with the opening, there are good moves and bad moves, and I'm trying to compute all the important ones, give them a score, and maybe discover that the starting position is Mate in 216 for white... :P (I'm serious about it!).

Well I want to add that the 'CAP' project worked on this already
for several years IIRC. AFAIK the project was stopped quite a
while ago.
Their goal was of course not to find mates from the opening,
IMHO this is impossible and will remain impossible for a looong
time(if not forever).
Results from the CAP project are integrated e.g. in ChessAssistant
and IIRC you will find some informations about it at Dann Corbits
site or just do a google search for 'CAP' in relation with computerchess.

Guenther

Ok, so if I solve chess, I will put a SPOILER warning on my pages...

In order to know the best move, an engine should search them all, this score, it's the one compared against the best score, if the score is higher, the move is better

Some engines allows you to select a noise option so in the debug file you get to know the best moves other than the original.

Anyway, I think that with this, even if I don't make something like BookBases, I may make a very powerfull opening book with the "Elite Moves". I tryed this with crafty, and althogh a book with only 1000 moves looked way pretty decent (Making crafty have a positive score always that it left book, and sometimes hold the book while some engines that had their own book like Delfi or GreenLight Chess would leave book early and still be losing), its flaw was the 'Horizon effect', I had to change the book from line to line several times, because I only had the best 2 moves (The best crafty picked and one that was selected as the best in previus searches, that is not the second best in actual ply, Or maybe it is since Crafty doesn't show its info), then sometimes the 'best move' was worse than the second book move, or maybe a 'random' move by the enemy gave a better score (That's how I discovered that the Queen's gambit is actually much better than clueless Crafty thought), the horizon effect won't go away, of course, but I only had tested crafty and changed its book accordingly about its games; but now that I will have a list of all moves of the elite lines, being fooled is going to be hard.

That is based on the fact that an engine that covers the first 60 moves of an opening, wont need to think till move 61, now that's a big time advantage, and more if you get a positive score all the time.

I'm already the happiest man in earth!! (At least the next two weeks) If you want to help me in any manner you'll get credit for it, so it will be ?lycrom and [insert your nick here] solved the first 20 moves of chess in year 2014

Well, thanks for your answers [Though the forums' owner could PMed me instead of postig... How did he know I'm a troll??], I'm gonna take a look to the cat... err... The CAP.

I'm actually surprised people take chess too seriouly and can't handle a bad joke right, blinking won't work, they may be flame war starters on their dreams .

So the annoying factor varies from persons to person, but take chess' posts without humor is the worse technique.

Ulysses Omycron wrote:Well, thanks for your answers (specially to Volker Pittlik, that although his answers were more close to SPAM than my fingers to my keyboard, he PMed me!! what a honor!! ), I'm gonna take a look to the cat... err... The CAP.

This kind of comments are not welcome, and of course it has no basis. So please, don't speak like that again, or moderation action will be taken.
As for winboard engines, and most other engines, only the first move is searched for a score, the rest are just refuted, compared to the first, so no actual score is calculated.

Ulysses Omycron wrote:But are you being serious about that ALL winboard engines search ALL moves on a given position

Is who being serious? I didn't see anybody say "ALL that winboard engines search ALL moves on a given position".

But if they did they don't.

Dan H.

Actually, as Fritz 8 shows, they do.

I greatly appreciate this topic to be deleted, since this is a top secret idea , or already PM all the members something like "See that this dude ulises omicrash did! he called Volker Pittlik a SPAMer!!", I didn't try to be offensive, that joke is a good one here in M?xico, but it seem the wrong place to post it, so I apologize for having a twisted sense of humor and I promise to not do it anymore on purpose.

?lycrom, the troll that is so optimistic in solving chess in the not near future.