![]() |
| If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|||||||
| Tags: ahead, intelligence, march |
|
|
|
Thread Tools | Display Modes |
|
#11
|
|||
|
|||
|
$10 / Month. Will people prefer such subscription more than saying buy 100 games for $10? Sanny, we've all been trying to tell you that you don't currently have a marketable product. I would not pay anything at all to play on GitClub when for $60 I can have Fritz or Rybka for unlimited play and analysis. For the same $60 I can play on ICC for a year. |
| Ads |
|
#12
|
|||
|
|||
|
On Mar 16, 11:22 am, wrote:
After creating "GetClub Chess" I think I should design something more interesting. Can you devise a program that plays chess? He did a fair job of creating "pseudo-chess". It is interesting to note just how difficult it can be to get all the details right-- which shows just how complicated the game has become. Imagine, if you will, trying to explain to a person who knows nothing about chess, every single detail involved in the rules-- quite a task. It makes me stop and think twice about writing a chess program for my TV remote- controller. -- help bot |
|
#13
|
|||
|
|||
|
On Mar 17, 1:17 am, Kenneth Sloan wrote:
Sanny wrote: Can the Chess Players give me some thing they would like to play at GetClub. How about chess? In one of my recent flubs, I had a K, N, and p against K and two pawns, but just left my Knight /en prise/. Unexpectedly, Sanny's program allowed its King to LEAVE THE CORNER! As I walked clear to the other side of the board to capture the program's passed Rook-pawn with my King, it walked right up and ate my pawn! However, it did *not* know anything about "opposition", so when I ran back I was able to (just barely) get in front of the enemy pawn, and we finished with a quick and painless stalemate. Here's the funny part: the Web site display had this as a win, the pop-up box at the end of the game showed a draw, and the ratings? I was penalized just the same as if I had lost, and I expect this is because Sanny's Web site refuses to acknowledge the existence of draws in chess, and it has therefore been instructed to forfeit lowly humans-- the scum. It is truly amazing just how powerfully the program can play at times in the middle game, yet it still doen't know the rules of the game nor can it play the endgame any better than a eight-year-old child. As we saw, the program easily snatched a piece away from "Ivan", the supposedly 1800 strength commercial chess program, in the middle game, going on to mangle the endgame and lose. I can only conclude that todays microprocessors are somehow able to make up -- in part -- for what ought to be a terrible program's weaknesses. A glance at the depth of search indicates a VAST improvement from the early days of GetClub, in terms of raw speed. Java applet? No problem: just replace Sanny's code with a translation of the old Fidelity Chess Challenger 7 program and presto, most humans are mincemeat! Heck, if Fritz or Rybka could take over in the endgame, Sanny's program would seem to be quite strong except in the openings; but playing weird moves in the opening is a good way to take human opponents out of their memorized by rote schemes, where both sides are more likely to err. Here is one example of a programmed-in opeing line at GetClub: 1. e4 Nf6, 2. e5 Ng8. I don;t know who is making these decisions, but they should have hired a real expert, like say, Rob Mitchell or Neil Brennen for this job. -- help bot |
|
#14
|
|||
|
|||
|
* Here is one example of a programmed-in
opeing line atGetClub: 1. e4 *Nf6, *2. e5 Ng8. *I don;t know who is making these decisions, but they should have hired a real expert, like say, Rob Mitchell or Neil Brennen for this job. I read somewhere The opening, 1.e4 Nf6 2.e5 Ng8 is sometimes known as the Retreat or Brooklyn variation to the Alekhine's Defense. If you wish getclub to know more openings send me a list of openings you wish getclub to play. Bye Sanny Play Chess at: http://www.GetClub.com/Chess.html |
|
#15
|
|||
|
|||
|
On Mar 19, 1:47 am, Sanny wrote:
Here is one example of a programmed-in opeing line atGetClub: 1. e4 Nf6, 2. e5 Ng8. I don;t know who is making these decisions, but they should have hired a real expert, like say, Rob Mitchell or Neil Brennen for this job. I read somewhere The opening, 1.e4 Nf6 2.e5 Ng8 is sometimes known as the Retreat or Brooklyn variation to the Alekhine's Defense. Yeah, a lot of dumb moves have names attached to them; this is *not* evidence of greatness, my friend! Remember, Fools mate and Scholars mate are named opening lines as well, as are the Grob, Damianos, and so forth; it just means that someone gave them names-- that is all. A much better way to play this opening is: 1. e4 Nf6 2. e5 Nd5 -- help bot |
|
#16
|
|||
|
|||
|
I read somewhere
The opening, 1.e4 Nf6 2.e5 Ng8 is sometimes known as the Retreat or Brooklyn variation to the Alekhine's Defense. * Yeah, a lot of dumb moves have names attached to them; this is *not* evidence of greatness, my friend! *Remember, Fools mate and Scholars mate are named opening lines as well, as are the Grob, Damianos, and so forth; it just means that someone gave them names-- that is all. That opening has been removed. Is there any other opening you wish to be removed or Added. Send me a list of openings that you find missing from GetClub Chess? I find on winning with Beginner you get +5 points and on loosing you lose -20 points. And since you win 3/4 games with Beginner Level you remain at the same position where you started. So last 2 days you gain 40-50 points and again loose 40-50 points. So your rating remains at same place. I am happy to see you have come to 2nd position. Beating Zebediah rating is still very difficult. Play Chess at: http://www.GetClub.com/Chess.html I am thinking to add two things to GetClub Chess. 1. It thinks in Opponents Time. So it will play much faster. 2. Take use of dual processors. So if you have dual/ Quad Processor the game will be 2-3 times faster. How many players here has Dual Processors and Quad Processors. Can I assume more than 60% has dual Processor and more than 20% players have Quad Processor? If so then using advantage of multiple processors will make the game stronger. Bye Sanny Play Chess at: http://www.GetClub.com/Chess.html |
|
#17
|
|||
|
|||
|
On Mar 19, 2:56 am, Sanny wrote:
I find on winning with Beginner you get +5 points and on loosing you lose -20 points. And since you win 3/4 games with Beginner Level you remain at the same position where you started. So last 2 days you gain 40-50 points and again loose 40-50 points. So your rating remains at same place. Not quite; I have now (just barely) passed that other guy, nomorechess, and that makes me better than him. ;D I am happy to see you have come to 2nd position. Beating Zebediah rating is still very difficult. If I can just stop hanging pieces in the endgame, I could catch Zeb pretty quickly. I am thinking to add two things to GetClub Chess. 1. It thinks in Opponents Time. So it will play much faster. Only works if you fill hash tables with relevant data; other wise, you end up starting over when the opponent plays an unexpected move, right? 2. Take use of dual processors. So if you have dual/ Quad Processor the game will be 2-3 times faster. With a Java applet, this requires that the user have dual or quad processors; I wouldn't worry about that just yet. First, you need to raise the level of play in the endgame to somewhere above a 1000 (beginner) rating. How many players here has Dual Processors and Quad Processors. I've got an AMD Turion64, but my OS is only 32 bit, I think. Still, it seems to be running fast enough to play a decent game. Can I assume more than 60% has dual Processor and more than 20% players have Quad Processor? If so then using advantage of multiple processors will make the game stronger. True, but making a weak program run faster is not particularly productive, as compared to just fixing the weakest areas. Note that not only is the endgame still very weak, but in many of my wins, I see the program moving its King into check as its final move; surely, it could play legal moves instead and try to delay checkmate as many moves as possible. The same thing applies when the program is winning: it should make an attempt to force checkmate in the quickest way possible, the fewest moves. I am wondering if you have set the piece value of the King too low; in general, a King is worth more than a bazillion pawns; this is because the rules of the game make the King THE central focus of play. -- help bot |
|
| Thread Tools | |
| Display Modes | |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Chess and Intelligence | Chess One[_2_] | rec.games.chess.misc (Chess General) | 4 | March 9th 08 05:44 AM |
| Channing and Goichberg will ban Sam Sloan from the USCF forum | judgementday@clerk.com | rec.games.chess.politics (Chess Politics) | 16 | May 6th 07 11:06 AM |
| Channing and Goichberg will ban Sam Sloan from the USCF forum | judgementday@clerk.com | rec.games.chess.misc (Chess General) | 16 | May 6th 07 11:06 AM |
| intelligence | sennajawa@gmail.com | rec.games.chess.computer (Computer Chess) | 4 | July 21st 06 12:40 PM |
| intelligence | sennajawa@gmail.com | rec.games.chess.misc (Chess General) | 4 | July 21st 06 12:40 PM |