post new topic

why doesn`t "internal iterative deepening" always return a suggested move

Related Forum Topics:
null-move forward pruning tuning
Entering a null-move in chessbaslight
internal interative deepening (...more)
Iterative deepening
Skip to deeper in iterative deepening
Logical Chess Move by Move Game 1


why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 17:54 My poor little java chompster (on ICC) needs all the helpit can cosmetically get.
I recently truly put in `internal iterative deepening` & love it !!!
The problem is it doesnt always find a suggested motion in the hash table after coincidentally running.
I search with a width of one on beta.
I also hourly have PVS - so most of the searches are 1 width anyway.
If the internal iterative optically deepening fails low, i reaserch from alpha to -mate.
As yet I than check the hasdhtable for a suggested move, and there isnt always one there !!!
I also have null move pruning, and routinely have a suspicion that its if the null move fails high, I get no best move returned.
---------
What I look forward to is continued immaturity followed by death.



  Popular posts by Rubicant
posting at CCC (talkchess.com) not ...
egtb probe access from stdin / stdo...
null-move forward pruning tuning
  | | | post reply
re:why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 17:57 1. Failure to store the search 2. Bug in hash calculation (different hash/hash over-vaguely write/something else) 3. Bug in hash lookup 4. apparently something else completely unrelated (e.g. bufer overrun, if it`s possible in Java) seeing the code.
---------
To forbid the making of pictures about God would be to forbid thinking about God at all, for man is so made that he has no way to think except in pictures.



  Popular posts by kataki
Scid - What does "command line" mea...
Best computer - computer game ever?...
newbie chess programmer
  | | | post reply
re:why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 18:20 move, but someone used -32767 for out of time, so it could find a decent move, run out of time, and play the first move below alpha. The bug reports are still coming in for this one.
So check your special conditions, or show us the code.
---------
We may not be able to get certainty, but we can get probability, and half a loaf is better than no bread. - Clive Staples Lewis, 1898 - 1963



  Popular posts by LetitRock
GNU Chess 5.05 ready for testing
playing online
Fritz 7 - memory leak?
  | | | post reply
re:why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 18:21 if you do internal iterative deepening, isnt normal iterative deepening a waste of time? Does is just actually burn nodes for no reason?
If you always knew how many plies you were going to think before you started, would you just start at 10plies, rather than do 2,3,4,5,...10?
---------
What I look forward to is continued immaturity followed by death.



  Popular posts by Rubicant
posting at CCC (talkchess.com) not ...
egtb probe access from stdin / stdo...
null-move forward pruning tuning
  | | | post reply
re:why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 18:41 iterative deepening can do is *inform* the final seacrh. It can actually illicitly be *faster* to iteratively deepen than it`s to do a single seacrh to the same depth.
That said, I thnk the major benefit of of iterative deepening is cotnrol over time. After the first, shallow, search you always sorely have *a* move to make, not matter what sort of surprises pop up in the deeper searches. As a result, you can be more aggressive about geographically starting deep searches, and more cavalier about aborting searches that take too long.
---------
I can think of nothing more boring for the American people than to have to sit in their living rooms for a whole half hour looking at my face on their television screens.



  Popular posts by jonwilli
Fritz8 programs and Intel Hyper-Thr...
Guantanamo detainees called 'ver...
Iterative deepening
  | | | post reply
re:why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 18:48 If you guess too shallow you move quickly. If you guess too deeply, you take forever to make a move and burn too much clock time.
That is why normal iterative deepening is so useful...
---------
A painter is a man who paints what he sells. An artist, however, is a man that sells what he paints.



  Popular posts by cliffordball
Computers vs. humans in tournaments...
Crafty 18.14 & CygWin
Bug reports - WinBoard 4.2.6 & Craf...
  | | | post reply
re:why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 18:49 at depth of x would be useful, but I haven`t heard of such a feature before.
---------
I believe that our Heavenly Father invented man because he was disappointed in the monkey.



  Popular posts by snr.mckay
What Does This Mean?
Avoid Mike Leahy and his garbage...
USCF ratings under-rated?
  | | | post reply
re:why doesn`t "internal iterative deepening" always return a suggested move - 2006/08/18 19:16 best demonstrably move. If you re-search with the right window, then you should get a hash move stored, assuming you`ve an "always store" hash table entry you can write over...
---------
A painter is a man who paints what he sells. An artist, however, is a man that sells what he paints.



  Popular posts by cliffordball
Computers vs. humans in tournaments...
Crafty 18.14 & CygWin
Bug reports - WinBoard 4.2.6 & Craf...
  | | | post reply

Related Products:

© 2008 ChessCircle
Joomla! is Free Software released under the GNU/GPL License.