Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: ChessBase tree feature is definitely bugged. NO

Author: Henrik Dinesen

Date: 09:36:23 05/22/05

Go up one level in this thread


On May 22, 2005 at 11:47:59, chandler yergin wrote:

>On May 22, 2005 at 10:37:26, Komputer Korner wrote:
>
>>On May 22, 2005 at 03:54:58, chandler yergin wrote:
>>
>>>On May 21, 2005 at 22:09:01, Komputer Korner wrote:
>>>
>>>>On May 21, 2005 at 20:17:17, chandler yergin wrote:
>>>>
>>>>>You can include variations i.e. analysis that is included in games you input,
>>>>>but these are not included in the statistical analysis.
>>>>>Transpositions are handled perfectly, and even reverse color openings.
>>>>>Try 1.c3 e5 2.c4 the Program will follow the Sicilian with White.
>>>>
>>>>
>>>>You don't have to include whole games to the tree. It was designed to add
>>>>individual moves to the tree or a partial line of moves which is the same as
>>>>adding individual moves to an opening book.
>>>
>>>Yes when adding games, truncate them to usually 20 moves beyond the ECO
>>>Classification.
>>>
>>> The statistics were included by
>>>>Chessbase from games they imported and then pruned off most of the tree so that
>>>>just the opening was left. You can also do the same thing by adding games or
>>>>even just add statistics without adding the games.
>>>
>>>Statistics are compiled from Database searches of games played.
>>>You cannot 'add' statistics.
>>>
>>>Either way, your choice, but
>>>>my original post stands. There are bugs in the tree.
>>>
>>>No there are not. I suggest it's 'Operator Error'.
>>
>>
>>Please look at the Chessbase Help on trees. It plainly states that you can add
>>statistics to the tree based on an external database and you don't actually have
>>to add the game lines to the tree.
>
>Please Post exactly what it states; we can be the judge of what it means.
>
> The submenu is
>>Edit/OpeningsBook/LearnFromDatabase
>>
>>However that was not the focus of my original post. My original post clearly
>>stated that sometimes when adding a move to the tree, the tree will add bogus
>>statistics to the move even if no retro move shows from a possible
>>transposition. I have the retro moves feature turned on.
>
>Yes, and unless you have 'removed' the write protection; you have NOT added
>a move to the Tree. You have entered it on the board, that's all.
>This is probably why you think there is a Bug.
>Next..
>You are asserting again, things not in evidence.
>You cannot 'add' bogus statistics; statistics are calculated, from
>positions in the Tree.
>(what kind of juice have you been nippin on?)
>
>Statistics:
>If you switch on this  display option the exact statistical details of
>each move will be displayed at the bottom of the list.
>For any 'highlighted' move, the Program displays at the bottom, the
>exact number of games that were won, drawn or lost.
>The Bar Graphics make it easy to take in the information at a glance.


When you refer to "write protection", I don't suspect that you mean the general
write protection from the files, but that your're refering to "allow move
adding" - inside the GUI!??

In normal terms write protection deals with the files themself.
"Alow move adding" is just a Chessbase precaution against accidentially manual
adding of a move.
Write protection itself is disabled when the book is imported to the harddisk -
same as with the database.

And, BTW, of course the stats isn't changed when you add a move - or a serie of
moves - manually og import "games" _without_ results. Number of games(N) and the
percentages(%) are only influenced by results.
In both cases they are treated as variantions/single moves (but when adding
moves manually, those are set as 'main moves' by default, which isn't the case
if you use the 'Import games' feature).

Regards,
Henrik



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.