Author: leonid
Date: 13:43:21 02/26/01
Go up one level in this thread
On February 26, 2001 at 16:11:53, Pete Galati wrote:
>On February 25, 2001 at 17:04:35, leonid wrote:
>
>>On February 24, 2001 at 18:27:41, Pete Galati wrote:
>>[........]
>>>I keep meaning to ask, is your program available to download anywhere?
>>>
>>>Pete
>>
>>Actually, I don't remember exact address of my program but it is at the Web Page
>>of my best friend. Address is:
>>
>>http://www.theparticle.com
>>
>>Mate solver is the part of unfinished chess program but solver is done
>>completely. If you will use it then I must to say that for mate solver one
>>feature is not indicated in description. Just recently found this. This feature
>>is change of number of nodes that program will see in each ply during the
>>selective search. It is: CTRL + P(p).
>>
>>If you will find that something is not clear, just say me. When you will use it,
>>you will find that all those "crazy" positions that I depose here you can create
>>yourself with bewildering easiness.
>>
>>Leonid.
>
>Thanks Leonid, I had lost track of this thread. I just downloaded the program
>and the html docs today, hope to have time to try it today. A Dos program! I
>like them.
>
>Pete
Will give you some, maybe not necessary, basic explantion of how mate solver
work.
From initial position push: Alt + |\ this is to make program use only its mate
solver.
Alt + I(i) to make mate solver search for mate only
by one way. Usually program use two methods before giving up.
For changing way of solving the mate, push Ctrl + D(d) Between 0 and 3 it is
selective search. 4 is brute force search.
If you want to change number of depth search push F7. Program can see as far as
13 moves deep.
If you want to clean chess board push Ctrl + Z(z).
If you want depose new piece on board push Shift + /?
You can try already two positions with this solver. One is 7 mate move. Position
from witch we start talking. Put selective search at 0 and depth to 7. Solution
should be short. It was, on my AMD 400, 0.06 sec.
Today position you can solve, for sure, at 11 moves (I don't know much shorter
solution). Maybe Heiner will come with one. His solver, when search is done by
brute force and with hash, make all the difference. I saw here few excellent
results. His hash goes well beyond 100M.
On today position you must put maximum number of moves seen on each ply [Ctrl +
P(p)] to 6. Put selective search at zero. On my computer it took 2 seconds to
give a response.
Leonid.
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.