Author: Albert Silver
Date: 14:46:16 02/13/00
Go up one level in this thread
On February 13, 2000 at 16:36:10, Vincent Vega wrote:
>On February 13, 2000 at 01:15:15, Albert Silver wrote:
>
>>It isn't time, it's plies. You said: "I understand that somebody is working on
>>confirming whether there is a linear ELO increase with ply depth." I read that
>>as an Elo value for each _ply_.
>
>Nope. He is indeed using plies and this is valid when _one_ program is used.
So he is studying the Elo value of each ply for only one program? What's the
point? He'll have to analyze the value for each and every ply along the line,
and even then the results will only be for that one program.
>But you tried comparing _two_ programs using plies which is no-no. You have to
>use time to get anything significant
Why is it no-no, and why do you have to use time to get anything significant? If
you want to refute what I've said about it, I'd like to at least see the
reasoning.
>
>>MindBlank, as described, knows only the material values, which as far as
>>knowledge is concerned, seems to me to be the strict minimum. As such, unless it
>>reaches a forced material conclusion, extra time will show it nothing, as I
>>expounded above. An example: Mindblank after 10 plies has not found any forced
>>win of material for any side. So if it is forced to play now it will take all
>>the moves it calculated don't lose material, and flip a coin (metaphorically
>>speaking). If at 12 plies it still has not found any forced material win, then
>>the same thing will occur. In other words it is absolutely dependant on finding
>>something at a deeper ply in order to choose a move that is non-random. Not so
>>for CyberGM.
>
>The difference however is that MindBlank should be able to process thousands of
>positions in the time it takes CyberGM to look at a single position. So the
>probability that it finds a capture somewhere is much higher.
How much higher? In my opinion this probability is close to nil unless the
position has already been compromised; yet how will Mindblack achieve this great
position with random move choices?
> Also even if
>CyberGM’s evaluation is very good, there will still be a great deal of noise as
>higher plies are evaluated.
>
>BTW, a real minimalist program NoMind would just look for mates and not even
>worry about all that stupid material :-)
I imagine that writing an ultrafast program with no knowledge would be extremely
easy. Try it. I think you'll find that it will lose so fast it won't even be
funny. Do you think it will magically start coordinating an attack against the
enemy king without any knowledge?
I suspect, no offense, that although you know how to play, that you aren't very
strong. I say this because I tend to see these arguments that knowledge is
over-rated by people who do not realize just how deep knowledge can go. A GM has
_skills_ that a 2200 player simply doesn't have. It goes beyond the simple 'he
plays better positional moves and calculates deeper'.
Albert Silver
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.