Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Congratulations to Rebel Century

Author: Ricardo Gibert

Date: 08:43:47 10/04/99

Go up one level in this thread


On October 03, 1999 at 23:44:31, Robert Hyatt wrote:

>On October 03, 1999 at 23:17:29, Ricardo Gibert wrote:
>
>>[snip]
>>>
>>>my webster's defines 'sacrifice' as 'voluntarily giving up something of
>>>value'.  I have a hard time saying 'I will sacrifice a ten-dollar bill if
>>>you will give me a 20 dollar bill in return...'
>>>
>>>:)
>>
>>Ok, you got me. I neglected to explicitly state I was refering to the _chess_
>>version of the term.
>>
>
>
>then here is a 3-move sequence. Sacrifice or combination?
>
>RxB, NxR, RxN.
>
>RxB obviously dumps a rook for a knight.  or if you look to the end of the
>combination it wins two pieces for a rook which is a significant advantage.
>
>Sacrifice or combination?

Impossible to tell. You removed it so completely from context, it may even be
neither. At the end of a forced(?) sequence a rook was traded for 2 minors.
That's all I can say. A silly example any event, even if I make generous
assumptions. Let's move on...

>
>How is that different from QxP+, RxQ, RxR#??
>
>Dumping a queen for a pawn?  Or winning the king?
>

A modest Q sac to mate it "seems".

>
>>>
>>>But I don't object to the term being used..  I just think that for a computer,
>>>the concept 'sacrifice' is wrong.  It is just a perfectly computable
>>>combinational tree search...
>>
>>You can give up a bishop to obtain a draw by perpetual check and because you
>>never get the material back, it is a called a sacrifice. I know it seems trivial
>>and is not what people generally have in mind when they use the term
>>"sacrifice", but I do believe it's use in such cases is fairly universal.
>
>
>in the case of a computer, it isn't 'sacrificing'.  It _sees_ that it can
>draw or that it can win.  IE it isn't giving up _anything_.  A human might

No it doesn't. Taking the Rebel game as an example, it "sees" that within the
limit of its search that it nets material, but does not "know" whether an even
deeper search will require it to return the material (or get mated even) with
interest. It "assumes" that is not the case. A strong assumption, yes, but it is
essentially a "gamble" imposed on it by the programmer. The element of risk
remains. It isn't over until it's over. The only way a program "proves" anything
is by reaching "terminating" (win or draw) positions in _all_ relevant leaf
nodes in its search. There are other ways of proving things using schematic
thinking, but that is another subject not relevant here.

>toss a bishop 'thinking' (but not sure) than he can force a perpetual.  But
>a computer either 'proves' that it can force it, or it won't ever go for the
>move in the first place.  IE we (as humans) gamble on things all the time.  But
>would it be the same as saying "I'll flip a coin and if it is heads I win, and
>if it is tails you win" if I rig the coin so there is _no doubt_ that it will
>end up heads when I want?

I have catalog where you can buy such coins, but it will not help your computer
in playing chess ;-)

>
>That is the minor point here...  computers don't sacrifice in the traditional
>way usually.  There are exceptions like the famous chaos sacrifice vs chess
>4.x where chaos didn't see any materian coming back, but thought the position
>justified the Nxe6 sac anyway...  I see a number of those in Crafty.  More than
>I really want to see.  But they do come close to the definition of a sacrifice
>as nothing "real" is won back, just some intangible positional things that may
>well not be enough to win with.

When you sac bishop to draw, you don't get _any_ material back. A true sac!

>
>
>
>>
>>In discussions about the term, people seem to want to reserve the term to
>>describe the giving up of material for "uncertain" compensation. The "romantic"
>>view on sacrifices. But then in the next minute they will apply the term to the
>>case I describe above. The common thread is: it is a sac if it voluntarily gives
>>up material which you don't get back or do not get back in an obvious way.
>>So a sac can be short term and quite calculable.
>
>
>your last phrase "don't get back..." is the key I have been talking about.
>A computer (in the case of the rebel move) "sees" that it gets the material
>back, just like in the two examples I gave above with the rook and then with
>the queen..

But what about the example of a drawing sac? Why leave that one out?
Technically, mate (mate is really misnomer) is a "forced concession of the
battle", you don't actually kill the king. That's why we have a complicated
mating rule. It would be simpler to just capture the K, but the inventers wanted
to keep the game "polite" and leave the really _fun_ part out. Perhaps, because
the game was actually anticipated to be played by real kings, they decided it
was "wiser" to leave the last grisly detail out. Killing a king is not really
fun when a _real_ king can really kill you in return. But this not _my_ reason
for calling it a sac. It is a sac when I give something up and I don't collect
the material back, either because of mate or a forced draw or whatever (=your
version?). Being "prevented" from getting the material back, because an opponent
resigns or draw is agreed, does not count of course.

>
>but you are right, it is often used vaguely...  or incorrectly (IMHO)...
>
>
>>
>>As for sac'ing to obtain mate, it is semantically arguable that you are getting
>>your material back and therefore it is not a true sac, but I don't see that as
>>natural or useful. In common _chess_ usage, it is called a sac and is
>>practically speaking consistent, when you consider its use to describe giving up
>>material to obtain a draw.
>>
>
>
>yes.. which is why I prefer the term "mating combination" rather than "mating
>sacrifice" as the latter could easily be termed an oxymoron...

By my way of looking at things, a mating combination may or may not include a
sac. I prefer to keep the distinction, rather than meld them together as you do.
People enjoy the sac version of a combo more. Reason enough to keep the
distinction IMHO. Definitions are fairly arbitrary. I choose an objective
definition that can be applied by anyone. You choose one that is subjective.
Taking the Rebel game as an example. If a 1200 played that "sac" it would be a
true sac, since he cannot reasonably be expected to calculate all the
ramifications. If a GM plays the sac, maybe it is a sac or not, depending on the
player, time situation, etc. A GM could calculate it to the point where he gets
the material back or play it based on intuition. I don't care for this
ambiguity.

>
>
>
[snip]



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.