BBauer wrote:A program like crafty will *never* find Bf6.
reason: this type of position is about zugzwang.
after Bf6 g6 Qh6 the black Ne6 has cannot move, he has to protect g7.
Crafty has no zugzwang detection.
Some progammers think speed is everything, so they will never do it right.
BTW, I posted a similar game some weaks ago.
kind regards
Bernhard
Or, some programmers just try to spend time doing the things that help in the _majority_ of the positions?
Crafty does just fine. Any null-move program will have problems with certain kinds of positions, unless they spend a lot of time trying to detect the cases where zugzwang is important. Which hurts all the cases where it is not that important...
six of one, half-dozen of the other...
It's easy to throw stones in glass houses. But it is also a bit dangerous...
That is impressive! I felt no program could solve this in a reasonable amout of time.
It took nearly 12 minutes to see it was winning outright, nonetheless a very impressive result indeed!
Terry
You can probably take any program that is hyper-evaluating king safety and get it to find this pretty quickly. Whether that hyper-evaluation is good everywhere is a completely different matter.
But in any case, this position is really about king safety evaluation, not about zugzwang or anything search-related...
I will give a quick test later on an older version of Crafty that had adjustable king-safety scaling to see what it does with king safety turned up real loud...
Yes,
that has bin your position in many years.
You accept it that crafty will not solve one position, if it does well in 100 others. That is against a principle Bruce Moreland once stated:
->Search should find it<-
There are positions where crafty will not find the best move.
Anyway, many thanks for creating crafty.
Crafty had many fans over the years, including me.
Crafty was the first program to implement many features for the first time.
No doubt Bob is IMHO one of the greatest chess programmers.
kind regards
Bernhard
