On May 28, 3:48 am, help bot wrote:
On May 27, 5:10 am, help bot wrote:

I do not believe it possible either. I think Black had a win there.

Sam Sloan is obviously correct. (Take note, for
I only type those words about once in a decade.)

It is just that I have not been able to find it (and I have not tried
hard either).

That's probably because it doesn't exist. White is
clearly winning.

What I *meant* to say here was that Black is clearly
winning (not White).

The fact that White still has a pawn on a4 means
there are no annoying stalemates for Black to worry
about, and he can therefore concentrate on weaving
a mating net or just picking off the loose Rook. For
a human, this may take a bit of effort, but I'll wager
a modern chess computer can pull it off *effortlessly*.

-- help bot

Actually it seems to me that it is a fortress white is trying to build
- the a4 pawn is just let go... The idea is not stalemate but a
fortress. Finding these things are neat - recently, in one of my un-
but soon to be published studies, I found a fortress of B+2P that
draws vs a queen (and a similar one in the same problem that didn't -
it was tons of fun)

but a position like Kh2 Rf3 g2/ Ke2 Qa4 h6(FEN: 8/8/7p/8/q7/5R2/3k2PK/
8 w - - 0 1)- can this work - seems hardly possible with the h pawn
there to cause trouble (attack g2 w/K+Q, and play the P to h4 to drive
the R from g3, etc). That's from the human perspective, and is quite
wrong.

But yes, a modern chess computer can pull it off effortlessly - just
look it up in the Nalimov tablebases - it is a draw with white or
black to move. Practically, black would have some good winning
chances, but with *best play* it is drawn.

However, Fritz and others evaluate this position as "lost" for White,
down by a factor of 5. But that is because all the computer can do is
count; it cannot play chess. Would it beat you? Probably, but....

This sort of thing is covered in the MAMS book that I just
reviewed..... And it looks to me like Whitaker was correct.