<lost in space> I can understand your frustration, but it would be a real shame for us to lose your analysis on Saturdays and Sundays. Let's cross our fingers that this was a one-off.

Fritz 11 gives the position after 26. Bxd4 as -1.2. <lost in space> has already posted this line (which Fritz also thinks is best)

25 ... Nd4 26. Bxd4 Rh6 27. Rg1 exd4 28. Qc4 Rh4 29. Rg2 Qf6 30. Qc5 b6 31. Qc6 Qe7 32. Bf1 dxe3 33. fxe3 Qxe3

click for larger view

I think I would say this is more than <=>. Black has won a pawn and the white pawns are all isolated. His knight is better than white's bishop and his rooks are more active. Fritz calls this -1.55, and I think I agree.

So I will allow that the original puzzle on move 25 just about satisfies the criteria for a Sunday. However white responds, black has a strong initiative.

But to illustate just how tough this was, Fritz took several seconds on the position after 25. b5 to find 25. ... Nd4. Normally it finds the puzzle solutions in a heartbeat.

I was very unfair with <Jason Frost>. The reason was my anger with <CG>. I gave him simply a line I found in my old scribbles from 8-9 hours ago.

Sorry, I published 3 times an answer to your question and deleted them directly afterward because I found a lot of locical mistakes.

Seems, that I am too tired now and have to stop here.

33. Qd4 is a good move compared to Qc6 and Qxd5.

Tomorrow is an other days and this day is called monday!

Material: Down 2N+P for 2B. The White Kh1 is tucked away in the corner with 2 legal moves, but is has no defenders. Black has a battery Rd8 and Rd6 obscured by the Black Nd5. White threatens 26.bxc6 or 26.e4 forking Nd5 and Qf5. The Black Nc6 exposes a weak Pc7 if it moves. No check or capture seems a feasible candidate, so on to threats.

Candidates (25…): Rh6, Qh3, Ndb4, Ncb4, Nd4, Nf4 Rg6

25…Rh6 (threatening 26…Qh3 [or Qh5] 27…Qxh2#)

25... Nd4 26. Bxd4 Rh6 27. Rg1 27...exd4 28. Qc5 dxe3 29. fxe3 29...Qe5 30. f4 Qb2 31. Rxc2 Qf6 32. Bf3 b6

33. Qd4 Qxd4(forced)
34. exd4 Rhd6

[34...Nxf4 35. Rxc7 Rhd6 36.Bc6 =]

35. f5! Ne3

[ 35. Bxd5? Rxd5 36. Rxc7 Rxb5 37. Re1 g6 38. Ree7 Rxd4 39. Rxf7 Rxd3 40. Rg7+ Kf8 41. Rgf7+ Ke8 42. Rce7+ Kd8 43. Rxh7 Rb1+ 44 Kf2 Rb2+ 45. Kf1 Rxh2 46. Rxh2 Kxe7 47. Rb2 Rd6 and despite only bieng down a pawn the rook ending appears lost for white, as the king is cut of from stopping the b pawn ]

36. Rxc7 Nxf5
37. Bc6 Nxd4
38. Ra1 h6

[38...Ne6 39. Rb7 Rxd3 40. Rxb6=]
[38...Nxc6?? 39. bxc6! Rxd3 40. Rb7! and black is actually loosing ]

39. h4! Rg6
40. Re1! Rg4
41. Re4 Rxe4
42. dxe4 Rd6
43. Kg2 g6
44. Kf2 Kf8
45. Ke3 and it looks like white should be able to hold for a draw

Not sure this is 100% correct but with best play it looks like white can hold a draw after 26. Bxd4

Do you know of a better site to test your observational skills over a week's period with generally increasing levels of difficulty? I would be very surprised if you knew of a better site. The quality of the real life games selected here for analysis can't be beat, IMH

I have the impression you want to feel important in this community. Interesting comments and good analysis no matter what the puzzle is what will attract us to your comments.

No need to go at the host with a panga/paper shredder/scythe/wrecking ball/sudicide bomb (depending on your culture/upbringing).

You are no doubt aware of the CG.com policy of throwing in spoilers. Whether this is a CYA policy or intentional sometimes, its still worth pointing out to people who may not have read "about our puzzles":

<Finally, we occasionally show a puzzle that we call a spoiler. These are positions where there is no move that clearly wins the game, but instead a variety of solid moves which are all playable. Usually these positions present the lure of a sacrifice which is unsound. You are expected to recognize the unsoundness of the tempting sacrifice and instead conclude that the best move is one of the "quiet" moves. We show spoilers from time to time because it encourages people to think combinations all the way through, instead of simply finding a move which looks like it initiates an attack, without considering the defenses.>

Thinking's the thing, by which we will capture the king.

Actually, there is a need.

<Once> and <lost in space> contribute cogent comments, which I greatly appreciate. I get on the site late, which has shielded me from its quirks today and yesterday, so I have no personal feelings to vent. I can well imagine, however, the frustration these two good gentleman might feel over being told that they devoted their considerable time and energy to misposed problems.

The problem was that the puzzle position changed during the day, with the effect that some of us were commenting on one position and some of us were talking about another. Cue confusion and frustration.

But, hey, we're all human and that means we makes mistakes from time to time. By pointing it out in the comments, we hopefully helped CG to spot it and put it right later in the day.

That emphatically does not mean that we are being critical of Chessgames.com, that <lost in space> doesn't think that this is a good site, or that I need to have the guidance notes repeated back to me.

I think we may need to agree to disagree on this one. This position is way beyond my ability to calculate as both white and black have so many choices. That's when I rely on silicon assistance.

Fritz 11 deviates from your line immediately with: 26. Bxd4 Rh6 27. Rg1 exd4 28. Qc5 Rh4 29. f4 Qh3 30. Rg2 dxe3 31. fxe3 g6

click for larger view