|Dec-07-17|| ||amateur05: Incredible!|
|Dec-07-17|| ||posoo: another ULTAMUT smashing for da FORMER king NO MORE CHAMPUNSHIPS for da stookfish,|
WELCOME to da NEW ERA OF SMASHING!
|Dec-07-17|| ||posoo: LOOK imma put my QUENE over HERE HAHAHAHAH BOOM BOOM rubka-sniffers!|
|Dec-07-17|| ||rodrigoporto: Skynet, is that you ?!|
|Dec-07-17|| ||4tmac: 36. Q-K6! very nice .... kodinifish needs more than a minute a move|
|Dec-08-17|| ||QuantumMechanic: This game is maybe the most incredible of AlphaZero. |
Would be enjoyable to see a match of AlphaZero versus World Champion Carlsen assisted by computers like Stockfish and all opening books.
|Dec-08-17|| ||scholes: This should had been game of the day. Best game of the match IMHO.|
|Dec-08-17|| ||Dave12: 36.Qe6 is an amazingly powerful move. The main idea of course is the threat Qe5+. The problem is, that if the queen doesn't move, than Qe5+ Qf6 Rh7+! will do. but Qf6 for example doesn't work either- because of Qe8 Qf8 and we are back in the same problem of Qe5+. So this move puts black in the jam.|
|Dec-08-17|| ||talwnbe4: Stockfish 6 on a single core processor considers 34.. Rd8 ?? (+-1.8) as a bad mistake in less than a minute.. 34..Re5 as =, (33..Re6! another alternative =) |
Don't know what was running here in this match.
The sac 19. Re1 is sound but not a forced win after 19.. Kxh6
|Dec-08-17|| ||birimbombum: Yes. Something wrong here. Move 34 is highly suspect...|
|Dec-08-17|| ||cata008: wow it seems that a0 just seems way deeper than stockfish|
|Dec-08-17|| ||whiteshark: REQUEST ANALYSIS
click for larger view
Black to move
1) =0.00 (27 ply) 34...a6 35.Rd6 Rd8 36.Re6 Re8 37.Rd6
2) =0.00 (26 ply) 34...bxc4 35.Rd6 Rd8 36.Qh6+ Kf7 37.Rhd1 Rxd6 38.Rxd6 Ke8 39.Qxg5 Nd7 40.Rxc6 Qh7 41.Rxc4 Rb8 42.Re4+ Kf8 43.Rf4+ Ke8 44.Re4+
3) +0.58 (26 ply) 34...Re6 35.Rd8 Qxd8 36.Qxe6 Qc7 37.Qe8 Kf6 38.cxb5 cxb5 39.Qf8+ Qf7 40.Qd8+ Qe7 41.Qd5 Qc7 42.Qf3+ Kg7 43.Qxa8 b4 44.Kh2 Qc5 45.Qb7+ Kf6 46.Qf3+ Qf5 47.Qe3 Qd5 48.Qxa7 Nd7 49.Rb1 Qe4 50.Qa2 Ne5 51.Qa1
1.0 minute analysis by Stockfish 8 v270317
|Dec-09-17|| ||talwnbe4: Seems my own version of winboard and stockfish 6 was buggy ! using Arena and Stockfish 8 gave completely different evals and best lines. This 2 minutes with a single Intel core 2, 2.4 Ghz, Stockfish 8 x64 :|
34...Re6?? 35.Rd8 Qxd8 36.Qxe6 Qc7 37.Qe3 +- 4.5
34... Re5?? 35. Qh6+ Kf7 36. Rhd1 bxc4
37. Rd8 Re8 38. Qxg5 +-2.5
34.. Rd8 +- 1.5
34.. best line, black is losing, 2 minutes,
best 34.. a6 +-1.2
|Dec-10-17|| ||Eatman: For this FM this is the most impressive game by AlphaZero. It is like Tal combined with Karpov. |
Deluded Stockfish thinks it is up +1/+2 in mid 20s but Black is quite lost at move 34, so Rd8 is not as strange as it seems.
Try running Stockfish 6/7/8 on your quad core computers for about 10 minutes and enter the suggestions that are supposed to be 0.00.
You will see that 0.00 turns into +2.5+ after the thematic Rd3 is played.
In most lines(such as 34. ... a6/a5) 35. Rd3 is the killer.
Sample line given by poor Stockfish 35. ... Ra7 36. Rf3 Qh8 37. Qg4 Qxh1 and that is lost.
So in effect Alpha Zero is seeing at least 2 ply deeper and also evaluating much better.
|Dec-23-17|| ||cormier: 1) =0.00 (36 ply) 11.Nc3 d5 12.e4 Bf6 13.exd5 cxd5 14.Re1 Nba6 15.Bf4 Bc8 16.Nd4 Bb7 17.Nf5|
10.0 minute analysis by Stockfish 8
|Dec-23-17|| ||cormier: 1) =0.00 (32 ply) 12.Nc3 Bf6 13.exd5 cxd5 14.Bf4 Nba6 15.Re1 Bc8 16.Nd4 Bb7 17.Nf5|
2) =0.00 (31 ply) 12.exd5 cxd5 13.Nc3 Bf6 14.Bf4 Nba6 15.Re1 Bc8 16.Nd4 Bb7 17.Nf5
3.0 minute analysis by Stockfish 8
|Dec-23-17|| ||ChessHigherCat: What struck me was that 32.c4 seemed like a bad or at best trivial move but it was just way over my head because it prevented 34. Qd5+ and more importantly 36. Qd5+, which would have forced the trade of queens. |
Has anyone got close enough to the secret prototype of AlphaZero to notice whether it has Vulcan ears?
|Dec-24-17|| ||saffuna: 26. Qh1...yeah, that's the move I saw right away...|
|Dec-31-17|| ||duvarsiz: Move: 27.Be4
Engine: Komodo 11.2 x64 4CPU
Analyse Time: 2 hour
1.206.922 TB hits
Principal Variation: Bxe4 Qxe4 Kg8
|Dec-31-17|| ||weisyschwarz: <Eatman>, what about <Whiteshark's> line,
34...bxc4, where 35. Rd3 is not possible?|
|Jan-10-18|| ||amateur05: Playing 26. Qh1 with a piece and a pawn down ... beyond human imagination.|
|Jan-14-18|| ||Eduardo Bermudez B.: There are too many counterintuitive moves for most humans|
|Feb-03-18|| ||keypusher: <weisyschwarz: <Eatman>, what about <Whiteshark's> line, 34...bxc4, where 35. Rd3 is not possible?>|
Try 35.g4, clearing the queen's path to c3 (e.g. 35....Rd8 36.Rxd8 Qxd8 37.Qc3+ +-). Seems completely winning.
|Feb-03-18|| ||Ron: At the end of Black's 19th move, White is down a knight and a pawn. However, after Black's 19th move, the Black King is on its third rank.|
This online article about King Safety in chess programming discusses Stockfish toward the end.
Perhaps Stockfish programmers--and chess players--need to rethink King Safety. Perhaps King Safety is under-evaluated.