Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Rebel, chrisw

BrendanJNorman
Posts: 2526
Joined: Mon Feb 08, 2016 12:43 am
Full name: Brendan J Norman

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by BrendanJNorman »

Alayan wrote: Sat Apr 11, 2020 10:48 am
Dann Corbit wrote: Sat Apr 11, 2020 4:24 am By this measure, neither the Evergreen game nor the Immortal game is beautiful.
The "magnificent" move of the evergreen is a mistake, making a winning position into a draw against proper defense, so I'd say it's overrated.
Aesthetics in chess have never been defined by the computer says.

Even books listing Kasparov/Alekhine/whoever's most "brilliant" games are FULL of mistakes by computer standards.

So either we accept that humans are unable to play beautiful chess (much to the delight of the mean-spirited patzer armed with Stockfish), or we keep traditional definitions of what makes a game of chess beautiful.

And usually "beauty" in chess is defined as something like:

"A concept that grossly violates traditional principles of expectations (material considerations, positional considerations, David vs Goliath scenarios, king safety or similar)."

So when somebody sacrifices their queen and dominates the game due to the unique harmony of their minor pieces...this is beautiful.

When somebody allows ALL of their pawns to be doubled and isolated (as Fischer once did), but it simply doesn't matter due to other factors...this is beautiful.

When somebody as white runs their king right up to the e4 square in a complicated middlegame (as Karpov once did) but is unable to be mated due to the coordination of his pieces in protecting the king...this is beautiful.

When somebody in a board full of pieces is suddenly reduced to zugzwang...and any move will throw himself upon the sword (while skipping a move would be 100% fine)...this is beautiful.

The level of enemy resistance is important, but far from the most important element in deciding if a game is beautiful.

Most of Morphy's opponents were weak (weak as in, below 2000 Elo!) by today's standards, but he is regarded as one of the most brilliant players in history.

And what about Mikhail Tal? The engine laughs at a ton of his sacrifices and refutes them without breaking a sweat...this would lead the "patzer with Stockfish" to call him weak.

But we know he was an amazingly brilliant chessplayer.

It could be said that a beautiful chess game actually REQUIRES mistakes, actually.
Dann Corbit
Posts: 12538
Joined: Wed Mar 08, 2006 8:57 pm
Location: Redmond, WA USA

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by Dann Corbit »

If you were playing OTB and had never studied the game, how would you have fared?
Part of the difference is cultural also, as formerly it was ungentlemanly not to accept a sacrifice.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
BrendanJNorman
Posts: 2526
Joined: Mon Feb 08, 2016 12:43 am
Full name: Brendan J Norman

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by BrendanJNorman »

Dann Corbit wrote: Sat Apr 11, 2020 12:26 pm If you were playing OTB and had never studied the game, how would you have fared?
Many people miss this point...or ignore it deliberately.

The losing side of this beautiful game is 3400 Elo.

Anybody on this site...the strongest players, from Larry K to Me to Chris W (similar strength to me I think) to Uri (pretty strong too IIRC) to anyone (I wish I knew how strong our members were :oops: )...

We'd have all been humiliated on the white side of this game...queen or no queen.

Actually, this is another reason why training games against these guys, even weakened, is sooo fun. 8-)
Alayan
Posts: 550
Joined: Tue Nov 19, 2019 8:48 pm
Full name: Alayan Feh

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by Alayan »

BrendanJNorman wrote: Sat Apr 11, 2020 11:18 am It could be said that a beautiful chess game actually REQUIRES mistakes, actually.
It does require mistakes, but by the losing side. Mistakes by the strong side (significant ones, inaccuracies are irrelevant) can make the game worse, especially when the mistake is the supposedly brilliant move (if someone plays a sound winning queen sac, then 20 moves later in the endgame blunder into a theoretically drawn position through some deep tactic, before winning anyway ; it doesn't detract from the beauty of the sac anywhere as much as if the sac was unsound and the other side just crumbled).

I have won my fair share of blitz games against low-elo players like me where I went into strong attacks with some sacrifices. Did I like the games, did I find them nice ? Yes. But would I claim them as masterpieces ? No, because I made some really unsound moves that would have me losing against strong opposition - and the defense I faced wasn't that challenging.

If the "brilliant" move is unsound, while the game can still be beautiful to watch, it's not worthy to be called "immortal" or "masterpiece" and such.

From that perspective, this queen-for-bishop sac by Leela game against Ethereal is indeed crazy and beautiful.
Fuddur
Posts: 50
Joined: Sun Mar 18, 2018 6:35 am

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by Fuddur »

BrendanJNorman wrote: Sat Apr 11, 2020 11:18 am
Alayan wrote: Sat Apr 11, 2020 10:48 am
Dann Corbit wrote: Sat Apr 11, 2020 4:24 am By this measure, neither the Evergreen game nor the Immortal game is beautiful.
The "magnificent" move of the evergreen is a mistake, making a winning position into a draw against proper defense, so I'd say it's overrated.
Aesthetics in chess have never been defined by the computer says.

Even books listing Kasparov/Alekhine/whoever's most "brilliant" games are FULL of mistakes by computer standards.

So either we accept that humans are unable to play beautiful chess (much to the delight of the mean-spirited patzer armed with Stockfish), or we keep traditional definitions of what makes a game of chess beautiful.

And usually "beauty" in chess is defined as something like:

"A concept that grossly violates traditional principles of expectations (material considerations, positional considerations, David vs Goliath scenarios, king safety or similar)."

So when somebody sacrifices their queen and dominates the game due to the unique harmony of their minor pieces...this is beautiful.

When somebody allows ALL of their pawns to be doubled and isolated (as Fischer once did), but it simply doesn't matter due to other factors...this is beautiful.

When somebody as white runs their king right up to the e4 square in a complicated middlegame (as Karpov once did) but is unable to be mated due to the coordination of his pieces in protecting the king...this is beautiful.

When somebody in a board full of pieces is suddenly reduced to zugzwang...and any move will throw himself upon the sword (while skipping a move would be 100% fine)...this is beautiful.

The level of enemy resistance is important, but far from the most important element in deciding if a game is beautiful.

Most of Morphy's opponents were weak (weak as in, below 2000 Elo!) by today's standards, but he is regarded as one of the most brilliant players in history.

And what about Mikhail Tal? The engine laughs at a ton of his sacrifices and refutes them without breaking a sweat...this would lead the "patzer with Stockfish" to call him weak.

But we know he was an amazingly brilliant chessplayer.

It could be said that a beautiful chess game actually REQUIRES mistakes, actually.
I truly agreed with you.
Thanks
BrendanJNorman
Posts: 2526
Joined: Mon Feb 08, 2016 12:43 am
Full name: Brendan J Norman

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by BrendanJNorman »

Alayan wrote: Sat Apr 11, 2020 1:35 pm
BrendanJNorman wrote: Sat Apr 11, 2020 11:18 am It could be said that a beautiful chess game actually REQUIRES mistakes, actually.
It does require mistakes, but by the losing side. Mistakes by the strong side (significant ones, inaccuracies are irrelevant) can make the game worse, especially when the mistake is the supposedly brilliant move (if someone plays a sound winning queen sac, then 20 moves later in the endgame blunder into a theoretically drawn position through some deep tactic, before winning anyway ; it doesn't detract from the beauty of the sac anywhere as much as if the sac was unsound and the other side just crumbled).

I have won my fair share of blitz games against low-elo players like me where I went into strong attacks with some sacrifices. Did I like the games, did I find them nice ? Yes. But would I claim them as masterpieces ? No, because I made some really unsound moves that would have me losing against strong opposition - and the defense I faced wasn't that challenging.

If the "brilliant" move is unsound, while the game can still be beautiful to watch, it's not worthy to be called "immortal" or "masterpiece" and such.

From that perspective, this queen-for-bishop sac by Leela game against Ethereal is indeed crazy and beautiful.
Yep. I agree with this. I have played games online where I came up with some gorgeous concept, completely outplayed a strong opponent and won...

...only to find afterwards that in one stage of the game, there was a glaringly simple way to just induce resignation on the spot, or I allowed serious counterplay or something like this.

I always feel angry at myself, not for missing the moves, but for "spoiling" the game...whether I won or not.

To me, I felt like an artist who accidentally splashes black paint on some work he was proud of and ruins it.

Whether I sold it afterward or not, is not the point. I have never played chess for trophies or Elo or cash prizes (although I accumulated all of them at various times) but mostly for nice games and "scalps" (beating strong players).
jp
Posts: 1470
Joined: Mon Apr 23, 2018 7:54 am

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by jp »

BrendanJNorman wrote: Sat Apr 11, 2020 1:23 pm Many people miss this point...or ignore it deliberately.

The losing side of this beautiful game is 3400 Elo.
First, it's not 3400 "classical" (whatever that means) when you play it micro-bullet or whatever the TC was. And we all know that 2800 elo humans have blundered mate in 1. Are we supposed to believe that was a "beautiful" mate, just because the loser was 2800 elo? You could say the human was 2800 elo but played that one move like he was 1500 elo. People have already in this thread suggested Ethereal may have a bug that would be the equivalent of that.

What is your belief about the objective eval. of the position immediately after the queen sac?
BrendanJNorman
Posts: 2526
Joined: Mon Feb 08, 2016 12:43 am
Full name: Brendan J Norman

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by BrendanJNorman »

jp wrote: Sat Apr 11, 2020 1:47 pm
BrendanJNorman wrote: Sat Apr 11, 2020 1:23 pm Many people miss this point...or ignore it deliberately.

The losing side of this beautiful game is 3400 Elo.
First, it's not 3400 "classical" (whatever that means) when you play it micro-bullet or whatever the TC was. And we all know that 2800 elo humans have blundered mate in 1. Are we supposed to believe that was a "beautiful" mate, just because the loser was 2800 elo? You could say the human was 2800 elo but played that one move like he was 1500 elo. People have already in this thread suggested Ethereal may have a bug that would be the equivalent of that.

What is your belief about the objective eval. of the position immediately after the queen sac?
Your argument is full of holes, so I assume you must be skimming this thread.

Nevertheless, I'll lay them out.

1. The game was 3+2 which is not "micro-bullet" but the international standard time control for blitz chess. According to CCRL, Ethereal 12 is 3441 at blitz chess.

2. Nowhere did I suggest that the strength of opposition is what makes a game beautiful, I only added that this game can be considered MORE "amazing" when considering the strength of the opposition. Obviously (duh! come on!) the beautiful thing is the domination motif present in the game.

3. One person suggested a possible "bug" and then retracted the suggestion citing the time control. Regardless of their retraction, ONE person suggesting there *might* be a bug because *another* person thought Ethereal blundered (Vivien was wrong about this), doesn't in any way mean that there is any legitimacy to the claim.

You are mistaken on EVERYTHING. Try to read carefully before posting a response.
jp
Posts: 1470
Joined: Mon Apr 23, 2018 7:54 am

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by jp »

BrendanJNorman wrote: Sat Apr 11, 2020 2:00 pm 1. The game was 3+2 which is not "micro-bullet" but the international standard time control for blitz chess. According to CCRL, Ethereal 12 is 3441 at blitz chess.

2. Nowhere did I suggest that the strength of opposition is what makes a game beautiful, I only added that this game can be considered MORE "amazing" when considering the strength of the opposition. Obviously (duh! come on!) the beautiful thing is the domination motif present in the game.

3. One person suggested a possible "bug" and then retracted the suggestion citing the time control. Regardless of their retraction, ONE person suggesting there *might* be a bug because *another* person thought Ethereal blundered (Vivien was wrong about this), doesn't in any way mean that there is any legitimacy to the claim.

You are mistaken on EVERYTHING. Try to read carefully before posting a response.
1. Okay, I exaggerated, but 3441 is the blitz rating, not the classical rating. Of course, all engines or humans are much weaker at blitz TCs than at classical. Why don't you play from that position at classical TC and see what happens?


2. "Amazing" is fine. "Beautiful" depends on its correctness. (If it does turn out to be a theoretical draw, then at least that's better than being a theoretical loss for Black, which would be very, very bad.)

And just look at the rant you wrote above, which was only about the strength of the opposition. It's pretty clear what that rant was implying, even if you're now trying to walk it back.
BrendanJNorman wrote: Sat Apr 11, 2020 1:23 pm Many people miss this point...or ignore it deliberately.

The losing side of this beautiful game is 3400 Elo.

Anybody on this site...the strongest players, from Larry K to Me to Chris W (similar strength to me I think) to Uri (pretty strong too IIRC) to anyone (I wish I knew how strong our members were :oops: )...

We'd have all been humiliated on the white side of this game...queen or no queen.

Actually, this is another reason why training games against these guys, even weakened, is sooo fun. 8-)

3. Whether or not it has a bug, you have to judge the move on its objective merits, just as you have to judge the 2800 elo's simple 1-move mate blunder on its objective merits. We don't need to study the psychological reasons why a 2800 elo player blunders mate in 1 to know that it's a bad move.


What are the objective merits of the Q. sac? The best ideas we have about beauty in chess are from endgame studies. Do you think an endgame study that is cooked will win any prizes just because it has a beautiful domination motif. The endgame study composer has to make the motifs work. Otherwise, we'd all be "great" endgame composers, producing crap "endgame studies" that don't work.

BrendanJNorman wrote: Sat Apr 11, 2020 2:00 pm You are mistaken on EVERYTHING. Try to read carefully before posting a response.
No, I'm not. See above. (Do you really deny that every player is much weaker the shorter the TC, and that blitz ratings are not calibrated on the same scale as classical ratings?) Try to be more objective before posting a response. You have obviously fallen in love with this game and cannot see things objectively.
BrendanJNorman
Posts: 2526
Joined: Mon Feb 08, 2016 12:43 am
Full name: Brendan J Norman

Re: Ethereal 12 (3400) loses to God! (Most Amazing Game I've Seen)

Post by BrendanJNorman »

jp wrote: Sat Apr 11, 2020 2:19 pm
BrendanJNorman wrote: Sat Apr 11, 2020 2:00 pm 1. The game was 3+2 which is not "micro-bullet" but the international standard time control for blitz chess. According to CCRL, Ethereal 12 is 3441 at blitz chess.

2. Nowhere did I suggest that the strength of opposition is what makes a game beautiful, I only added that this game can be considered MORE "amazing" when considering the strength of the opposition. Obviously (duh! come on!) the beautiful thing is the domination motif present in the game.

3. One person suggested a possible "bug" and then retracted the suggestion citing the time control. Regardless of their retraction, ONE person suggesting there *might* be a bug because *another* person thought Ethereal blundered (Vivien was wrong about this), doesn't in any way mean that there is any legitimacy to the claim.

You are mistaken on EVERYTHING. Try to read carefully before posting a response.
1. Okay, I exaggerated, but 3441 is the blitz rating, not the classical rating. Of course, all engines or humans are much weaker at blitz TCs than at classical. Why don't you play from that position at classical TC and see what happens?
2. "Amazing" is fine. "Beautiful" depends on its correctness. (If it does turn out to be a theoretical draw, then at least that's better than being a theoretical loss for Black, which would be very, very bad.)
3. Whether or not it has a bug, you have to judge the move on its objective merits, just as you have to judge the 2800 elo's simple 1-move mate blunder on its objective merits. We don't need to study the psychological reasons why a 2800 elo player blunders mate in 1 to know that it's a bad move.


What are the objective merits of the Q. sac? The best ideas we have about beauty in chess are from endgame studies. Do you think an endgame study that is cooked will win any prizes just because it has a beautiful domination motif. The endgame study composer has to make the motifs work. Otherwise, we'd all be "great" endgame composers, producing crap "endgame studies" that don't work.
1. I already ran a gauntlet with Stockfish 11 playing as white vs various lc0 networks at long time controls. All games were drawn, despite ridiculous material imbalance.

2. All real chess players worldwide...from any culture or language would call black's idea beautiful... Anyone else is just a zealous contrarian. If I said "look at how stupid lc0 blundered its queen and Stockfish still couldn't win" you'd probably still be disagreeing out of habit. Truth is, If Alekhine or Tal could be resurrected and see such chess being played, they would be GREATLY inspired and impressed...despite being praised as the greatest combinational geniuses in history. Pettiness about the meaning of certain adjectives (beautiful, amazing, whatever) notwithstanding.

3. Kramnik blundered into a mate in 1 against a computer. This is not a rule amongst elite players, so stop pretending it is. You are using an exception to prove a rule with an unsound foundation. Regardless, you have created a strawman here...there is no bug and Ethereal's move (which Vivien proposed to be a blunder) was not a blunder.

4. Many of the old school studies have been refuted by modern engines, but this in no way takes away from the beauty of the "key" contained therein.

Computer nerds would snicker through their bucked teeth, but real chessplayers don't budge. Beauty is beauty.

It seems to me that you (and 80-90% of this site) just like to argue (and shit on others and be "contrarian") just for the sake of it. I prefer to enjoy my passion for chess and share nice things I come across with like-minded people. If you just want to shit on nice ideas to boost your own feeling of significance, then that's your own issue buddy.