"...involved exactly 3 of the key elements identified by Sandler and Regan as key to the successful attacking technique of AlphaZero against a castled King--namely, the combination of:

1. controlled diagonal
2. controlled file
3. push of h pawn."

As fully explained with examples dating back to Greco by Vukovic in 'The Art of Attack in Chess.' (1965), he too took a lot of his material from earlier writers on the subject, Tarrasch, Spielmann and Alekhine...

Chapter 7 "Ranks, Files and Diagonals in the attack on the Castled King."

Alpha Zero, and it's fans, (a machine with fans...good grief) some think this tin can has revolutionised chess. It is simply re-inventing the wheel - it's added a few new spokes, I'll give it that, but we cannot allow it to remove the saddle and handle bars. We are still in the seat and steer chess. It's a human v human game.

AlphaZero is a chessplayer. Questions of hardware and configuration aside, it's arguably the strongest chessplayer on earth, certainly far stronger than Carlsen or any human, and it plays beautiful chess. It's never hijacked a national chess federation, or dodged an opponent, or accused the Jews of trying to take over the world. It's also a fascinating application of reinforcement learning. Why shouldn't it have fans?

<As fully explained with examples dating back to Greco by Vukovic in 'The Art of Attack in Chess.' (1965), he too took a lot of his material from earlier writers on the subject, Tarrasch, Spielmann and Alekhine...>

Since there is nothing completely new under the sun, this sort of argument can easily be used to claim that no one has learned from anyone post-Morphy. But the style of the strongest players always tends to influence everyone else. Chessplayers were surprised, in my recollection, when Kasparov came along and showed that Karpov's style was not the only style that could succeed at the very top. When Stockfish and other engines seemed to show that nearly any position could be defended, that had an effect on how humans played the game. Now that AlphaZero has come along and shown that even Stockfish can be overcome with aggressive, dynamic play, that's also having an effect.

I think some over-the-top claims of its influence on Carlsen have been made, but he does seem to be playing rather differently this year, and the change has been welcomed by everyone, except his opponents. Has A0 had an effect on him? Yeah, I think so.

<It's a human v human game.>

Hi K.P.

(here we go again, usually we debate in a civil manner, let's keep it so.)

Not afraid, just bemused at the glorification being bestowed on a computer program which does not even know it's playing a game.

First this bit:

"It's never hijacked a national chess federation, or dodged an opponent...etc."

You are correct there, computers don't hurt, cheat or have any racial discrimination - only people have that particular gift.

"...AlphaZero has come along and shown that even Stockfish can be overcome with aggressive, dynamic play."

I howled like a lost dog in the wilderness for years decrying the faults in Stockfish. Alpha is highlighting these defects.

Thank you Alpha. I'm a human with a ton of baggage, the last three letters 'age'. being my biggest suitcase.

(I'm 68 on June 30th K.P. - I'll be touring in the desolate blissful peaceful wastes of north Scotland - but you can still wish me a Happy Birthday.)

Where was I?....Oh yes thanking Alpha because it can calculate deeper without tiring, without worrying about winning or losing and frying S.F's chips (fish and chips...geddit...).

"... no one has learned from anyone post-Morphy."

We could go back to Greco. Every combo played in one form or another can be found in the manuscripts of Greco.

Greco's analysis is dodgy (all the games are made up) but the idea behind the combo's are sound.

Later players came along used these combo's in different settings and then figured out and taught how to set them up by looking for landmarks in positions.

As decades passed the level of the game rose higher and higher not only opening analysis but in the areas of positional play, endgame theory and dynamic play.

Alpha is showing us nothing new, it's been infront of us all the time. This Alpha tool is simply blowing fresh (and welcome) air.

I cannot credit Carlsen's amazing run in the past six months down to Alpha. (I think we agree there) He is a genius who blows hot and cold.

His cold makes him a standard good GM.and admits when he blows cold he's been lucky. When he is hot he is magnificent.

"....and it plays beautiful chess."

Beauty is in the eye of the beholder, if you like what it does, then so be it. Afraid I cannot get excited about a machine beating another machine (Stockfish) which I know was flawed.

Need to see Alpha playing a 3rd choice move in a lost position going for a swindle. That is human v human chess. (our game) playing v human foibles.

That to me from a computer would be, for want of another word, 'beautiful' because it's not the move but the idea behind it I've always found beautiful and stimulating, Alpha does not have ideas.

Reply soon mate, at noon tomorrow the mobile is switched off and I'm off planet till the end of next week.

Anyway, I've pointed out Stockfish vs AlphaZero, 2018 a few times. That's as close as a computer can come to playing as badly as you want it to.

<Beauty is in the eye of the beholder, if you like what it does, then so be it. Afraid I cannot get excited about a machine beating another machine (Stockfish) which I know was flawed.>

You should be ashamed to write something so head-poundingly dumb.

Never said **** traps, I'm talking about a swindle based on a two move trick when a player realises he going down and switches into rat mode.

You can set traps when you are winning, level or losing and you can have such a thing as coincidental trap.

I want it to realise it is losing and go for a roll of the dice. A swindle, which is a very important part of human - human chess. It even has it's wiki page.

https://en.wikipedia.org/wiki/Swind...

Never seen the game you linked to, Thanks. I'll reply there as it is more relevant. Stockfish vs AlphaZero, 2018.

A lot of people do not like Giri. That is because he talks too much.

During my playing career, I met two world champions and got to play a number of GMs. The titleholders were both regarded as likeable in their day and did not disappoint in that regard.

Some of the worst offenders in this way in my experience have been averagely good (1800-2000) players who, as you say, had overblown egos. For the most part, the stronger players I have known were well aware they could play and had no need to advertise.

In the 1984 tournament book of the Canadian Closed, Danny Kopec wrote appreciatively in his thumbnail sketch of Kevin Spraggett, noting that he was one instance of a strong player who did not come it over on other players at all, whereas this was not necessarily the case with other participants in that zonal.

<...Giri is still too young to do that but he might learn that in time though some older guys (Kramnik, Kasparov ...) never did learn that. :-)>

Hi FSR,

It is a very good article.

https://en.wikipedia.org/wiki/Swind...

I'v often used and preached option 2.

(2) Do not be afraid of losing. "Once you've accepted that your position is lost, you should be in a position of psychological strength."

Sometimes the worst thing you can have is a technically won game. I used to play the Latvian with the idea of getting a lost (but very active) position ASAP.

The knack is, as you say, don't be afraid of losing. It's only a game. Enjoy yourself.

<Sally Simpson: ***

...

(2) Do not be afraid of losing. ...>

Well, if you are afraid of losing, just do not play.

<Sometimes the worst thing you can have is a technically won game.>

Well, paraphrasing Tarrasch, I suppose.

<I used to play the Latvian with the idea of getting a lost (but very active) position ASAP. >

Hi, J.B.

Score with Latvian was good. Won a lot more than I lost, very few draws but funnily enough got two draws v Leonard Shamkovich and Jacob Aagaard in simuls with the Latvian. I was all ready to play it v Adorjan in 1974 but he opened 1.d4. Pity I may have got another mention in the Black is OK series.

Recall a Scottish Chess magazine from the late 70's saying the juniors are playing the Latvian and losing and blaming me for starting a craze. (much like everyone did for a few months after Miles did his 1...a6 v Karpov. The copycats got walloped and that craze soon died out.)

Strangly enough never scored too well with the pure Latvian in reverse. (about 50-50) 1.e4 Nf6 2,f3(!?) e5.3.f4!

click for larger view

One of my efforts there does get a mention in the 'Black is OK' series, a loss, G Chandler vs J B Henderson, 1994 (another J.B.!)

"Do not be afraid of losing...."

Is not a pre-game feeling it's when you finally admit to yourself you have cocked things up during a game. (in the above game my failure to play Kb1 because I thought I could save a tempo - some you win...some you lose.)

So go for a trick/trap or two, might as well, you are lost, no pressure, no fear. Hence my last throw of the dice from the above game here...

click for larger view

I played 32.Rxd4 I recall seeing 35...b4 coming when I played it and thinking how unfair is that? But I won/swindled quite a lot of lost games. I don't mind giving a few back if a swindle fails, the important thing is I spotted it and tried it. I've won from a lot worse positions. Never underestimate the human ability to find the losing move from a choice of good moves. (another Tarrasch paraphrase.)

Do you/did you have a pet line that scored well.

Yesterdays 'Guardian'.

"...but it is clear he [Carlsen] derives more satisfaction from the way he mentally broke down Giri again a month later in 23 moves.

He surprised me in the opening. And I just decided that I was going to play something very dubious since I knew that was going to be unpleasant for him psychologically – in other words I had confidence to play the man and not the position,” he says.

“And I could see him for the next hour just agonising,”

he continues with growing relish.

“He realised that his position was probably good but it would be very hard to prove it. So I was still in the driver’s seat, because I was not afraid and he was. I didn’t really care about the result – I was just making a very good decision psychologically and it showed.”