Review of Game Changer

Game Changer: AlphaZero’s Groundbreaking Chess Strategies and the Promise of AI by Matthew Sadler & Natasha Regan 2019 New in Chess 416pp

In December of 2017 DeepMind released a paper showing that their self-learning AI, AlphaZero, had defeated the powerful and popular engine Stockfish in a 100 game match by what seemed to be an inconceivable score of 28 wins, 72 draws, and no losses.

However, a deeper look showed that the terms of the match were deeply flawed. The playing field was nowhere near level, and so as many people in the chess world went all agog at the results, I was in the small group of non-believers. Yes, it was impressive *how* AlphaZero played, making speculative sacrifices, etc. but as my friend Hikaru Nakamura said in an interview “I don’t necessarily put a lot of credibility in the results simply because my understanding is that AlphaZero is basically using the Google supercomputer and Stockfish doesn’t run on that hardware; Stockfish was basically running on what would be my laptop. If you wanna have a match that’s comparable you have to have Stockfish running on a supercomputer as well.”

And as far as I was concerned that was that.

Then, a few weeks ago I was listening to Ben Johnson’s excellent podcast Perpetual Chess. The authors of this book were on and one of the things that they mentioned early on was that the DeepMind staff was also receptive to the criticism and as a result they had set up a second match.

As Wikipedia notes about the second match “In the final results, Stockfish ran under the same conditions as in the TCEC superfinal: 44 CPU cores, Syzygy endgame tablebases, and a 32GB hash size. Instead of a fixed time control of one move per minute, both engines were given 3 hours plus 15 seconds per move to finish the game. The version of Stockfish used was version 8. AlphaZero won with a score of 155 wins to 6 losses, with the rest drawn. DeepMind also played a series of games using the TCEC opening positions. AlphaZero won 95 out of the 100 mini-matches from these positions.”

Suddenly I found my interest in AlphaZero piqued. Perhaps there was more to this after all.

Then, like a beam of light shot straight into my soul I come home one day to find the book Game Changer in my mailbox.

The content is laid out in eighteen chapters in five parts.

Part I AlphaZero’s history

Chapter 1 A quick tour of computer chess competition

Chapter 2 ZeroZeroZero

Chapter 3 Demis Hassabis, DeepMind and AI

Part II Inside the box

Chapter 4 How AlphaZero thinks

Chapter 5 AlphaZero’s style – meeting in the middle

Part III Themes in AlphaZero’s play

Chapter 6 Introduction to our selected AlphaZero themes

Chapter 7 Piece mobility: outpost

Chapter 8 Piece mobility: activity

Chapter 9 Attacking the king: the march of the rook’s pawn

Chapter 10 Attacking the king: colour complexes

Chapter 11 Attacking the king: sacrifices for time, space and damage

Chapter 12 Attacking the king: opposite-side castling

Chapter 13 Attacking the king: defense

Part IV AlphaZero’s opening choices

Chapter 14 AlphaZero’s opening repertoire

Chapter 15 The King’s Indian Samisch

Chapter 16 The Carlsbad

Part V Conclusion

Chapter 17 Epilogue

Chapter 18 Technical note

After playing through the games of AlphaZero I really can’t rave about the games enough.

In addition to the book, the authors have created a YouTube channel where they only include games not otherwise included in the book.

The truth about the games of AlphaZero is that they are amazing works of art in so many cases.

Rather than listening to me ramble on, look at this game from the YouTube channel.

Then listen to the authors on Perpetual Chess:

Then go buy this book.  You won’t regret it.

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Too Much Garry? No Such Thing!

So I come across this position in Timman’s latest book The Longest Game about the five Kasparov-Karpov matches from 1984-1990.

Here Garry plays 13.Bf4 with Timman noting “Against Marjanovic in La Valetta 1980, Kasparov had played 13.exd5 and won convincingly.”

I was wondering about that game and so I looked it up. Wow…just wow.

The blunder happens in this position:

The threat of course is 17.Nh6+ with the discovered attack on the queen. This game is the stem game in this line, and Marjanovic chooses the worst way to deal with the threat by moving his king to h8, after which his position implodes since the knight on c3 is able to come into the attack via e4 with tempo since the queen is unguarded on d7.

An interesting factoid here is that this game also seems to be a possible example of who was staying current in chess literature at the time and who wasn’t due to the very next game in this line, which took place the following year in Buenos Aires between Argentinian IM Raimundo Garcia and Columbian Augusto Pereira.  I can find ratings in the high 2200’s for Pereira so it seems likely he was close to FM strength although he never got the title.

The two games are the same through White’s 19th move:

Here Pereira deviated with 19…Qc5 rather than 19…Qf6. No matter, he still lost quickly.

The reason for my comment about staying current in literature is that back in these days there were no databases and it could be hard to find recent games. Those players who were better at it often had an advantage over those who weren’t.

The Kasparov game had been published in Informant 30, but unless players took the time to truly read and digest the Informants they would often be at the mercy of their better prepared opponents.

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Modern Advantages

Currently I’m reading Jan Timman’s latest book The Longest Game in preparation for a review.

I’ve long spoken about certain modern advantages about buying books that exist today, but didn’t 30 years ago when I first was playing in tournaments. Chief among them the fact that back in the late 80’s you’d see a book title in a catalog, and maybe a two-line description and that’s what you needed to use to decide whether or not to buy something.

These days you can go online and read excerpts along with finding numerous reviews, etc. This is why it’s generally agreed that we’re in a Golden Age for chess books.

However, there’s a huge advantage these days in the reading of chess books which I’d like to talk about a little today.

30 years ago books were the primary source of learning and improvement. Today they are just the springboard.

Let’s take a look at this position from the second game of the first K-K match.

Here Kasparov plays 8.Nh4 about which Timman says “The critical move, which had been introduced into practice by Polugaevsky in 1980. Before that, 8.Nd4 had been the usual move. However, in that case, Black can return the pawn by 8…Nc6 (Korchnoi’s 8…Bc6 is also possible, intending 9.Nxc6 dxc6) 9.cxd5 Nxd4 10.Qxd4 c5 11.Qd2 d6 with a playable Benoni position.”

This is the final position from that note above:

I sat there looking at that and asking myself “But why 11.Qd2 in that position?  Why block in the c1 bishop like that?” And that takes us to the point I want to make today.

Back in the times in which this game was played, when a club player would see a note like that from a GM commentating on a game it didn’t matter if we understood it or not. If you couldn’t reason out the idea behind the move then you just had to move on. You were also limited to the idea fed to you by the annotator of the game. Whatever they decided to mention as alternatives is what you got to look at. The end.

These days we’re not so hampered.  I simply opened ChessBase and put in the moves to take us to this position before 11.Qd2

Here I’m able to do a number of things. First, I run Stockfish 10 which tells me after a depth of 36 that 11.Qd2 is 0.64, 11.Qd3 is 0.60, and 11.Qd1 is 0.60.  OK, so this shows that there’s nothing inherently wrong with putting the queen on d2. My instinct of “but the bishop” is a knee jerk one, but in this day and age I can learn why.

So I go to the reference tab in my database. I can see that in the 8.5 million games in my database this position was reached 19 times and that there have been four moves played. 11.Qd3 has been played twelve times (last played in 1980), 11.Qd2 five times (last played in 1980), and 11.dxc6 (in 1978) and 11.Qd1 (in 1966) once each.

So now we can look at the games in the 11.Qd2 line. Of the five, four are GM games with one taking place between amateurs. If we disregard the amateur game and look at the four games we see that interestingly, Yuri Balashov was white in three of them, with the aforementioned Korchnoi having white in the other game. The players of the black bits were Timman, Inkiov, Furman, and Karpov.

Since Timman mentions Korchnoi, and since it’s the stem game,  let’s look at that game first.

Now let’s look at the four Balashov Whites in chronological order:

So in looking through those games it becomes easy to see that the idea for White was simply to play b3 and then develop the bishop that way. Therefore the queen being on d2 was not a detriment at all.

It’s also interesting to see that Korchnoi’s plan of b3-a4-Rb1-Ba3-b4 was determined to be inferior by Balashov who then developed the bishop to b2 instead. The engine agrees with the analysis as well.

I’m often amused at the view espoused by some that books are somehow outdated due to the fact that we have so many technological tools at our disposal. To me, books are enhanced rather than downgraded by these advances of the modern age.

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Tough Game, Good Lesson

I played a game tonight where first I built up a good position:

Here I found the nice move 19.Nd5. I probably could have played it a move or two earlier as well.

Then I wind up with this position:

Here I play 28.e5? I completely missed that 28.Rxf6 is resigns as the rook can’t be taken since it leads to mate. 28…gxf6 29.Qxf6+ Qg7 30.Qxd8+ Qg8 31.Qxg8#

Naturally I saw it during the game after I didn’t play it.

This weighed very heavy on my mind, but I was able to do something which I haven’t always been able to do in the past and continue to grind and try to convert.

I missed a few other winning ideas along the way, but played until this position was reached:

So the result certainly didn’t go the way I wanted it to, but the fact that I continued to press instead of offering a draw out of disgust seems to me like a good sign.

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Crush by Petrosian

Earlier today I saw my friend Elshan ask on Facebook for games with Tigran Petrosian on the White side of the King’s Indian. GM Jacob Aagaard just posted “Qh8+” as a response.

That sent me on a mission, and this is where I wound up…the final position from Petrosian – Spassky in Game 10 of the 1966 World Championship.

Here is the entire game. Thank you Elshan and Jacob for getting me to look for this!

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Don’t Ignore the Center Larsen – Spassky 0-1

When they think of Bent Larsen, most American’s probably think “Isn’t he that guy who lost to Fischer 6-0?”

However, there is a lot more to Bent than that footnote would suggest.  He was clearly the second best player in the West for many years behind only Fischer.

In fact, in 1970 during the first “USSR vs The Rest of the World” match, the captain of the World Team, former world champion Max Euwe, had decided that he would use Elo ratings to determine board order.

Bent Larsen would not accept this as Fischer had been inactive for some time at this point and he himself had had several successes leading up to this event.

In a gesture that surprised many, Fischer agreed and stepped down to play Tigran Petrosian on Board Two while Larsen faced world champion Boris Spassky on Board One.

After a draw in their first game, Larsen sat down to the board for Game Two and promptly committed a cardinal sin in the chess world by completely ignoring the center in the second game. This allowed Spassky to create the following miniature.

Take this as a lesson…ignore the center at your own peril!

And yes, these days the Nimzo-Larsen attack (1.b3 or 1.Nf3 2.b3) is often used by strong players, but these days the theory and understanding are quite far advanced compared to where they were when Bent was pioneering this setup.

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Shameless Request for Donations

It’s not often that I make an actual shameless request for donations.  Sure, I have the Patreon thing at the bottom of each post, but this is not that.

It costs me around $170 a year to maintain this blog, and the truth is that I really enjoy doing it.

I pay for three years at a time, and it’s coming due in a few days.  If you read this content and feel that you get something out of it, and if you can spare a few bucks I’d sincerely appreciate a PayPal donation.  $1, $2, $10, whatever.

It all adds up and it’s all extremely helpful.

So if you can’t afford to donate, please don’t.  But if you can, then this would be the time to do so.

All help is gratefully appreciated.

Thank you from the bottom of my heart,

Chris Wainscott

Positional Chess Books by Sakaev & Landa

For those of you who may not be familiar with these books, let me start by telling you that they are very well done.

They’re far different than most books I’ve seen on positional chess. One of the main reasons for this is that rather than breaking the material up into the expected chapters of open files, outposts, strong squares, etc. the two volumes are split with Volume One covering openings and middlegames, and Volume Two focusing on structures and dynamics. Yes, dynamics. After all, as Botvinnik noted, “tactics are the servant of strategy.” Which is a way of saying that positional play is often aimed at gaining a large enough advantage to crash through tactically.

This post is not a review. If you would like to see my review of Volume Two of this set you may click here.

If you would like to read a review from several years ago on my original blog about a Shirov DVD showing how tactics crown positional advantages, then click here.

So if this post is not a review, then what is it?  I’d say it’s a call to action. You should purchase these books if you enjoy well presented material.

For instance, in Chapter One of Volume One the theme is “A Lead in Development.” There are five well annotated games which are shown first. I present – without annotation since that would take forever to copy from the book – four of those five. The fifth is a training game of Sakaev’s and so is not in my database.

Tal – Uhlmann 1971

Tal – Toran 1961

Tal – Petrosian 1974

Seirawan – Karpov 1982

In this game Yasser fell behind in development and Karpov steamrolled him.

Then at the end of the chapter there is a section on “additional material” which simply lists a handful of games for the reader to analyze and review on their own.  Those games are:

Keres – Botvinnik 1941

Svidler – Dreev 1997

Kasparov – Polugaevsky 1978

Karpov – Karpov 1993

So when all is said and done you’ve have five games with detailed annotations and another four to review on your own.

I really enjoy books like this with that added element of “go do more work on your own” to them. Almost as much as I love the Soviet School of Chess.

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Right Idea, Wrong Execution

Today in a game I had this position as White:

I came up with the correct idea of sacking the exchange with 15.Rb4. Then after 15…Nxb4 16.axb4 Qb5 17.Rxa7 Nc6 18.Ra3 my opponent played 18…fxe5

Here I misplayed the position by taking back on e5 with 19.dxe5. However, had I played 19.Qa1! instead I would have had a sizeable advantage.

19…Kb7 20.Nxe5 Nxe5 21.Ra7+ Kc6 and now the nice shot 22.Bf4.

White is clearly better here.

Having said all that, I’m ok with this game since I correctly sacked the exchange.

Now if I can just get my strategic thinking and calculation up to par I’ll convert these as well.

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.

Very Nice Exchange Sac

I was looking at this game earlier between Yair Kraidman (whom I have never heard of) and Heikki Westerinen (whom of course I have heard of) and this is the position after White’s 16th move 16.Bb5

I’ll be the first to admit I’m not great with understanding the majority of the exchange sacrifices I see.  In this case I took a lot of time trying to puzzle it out, and as near as I can figure out the key here is the position after 22…Rxb2.  I think that must be the position that Westerinen saw when he went for it.

Of course I have no real idea, but if a strong player wants to throw me a bone here and let me know if I’m on the right track that would be awesome.

Here is the complete game:

Til Next Time,

Chris Wainscott

If you like this blog, please consider becoming a Patreon supporter. Any money I raise will go towards lessons and stronger tournaments.

If you can spare it, please click here and become a supporter. Even $1 a month can help me achieve my dream.