Depth and Beauty in Pervakov’s Endgame Studies

by Nagesh Havanur
5/6/2019 – Last time our columnist wrote on the life and career of Oleg Pervakov. This week he offers a glimpse of the eminent composer’s work from a recently published collection of endgame studies. He also sets a challenge for readers to solve three of them. Your move! | Photo:

ChessBase 15 - Mega package ChessBase 15 - Mega package

Find the right combination! ChessBase 15 program + new Mega Database 2019 with 7.6 million games and more than 70,000 master analyses. Plus ChessBase Magazine (DVD + magazine) and CB Premium membership for 1 year!


How to lose a book

A friend of mine and I were having a cup of tea.

“A new restaurant has opened on our street. The other day my wife and I visited the place. Elegant place and nice décor. One couldn’t take one’s eyes off the wall. Such fine paintings! But it took so long to place an order and the main course didn’t arrive at all….”

book cover

I burst out laughing. “I know what you are hinting at!”

“Do you? You write a long piece on Pervakov and there is not one study by him.  So where are they?”

“Patience is everything. Have you seen this book?"

Now that was a mistake. My friend is a Pervakov fan. The moment he saw the book, he seized it with both hands and leapt from his seat. I was stunned.

“What are you doing?”

“I have been waiting for a book like this. Thank you any way!”

In two bounds he was at the door.

“Hello! You can’t do that. It’s my book.”

I protested.

My friend was grinning. In a moment he was gone.

So you see, I no longer have the book. I shall be relying on my notes to tell readers what it’s about.  This work is a collection of 100 Pervakov studies composed during the period, 1983-2018. A few are joint compositions with fellow-authors like Karen Sumbatyan, Mikhail Gromov and Martin Minski.

A bit of nitpicking

PervakovBefore I offer a glimpse of these studies let me mention a few issues, minor as they may be. The introduction claims, Pervakov has composed nearly 500 studies. Not really. It is a little more than 350. While the first 86 studies are placed in strict chronological order, the remaining 14 (composed during 2016-2018) are all mixed and follow no particular order. The title is also a bit of misnomer and it carries an uncanny echo of the Soviet era. This book is anything but that.

Similarly, the subtitle alluding to the author has an element of hyperbole. Tkachenko was a member of the Ukranian Team that won the World Chess Composition Tournament in 1997. He wasn’t in the race for the individual world championship, though. In fairness to him it may be mentioned that he has the rare distinction of being both a fine composer and a chess historian.

His analysis of studies in this book is precise and presentation lucid. He has found many astonishing and beautiful lines that otherwise would not be known.


Are you all ready?

So enough of nitpicking. Let us look at these studies. Time and again Pervakov works wonders over the board.  He can offer you a position with very little material and weave magic on those squares. Take a look at the following diagram. The lone White king can hardly stop Black pawns from queening. What would you advise him to do?


You'll find the solutions to the problems at the end of this article

How was it? If you solved it at first go, congrats. If not, you are in good company. A friend of mine pushed the pawn on d4, and to his shock Black’s f-pawn raced to queen. Cursing himself, he first played 1.♔xf4 and then found there was no way of stopping the b-pawn. These pawns are naughty chaps!

The second study looks simple. Here White needs to promote his h-pawn to queen. However, two Black pawns are in the way and their king is preparing to reach h8 himself. Will the White king and bishop still make it? Your move.


Did you do that? I showed it to another friend of mine. “It’s all very deep and subtle, but hard for my grey cells,” he mumbled. Then he asked, “Why don’t you give me a fun position?”

So I picked up this one from the book and his eyes lit up. “Ha! I know what to do. I just play 1.b7-b8=Q and it’s over. Both the Black king and the rook are in danger. Either he loses the rook or he is mated. That’s it.”

As you will see, it all happens, but differently!


Did you like that? Not the Black monarch, though. He thought he was winning!

By the way, did any one get all the three right?  Please come forward. You deserve an applause. Now three tests are too many for one evening. Shall we take a break? Let us sit back and relax….

OK, are you all ready? Here we go again. The next two are a bit “crazy”. So you don’t have to exert yourself if you don’t wish to. You can give each a reasonable try and see how it goes.

The first one is a real slugfest.  White is on the way to have a second queen. However, his immediate concern is 1…♛xa5 mate. For the same reason he has no time for 1.♕xc1.

What should he do?


How was that? The waltz with the queens and rooks makes a stunning impression.

We have come to the end of our journey. In this last position both sides are fighting on the edge of precipice. White has played ambitiously and the monarch has reached as far as h6. However, his knight is pinned and the queen is under attack. So he needs to gain initiative.

Now who should be solving this? Only those who have still retained their sanity after checking out the last one! The solution is a bit long. Go as far as you can. Stop when you hit a road block. Let the composer show you the way.


I did show this position to a club player, and he nearly fell off the chair when he saw the first move. But you, dear reader are smart and with a bit of imagination, can find it. It’s the follow-up that has more surprises in store.

I spent days poring over the studies in this book. It’s hard for me to categorize Pervakov’s style. If you look at the purity of conception, economy of material and precise execution in the first two studies shown here, you feel, his work is right in the line of Grigoriev and Kasparian. But then he can also transport you to the realm of fantasy like Korolkov and Mitrofanov as we have seen in the last two studies here. So take your pick and happy hunting!

P.S.: This versatile composer continues to reach great heights. Recently he won the Raaphy Persitz Memorial Tourney (2019) with a star-studded field (Yuri Bazlov, Martin Minski, Yochanen Afek and Jan Timman competing):


1) There are 280 studies of Oleg Pervakov in Harold Heijden’s Database

HHdbV for the period, 1983-2015.

This book has 20 studies from recent years, 2016-2018.

2) More information on the book may be found at New in Chess.


The five positions above with their solutions in the order presented... 


Click or tap a game in the list to switch

Prof. Nagesh Havanur (otherwise known as "chessbibliophile") is a senior academic and research scholar. He taught English in Mumbai for three decades and has now settled in Bangalore, India. His interests include chess history, biography and opening theory. He has been writing on the Royal Game for more than two decades. His articles and reviews have appeared on several web sites and magazines.
Discussion and Feedback Join the public discussion or submit your feedback to the editors


Rules for reader comments


Not registered yet? Register

Dr Zeiss Dr Zeiss 5/13/2019 09:06
Mea culpa. In my analysis I had the white h-pawn on h2 instead of the correct h3. The lesson is that in chess even the tiniest difference can be decisive.
chessbibliophile chessbibliophile 5/10/2019 11:18
Some times is more important to ask questions than accept pat answers. You have just done that and I am glad to see your interest even if your lines on occasion are wrong. But then who is perfect? We all live and learn.
About computers in chess composition: Today no composer presents a problem or endgame study in a competition without cross-checking with databases and engines. Thereafter it is submitted to a jury of fellow composers who also look for possible duals, cooks before confirming authenticity and quality. There are rigorous standards of excellence in chess composition. Nevertheless, when it comes to tries and supplementary lines there is a possibility of human error. So one should verify.
Your analytical efforts are appreciated and you are always welcome.
chessbibliophile chessbibliophile 5/10/2019 11:14
For readers: The 1996 study by Pervakov here holds quite a few mysteries.
Here is summing up after all the discussion.
1.Rf7+ wins. With best play both sides 1.Rg8+ can only draw. Readers may check the following analysis.
Supplementary lines are marked (a) and (b).
1.Rg8+?! Kxg8 2.e8Q+ Kh7! (a) 3.Qh8+! (b) Kxh8 4.Qh6+ Kg8 5.Qxg6+= with perpetual check.
a)2...Kg7? 3.Qxd5 Qxa5+ 4.Qxa5 Rg1 5.Qxc3 Rg3 6.Qxg6+ Rxg6 7.Qxc2 Soon White wins by shepherding his pawns for queening.
b) Black has two tries here:
I.) 3.Qxc3? Rg1! Now White can capture the c-pawn or rush the queen to the defence of the king.
If 4.Qxc2 4...Qxa5+ 5.Qea4 Rg3+ 6.Qcb3 Qc5+ 7.Kb2 Qc3+ 8.Kb1 Rg1 mate.
Or 4.Qa4 c1Q+ 5.Qxc1 Qd3+ 6.Qb3 Rxb3+ 7.axb3 Rxc1 and Black wins.
II)3.Qxd5? Qxa5+! 4)Qxa5 Rf1!! 5)Qxc3 Rf3! An important line given in the book.
Kapawner Kapawner 5/9/2019 06:05
@Dr Zeiss, 4.Qh3 is impossible because the white pawn is on h3.
Dr Zeiss Dr Zeiss 5/9/2019 05:45
@Kapawner, Not so I am afraid, 3.-Rg1? 4.Qh3+ Kg7 5.Qh3-h8 checkmate. It seems that the composer himself overlooked this 1.Rg8+! dual. Perhaps he did not check with computer.
Kapawner Kapawner 5/9/2019 05:32
After 3.Qxc3? Black wins 3...Rg1! (instead of 3...Qxa5?) given by chessbibliophile.
Dr Zeiss Dr Zeiss 5/9/2019 04:59
@Kapawner, It seems that the variations are easy to confuse if not given in full. The cold irrefutable logic of the computer shows that 1.Rg8+! Kxg8 2.e8Q+ Kh7 3.Qxc3! Qxa5+ 4.Qxa5 and white is winning, e.g. 4.-Rf1 5.Qd2! (or Qc5! as I already mentioned) and the line continues as given below. Or some other rook move by black but white can avoid perpetual by giving one queen for a rook. After 5.Qd2! computer evalution is over +10 for white.
Kapawner Kapawner 5/9/2019 04:04
1.Rg8+? Kxg8 2.e8Q+ Kh7! is the only refutation after 3.Qxd5! (instead of 3.Qxc3?) Qxa5+! 4.Qxa5 Rf1! 5.Qxc3 Rf3! That makes the study and its idea correct. That is the solution from a reliable source. I don't know what is given in the book by Tkachenko.
Kapawner Kapawner 5/9/2019 11:44
Dr Zeiss, place the black pawn c3 on the board and check your lines again. The study is correct. Rf7+! wins Rg8+? does not.
Dr Zeiss Dr Zeiss 5/9/2019 08:19
@chessbibliophile, One more correction. After 4.-Rf1 Fritz gives either 5.Qd2! or 5.Qc5! and white wins. E.g. 5.-c1Q 6.Qxc1 Rxc1 7.d7 Rc3+ 8.Ka4 Rc6 9.Ka5 when white gets a new queen and can sacrifice one queen for rook to avoid perpetual if needed.
chessbibliophile chessbibliophile 5/9/2019 04:24
@ILAN RUBIN,That's elegantly expressed. For the non-Russian reader, партия has two meanings, game and party. It’s part of the play on words here. There was an entertaining discussion among composers on this subject in Sergiy Didukh’s blog. Steffen Nielsen, a fellow composer wrote, “According to my wife a game of chess is the exact opposite of a party!” It is not a party for the loser, for sure!
On the new Russian book (not the one reviewed here) opinions differ.
Didukh, a composer in his own right and a stern critic wrote, “The book contains many nice recent studies (most of them are by Pervakov) and will surely be liked by readers. As for me I don’t like the book very much because there isn’t any system or order in the presentation of the studies. They are just mixed up with positions from games showing more or less similar ideas. But of course, the intention was to show common things in studies and games and not to please a researching composer.”
ILAN RUBIN ILAN RUBIN 5/8/2019 07:19
Этюд и партия едины is a play on words, based on the Soviet Communist Party slogan Народ и партия едины ("the people and the party are one"). I'm glad I don't have to find an appropriate translation that conveys this subtlety...
chessbibliophile chessbibliophile 5/8/2019 05:36
@Dr Zeiss, Here is the same variation with more lines and engines confirm the verdict.
1.Rg8+?! Kxg8 2.e8Q+ Kh7!
(If 2...Kg7? 3.Qxd5 Qxa5+ 4.Qxa5 Rf1 5.Qxc3 Rf3 6.Qf8+ Rxf8 7.e6+ Rf6 8.d7 c1Q+ 9.Qxc1 Rxe6 10.Qc3+ Kh7 11.Qd3 wins)
3.Qh8+! [3.Qxc3? loses to Qxa5+! 4.Qxa5 Rf1!! (not 4...Rg1? 5.Qc3 Rg3 6.Qxg6+!) 5.Qc3 Rf3] Kxh8 4.Qh6+ Kg8 5.Qxg6+= with perpetual check
Dr Zeiss Dr Zeiss 5/8/2019 12:50
@chessbibliophile, Please check your variations with engine. Unfortunately it shows that white is winning both after 2.-Kh7 or Kg7.
chessbibliophile chessbibliophile 5/8/2019 11:00
@Dr Zeiss, In the fourth study, the plausible 1.Rg8+?! only draws after 1…Kxg8 2. e8 =Q+ Kh7! 3.Qh8+! Kxh8 4.Qh6+ followed by perpetual check.
If 3. Qxd5? Qxa5+! 4. Qxa5 Rf1!! (not 4...Rg1? 5. Qxc3 Rg3 6.Qxg6+!) and Black wins, though for the moment White is two queens up. There are several other spectacular variations in these sidelines.
chessbibliophile chessbibliophile 5/8/2019 10:32
@Frits Fritschy, For reasons of space I offered only the main line of analysis by Tkachenko in each case. Otherwise under this second study he also gives the same line as yours after 1…g6. That’s a nice coincidence. More interesting is your reply to the attempt, 1.h4. Quite a few players would like to see why White does not simply push the pawn and win. He can not as the Black king is faster and reaches the critical squares e5 and f4. This is your contribution and it is not in the book!
Dr Zeiss Dr Zeiss 5/8/2019 09:11
And in the fourth study (1.Rf7+) there is a dual with 1.Rg8+ leading to the same continuation.
chessbibliophile chessbibliophile 5/8/2019 05:27
As is known, Pervakov co-authored the book, "Studies for Practical Players" with late Mark Dvoretsky years ago. Recently he brought out another work with the late master, Этюд и партия едины (literally, “Study and Game are Equal”).It is published by the Russian Chess Federation and would also be translated into English. There are parallels between examples from game positions and endgame studies in this slim volume.
macauley macauley 5/7/2019 09:56
@vento56 @brian8871 - Thanks, fixed.
brian8871 brian8871 5/7/2019 12:21
The first puzzle should say "White to play and draw," not win.
vento56 vento56 5/7/2019 12:02
First study caption should be White TO move and DRAWS
Frits Fritschy Frits Fritschy 5/6/2019 11:16
In the second study, a relevant variation is 1... g6 2 Bxg6 Kc7 3 h4 Kd6 4 h5 Ke7 5 h6 and now 5... Kf6 6 Bh5 or 5... Kf8 6 Bh7 Kf7 7 Kxf3 Kf6 8 Kg4 and white is just in time to keep the black king from g5. And 1 h4 doesn't win: 1... Kc7 2 h5 Kd6 3 Kf2 Ke5 4 Bg6 Kf4. The bishop needs to be at f5 to prevent this second solution.