Reports | August 18, 2009 2:47

Jermuk R8: Ivanchuk beats Gelfand, Leko survives

Jermuk GPHe needed all his endgame experience, creativity and 98 moves, but Ivanchuk defeated Gelfand in the only decisive game of round 8 in Jermuk. He joined Leko in the lead; the Hungarian successfully defended a difficult ending against Jakovenko.

The 5th tournament in the FIDE Grand Prix Series takes place in Jermuk, Armenia. It's a 14-player round-robin with Aronian, Jakovenko, Leko, Gelfand, Bacrot, Kamsky, Karjakin, Eljanov, Alekseev, Akopian, Ivanchuk, Cheparinov, Inarkiev and Kasimdzhanov. More info on the GP and Jermuk in our preview.

Round 8

After such a horrible first half of the tournament, Inarkiev's quick draw against Eljanov was fully understandable. Not long after that the two Armenians did more or less the same, but at least the games were theoretically relevant.

Akopian-Bacrot was a theoretical battle where 16.Rad1 deviated from a game between the same players in the previous Grand Prix - Bacrot had probably looked at it and equalized easily. Aronian repeated the gambit line of the Queen's Indian he had beaten Leko with in Nice in March, but his 16.Qf4, also tried by Nyback last month against Wojtaszek, was answered with a novelty by Karjakin: 16...Re8. The idea was (a very nice way of) giving back the pawn for full mobilization, and makes you wonder till what point Karjakin had prepared!

Jermuk GP

Levon Aronian and Arianne Caoili listen to Gabriel Sargissian - was fruit juice the missing element in Levon's preparation?

The other four games lasted much longer: 60, 60, 98 and 121 moves! Let's continue from shortest to longest, which comes down to some sort of chronological sequence as well. Alekseev-Kasimdzhanov never seemed out of balance anywhere, but it's instructive to see such endings played out till the very end sometimes. Kamsky outplayed Cheparinov with Black in a Breyer and seems to have missed a win in the knight ending with 51...Nf2 52.h4 g4 53.Nd4 Ne4 54.Nf5 h5.

Jermuk GP

Ivanchuk after the game, at the press conference

Then the battle between the two 40-year-olds: Ivanchuk and Gelfand. It's not clear whether 18.Ba3 lost or sacrificed a pawn because White had clear compensation afterwards. Then 24.g4 was a great move, conquering the e4 square, and after the time cotrol Ivanchuk's compensation had turned into an advantage, which became close to winning after the mistake 43...Nxa4? where 43...Bd6 was necessary. However, Gelfand defended like a lion and might have still drawn with 51...Ra8 52.Nc3 Be7. The final phase was a pretty reward for Ivanchuk's persistance.

It meant that the Ukrainian was at least shared first in the standings, because on the last board left, Leko had been defending a Berlin Wall ending with a pawn down for ages already. It looks like Jakovenko missed wins with 89...Rh2 90.Nc4 Rxa2 91.Re8 a5 and 94...Rd3, so it was a lucky escape for the Hungarian! Black's final move must have been accompanied with a wink, finishing the marathon with a mutual smile.

Jermuk GP

Leko and Jakovenko at the start of their hard-fought Ruy Lopez

Leko is doing well, just like in Nalchik, Ivanchuk is again winning rating points back, Kasimdzhanov performs just excellent while Aronian is a point behind the leaders. But with two GP victories in the pocket already, he hasn't got too much to worry about.

Round 8 games

Game viewer by ChessTempo

Jermuk Grand Prix 2009 | Round 8 Standings

Jermuk Grand Prix 2009

Jermuk Grand Prix 2009 | Schedule & results

Jermuk GP

Aronian posing with tournament photographer Aram Karakhanyan, this time on the other side of the lens

All photos © Arman Kharakhanyan


Share |
Peter Doggers's picture
Author: Peter Doggers

Founder and editor-in-chief of, Peter is responsible for most of the chess news and tournament reports. Often visiting top events, he also provides photos and videos for the site. He's a 1.e4 player himself, likes Thai food and the Stones.


daniel's picture

I am so happy to see Ivanchuk winning. Fantastic game!

kabuki's picture

Ivanchuk should hit the top 10 today..replacing Gelfand..sweet victory indeed

tibone's picture

Peter you are writing that in Leko - Jakovenko 89...Rh2 would have won. Couldn´t Leko claim a draw due to the 50-move rule after that? I think ...Rh2 makes the 50 moves perfect.

tibone's picture

oops, sorry i started counting after 39...Kxb5 not after 40...c5 as i should have.

Thomas's picture

@ Castro: Don't make fun of me again regarding "counting moves manually"! :)
But you had good company (in counting wrongly): the tournament website and, so it seems, Jakovenko.
Regarding 89.-c4+, what is your follow-up after 90. Kc2 Rh2+ 91.Kb1 ? 91.-Bc3 allows simply 92.Nc4:, and after 91.-c3 92.Nc2 white also seems to keep sort of a fortress.
Altogether it's funny that the 50 move rule intervenes that often in recent top-level games. A third example was the Mainz rapid game Nepomniachtchi-Anand (Anand couldn't win a queen endgame two pawns up, thus losing his last chance to qualify for the final).

Castro's picture


The site is right, as you were on your 1st post. Your "correction" is wrong, as is Peter on the article.

89....Rh2 would be black's 50th, and the played 89.Re4 was white's 50th. So white could imediately claim a draw.

Instead of 89. ...Rh2= and instead of the played 89. ...a5?, Jakovenko should try 89. ...c4+! (resetting the counter), followed by 90. ...Rh2+ which, it seems, can be a win!

Castro's picture

:-) this is pandemic! Wrong me! 40. ...c5 is the last, yes. So 89. ...Rh2 was still posible (and wins). But 89. ...c4+ also wins, am I right here?
Also, it could be tried 88. ...Rh2 (directely) and, after 89.Rg2 c4+.

tibone's picture

No Thomas, Jakovekonko didn´t think 89...Rh2 is an immediate draw due to the 50-move rule. He can count.
He did not play it because of 90.Nd5 Rxa2 91.Nc3 (
I don´t know whether he didn´t see 89...Rh2 90.Nd5 c4+! or rejected it because he was thinking it leads to a draw.

tibone's picture

in fact the official site is wrong then. it says there:"Had black been able to play 89...Rh2, he would have won the a-pawn, and a win may have been likely, but with this move, he would have played the 50th move without a pawn push or a piece take, and thus was forced instead to play a5, and white maintained drawing chances. "
it seems that, like me, they can´t count.

Thomas's picture

Yes, 89.-Rh2 would have been the 49th move. But then Leko has the brilliant defence 90.Nc2!! Rd2+ 91. call the arbiter and claim a draw by the 50-move rule! Should I put a smiley? Seriously, this would have ended the game in this variation ... .

tibone's picture

Thomas, after 90.Nc2 c4+ (just in time) 91.Rxc4 Rd2+ wins. More tricky would have been: 89...Rh2 90.Nd5 because after 90...Rxa2 91.Nc3+ Bxc3 leads to a drawn rook ending (according to the tablebase, I don´t know anything about rook endings sadly).

Thomas's picture

Yes, actually (honestly) I had seen 90.-c4+ right after posting ... but wanted to give someone else the chance to post a correction :) .

As far as the possible rook ending is concerned: Yes, rook endings with a+c (or f+h) pawns are theoretically drawn [that's what I know, don't ask me how to defend them ...]. But the defense is rather tricky (even for top GMs) if you have to rely on increments. For example, Aronian lost such an ending against Kramnik (Corus 2008).

Castro's picture


Yes, indeed the tournament website, Jakovenko, tibone and me made the same mistake.
But (at least in my case and that of tibone's) it was not a counting issue, but a bad verifying in where to start counting. In any case I didn't count "manualy" hehe ;-) . Anyway, I too deserved to be made fun of, that's for sure!

You may also be right in that my sugestions wouldn't forcefully win either, but I gave them at least as seemingly superiour to 69. ...a5. And were also based on the wrong idea that 89. ...Rh2! was 50 moves rule draw.

Latest articles