This is really terrific. YouTube has the full video of Karpov-Korchnoi from the recent Pivdenny tournament in Odessa, and unless you already know the game well, it's worth the 20 minutes it takes to watch both parts. (Part 1 here, part 2 here [or find and follow the link after watching part 1].)
The game is fascinating on a number of levels: it's a very interesting game, and despite the G/10 + 5 second increment time control, the players, especially Karpov, feel free to take surprisingly long thinks from very early in the game - a good lesson for those of us who race through the opening like jackrabbits, whether we know what we're doing or not! As early as move 5 Karpov starts taking substantial "thinks", but more about this below. The game is also fascinating in its own right, with some neat tactical ideas from both players before the winner has the last laugh.
I don't want to spoil anything before you've had the chance to watch anything yourself, but for those who are interested in a guided tour of the game, offering various questions and exercises for you to solve during the videos, click on.
Question 1: The game starts 1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Qc2 d6, and now Karpov starts thinking. Why? Is 4...d6 especially terrifying?
Answer: No, it's not terrifying, and it's not new, either. It is unusual, though, and I imagine Karpov was thinking about one of two things. First, it's possible that he was trying to think of any potentially relevant predecessors - a game where someone else used 4...d6 against him, or a game where Korchnoi had employed it against someone else. Second and perhaps more likely, he hadn't faced it, and was either trying to remember a little of the theory and perhaps trying to determine if there was anything wrong with 5.a3, the move White uses against most normal moves. (There isn't, and so he played it after 1:14; ironically though, 5.Nf3 is the most common move there!)
Question 2: After 5.a3 Bxc3+ 6.Qxc3 c5, Karpov thinks for a full two minutes over his next move. What are the relevant issues?
Answer: There are at least two sorts of considerations that must be kept in mind. First, whenever a fundamental transformation of the pawn structure, especially in the center, can be effected, it's worth spending some time determining what ought to be done. Should White allow ...cxd4, and if so, should he recapture with a piece or a pawn? The second consideration is a typical one in the 4.Qc2 NimzoIndian when Black plays ...c5 at some moment: the possibility of ...Qa5+. White's kingside development is often slow in the 4.Qc2 Nimzo, and White has often suffered from quick attacks whipped up by ...Ne4 and ...Qa5+. Karpov's 7.Nf3 wasn't the only move, but the questions above needed to be answered to his satisfaction before he could play it.
Question 3: The game continued: 7.Nf3 O-O, and now Karpov thought for an incredible four minutes and 23 seconds, leaving him with just 3 minutes to go after only 8 moves! What could he have been thinking about this time?
Answer: In addition to various specific variations, there are further general issues to be addressed. First and foremost, there's the question of how to develop his bishops and get his king to safety. To the bishops: White could bring his bishop to g5, and then perhaps play e3 and bring the other bishop out. Or he could play g3, finish kingside development and only then worry about the Bc1. And then there's fianchettoing the queen's bishop, and that too can be done in various ways. Ultimately Karpov chose the latter method, playing 8.dxc5 dxc5 9.b4. The central tension is removed, and he no longer has to worry about his pieces getting displaced after ...cxd4.
But why not 8.Bg5? I'm not sure it's bad, but Black may have interesting avenues for counterplay after 8...h6 9.Bh4. Here's one wild, woolly and unforced variation: 9...cxd4 10.Nxd4 g5 11.Bg3 e5 12.Nb5 Nc6 13.O-O-O Ne4 with serious complications. Karpov's method keeps things under control, at least for the moment.
Question 4: What should Korchnoi play now, after Karpov's 9.b4?
Answer: Probably not the move he played. After his 9...Nc6, Karpov's rejoinder, the obvious (given his last move) 10.Bb2 nailed the knight down to f6 - no ...Ne4 leaps are permitted (again, at least for the moment). So the immediate 9...Ne4 may have been better, or at least worthy of consideration.
Question 5: After 9.b4 Nc6 10.Bb2, what now?
Answer: After (at least) four and a half minutes(!), Korchnoi played the dynamic 10...Qb6.
Question 6: Why?
Answer: The point is to force b5, after which the ...Ne4 and ...Qa5+ idea is again on. With White's bishop pair and space advantage, Black needs to make something happen before White finishes developing. Once that happens, White's long-term assets will give him a very comfortable and just about unloseable position with decent winning chances.
Question 7: The game continued 11.b5 Nd4 (11...Ne7 may be objectively best, but lets White finish developing in peace) , and now Black's idea is to meet 12.Nxd4 with Ne4 (no Qxg7# now). Can White cope with this idea, or should he abstain from the capture on d4?
Answer: After 1:53(!), Karpov decides that he can safely take on d4. After 12.Nxd4 Ne4 13.Qe3 (forced) Qa5+ 14.Kd1 Qa4+ happens quickly, and while it's clear that Karpov has at least a draw in hand after 15.Qb3 Qa5 16.Qe3.
Question 8: But after 16...Qa4+, repeating the position, does Karpov have more than a draw?
Answer: Yes! Karpov plays 17.Nb3, which doesn't seem so special at first due to 17...Nxf2+, when Black either wins the rook and seems to be doing fine, or else White takes on f2, when 18...Qxb3+ 19.Kc1 Rd8 20.Qe1 e5 probably wins outright (21...Bf5 is a huge threat, and 21.e4 Bg4 wins).
Question 9: So what's Karpov's idea?
Answer: To give away the rook, but win the queen! After 17.Nb3 Nxf2+ 18.Ke1! Nxh1 the bone-crushing 19.Qc3 forced immediate resignation, as White threatens both 20.Qxg7 mate and 20.Nxc5, winning the queen for free.
Bonus question: Karpov did make an error in this game. What was it?
Answer: 15.Qb3, offering the repetition the first time around. I'm sure Karpov did it for the extra 10 seconds on the clock it gave him (the increment for two automatic moves), but had Korchnoi traded queens and then played 16...Nxf2+, his disadvantage, though still serious, would be less than it was in the game.
Nice to see that neither player feels the slightest compunction in starting their move before their opponent has pressed the clock.