Looking at it again, I believe you are right. I was probably too focused on the computer's first assessment, and therefore I looked more at the computer move 16.e5.

Black can avoid that specific variation by playing 15...0-0 instead. Then 16.Nf3 is line c, while 16.e5 Bxh4 17.gxh4 Qxh4 transposes to what I chose as main line in line d. Now I only looked at 18.exd6, but of course, White could also choose other alternatives and stay a pawn down for good compensation. I will have to look at this more closely, but after looking at some possibilities here, my current impression is that White has enough compensation and his position is easier to play, but Black is hanging on and is not worse if he plays carefully.

If that is the case, you could say that with the current status of the 5...c5 line, both players could be satisfied with having this line in store. As White, you are very unlikely to face this line, and if you do, you have an answer where White is not worse and Black is the one who must be more careful. As Black, you are also unlikely to face this approach from White, and if you do, you are not worse and you have probably studied the position more than White.

Thank you for bumping this thread which I missed back in 2010. May I suggest another "mysterious" queen move: (after 10.Bg5 Qd7 11.cxd5 h6 12.Bxf6 Bxf6 13.Qd2) 13...Qe7!?

Black is spending a lot of time and is falling behind in development, but he has the bishop pair, and currently I do not see a White advantage in any line. The queen move serves at least three purposes:
- Black can now put his knight on d7.
- Black threatens, in some lines, to take twice on h4 (not yet a threat due to Nb5).
- Black opens up the possibility to take on h4 followed by ...g5.
Meanwhile, he stays flexible when it comes to castling.

Some fresh analysis:

a) Slow play like 14.Nf3 should not trouble Black, e.g. 14...Bg7 15.e4 Nd7 16.Rfe1 0-0 and it looks like a normal Benoni where Bb7 might be a bit misplaced, but White is missing his black-squared bishop.

b) The direct attempt 14.Ne4 Bxh4 15.gxh4 Nd7 16.Qf4 Ne5 also looks fine for Black. White is not getting any further, and his queen and knight are now blocking his own pawns. 17.Nf6+?! Kf8 achieves nothing, as Black kicks back the knight with ...Kg7 on his next move.

c) 14.f4 Nd7 15.Nf3 0-0 (15...0-0-0!?) 16.e4 Bg7 and again it looks like a fine Benoni for Black to me. Black must watch out for the e5 break, but without his black-squared bishop, White must also be careful not to over-stretch.

d) The most critical line is probably 14.e4 Nd7 15.f4 (15.Rfe1 Bxh4 16.gxh4 g5!? 17.b4 Ne5 and the only evaluation I dare give at the moment is "unclear.") 15...Bxh4 (15...0-0 16.e5 and Black probably has to take on h4 anyway) 16.e5! (16.gxh4 Ba6! and when the rook moves, Black takes on h4) with a messy position, e.g. 16...0-0 17.gxh4 (17.exd6 Qxd6 18.gxh4 Nf6 19.f5 Kg7 looks equal) 17...Qxh4 18.exd6 and again, "unclear" looks like the only safe evaluation. White has a rather unusual pair of d-pawns which could be either strong or weak.

(bumping up an old thread)
I've looked at this line as well some time ago, and my conclusion was that Black should play d6 on move 8 because I ran into problems after 8... Bg7 9. Bg5, so 6. d5 exd5 7. Nh4 g6 8. Nc3 d6 and if now 9. Bg5 then Be7, which works because of the misplaced knight on h4.
I've looked a bit at Pantu's proposal 9. O-O Bg7 10. Bg5 Qd7 11. cxd5 h6 12. Bxf6 Bxf6 13. Qd2 and here I still seem to like 13...Na6 14. e4 O-O-O, the f4-move (a way to make sense of Nh4) doesn't attack the black king here, of course now White can stage a queenside pawn storm but I don't know if that's so scary with a fianchettoed bishop around there. What about for example 15. f4 Bxh4 16. gxh4 Kb8 ...?
(but seeing that BPaulsen on another thread (http://www.chesspub.com/cgi-bin/yabb2/YaBB.pl?num=1334169475/17#17) considers this to be White's best attempt, I may have missed something of course)

edit: Houdini seems to overestimate 15...Bxh4 (Stockfish isn't fond of it) but in some lines he only 'admits' this around move 35, 15...Bd4+ 16. Kh1 Rhe8 looks like a better try

BPaulsen wrote on 12/11/10 at 12:17:41:


It seems white can practically force black into an inferior hedgehog with the 1.Nf3 move order vs. your "super accurate" Ne4.  Warning:  rank amateur analysis to follow...  and no Brian this idea didnt come from an engine


1. Nf3 Nf6 2. c4 b6 3. g3 Bb7 4. Bg2 e6 5. 0-0 Be7 6. Nc3 Ne4 7. Nxe4 Bxe4 8. d3 Bb7 9. Qc2!?

This move does 3 things:

The queen covers e4, hence renewing the threat to push d3-d4 and a later e2-e4.  Black should probably therefore prevent this plan with 9...c5.

White's queen is also targeting the h7 square.  This lingering threat makes it dangerous for black to castle too early.  White threatens to push d4 when Ng5 threatens mate on h7 and frees the g2 monster on the unopposed long diagonal.

Hence, white simply plays a good waiting move.  Fortunatenly his queen move to c2 gave him the opportunity to play an excellent waiting move...  10. Rd1!.  

Again d4 is threatened.  How does black respond?  Castling is met by d4, when black won't have time to prevent both d5 and Ng5.   Pawn moves seem too slow since d4 now threatens to open a file with white's rook opposite black's queen.

It seems black's best is 10...Nc6 to prevent d4.  But now black has a hedgehog with a less than ideal knight (on c6 instead of d7).  White can prepare d4 with a queenside fianchetto, where his Qc2 looks reasonably placed:  11.b3 0-0 12.Bb2.  

This seems like a reasonable edge for white to me.  Black can't prevent d4 and doesn't have any obvious counterplay.

Phil Adams wrote on 12/14/10 at 15:20:24:


I've noticed similar types of problems with evaluations in Marin's books as well. Ignoring some dynamism, White would be better, but wait, Black has....



Still great of course, just frustrating sometimes.

I decided to explore the alternative line 1. d4 Nf6 2. c4 e6 3. Nf3 b6 4. g3 Bb7 5. Bg2 c5 6. d5 exd5 7. Nh4 g6 8. Nc3 Bg7 9. O-O O-O 10. Bg5 Nc6 played in Saemisch-Reti, 1927. Reti sacrificed the d-pawn, gained decent compensation and won the game against the notorious time-trouble addict. 

Having analyzed the game, however, (see attachment) I do not think the line is satisfactory for Black. See what you think.

BPaulsen wrote on 12/13/10 at 01:23:28:


For now, I sort of agree that 12...0-0 13.Nf3 b5 14.h3 Ne5 15.Nxe5 Bxe5 16.Bh6 Re8 17.Qd2 could be a main line. However, Im not so sure that White is just comfortable. Maybe we could try to look at this line a bit more? 

First of all, it is worth noting that 16...b4 is possible and unlike the main line where 16...Re8 17.Qd2 has been inserted the White knight can no longer go to d1. However, White has 17.Bxf8 bxc3 18.Bh6 (18.Bxd6!?) and Im not entirely convinced that Black is ok. 

This lead me to investigate 14...b4. White now has two obvious choices: 
a) He can choose to take up the fight and go for hxg4
b) He can willfully retreat his knight to e2

I wont go through all of it but basically I found that 
a) 15.hxg4 bxc3 16.bxc3 Bxc3 17.Rc1! seems good for White


Then I was  looking at 12...0-0 13.Nf3 Re8. Now 14.Bh3 looks menancing but it seems to me that Black can just ignore the threat and play 14...Na6. Therefore my main suggestion for White was 14.h3 Ne5 15.Nxe5 Bxe5 16.Qd2 but then I realized that, from Black's perspective, this is probably just an inferior version of the line suggested by BPaulsen after: 12...0-0 13.Nf3 Na6 14.h3 Ne5 15.Nxe5 Bxe5 16.Bh6 Rfe8.
However, I get the impression that Black is not doing too bad here what do you think?

BPaulsen wrote on 12/13/10 at 00:44:53:



My intention was to avoid ...h6 all together if possible. Thus I would be inclined to play 12.e4 0-0!  The point here for Black is that now he has sucessfully avoided the trade of dark-squared Bishops as, generally, Qd2/Qc1 can be easily met by ...Re8 but in some specific variations also ...h6 or ...Bd4 as long as the Knight has still not been kicked backed from g4. 

Also, the advance of the e- and f-pawn is not easy as long as a) the black knight is still on g4 and b) as long as the White dark-squared bishop is "stuck behind enemy" lines on g5. Btw, a further point is that the Black knight might in some cases go to e5 and d3 instead of the "normal" f6. But anyway, thats of course not the main idea behind 11...Ng4 but merely a point to remember