1.e4 d5 2.d4 dxe4 3.Nc3 Nf6 4.f3 c6 5.Bc4 exf3 6.Nxf3 Bf5 7.0-0 e6 8.Ng5 Bg6 9.Ne2 Bd6 10.Nf4 Bxf4 11.Bxf4 Nd5 

I'm back to thinking 12.Qd2 is White's best try at compensation in this line.  I'm still not sure, however, if Black should go 12...h6 13.Nf3 Nxf4 to ensure the bishop pair or play 12...Nd7 or 12...0-0.  After 12...Nd7 13.Bg3, White does keep a grip on the e5-square and sometimes after ...N7f6 the Bishop is redeployed to e5 with good effect.  After the immediate 13...N7f6 maybe White should try 14.Bd3 stopping Ne4 for the moment at least.  Anyway, I still think Black is better, but I have been unable to find a plan for Black that crystallizes the situation.  If I can get my analysis a little more organized then I'll post it so others can poke holes in it.

Sevenviolets wrote on 05/26/06 at 07:30:51:


I think where black goes wrong in your variation is 16.hxg6?!, as one can see afterwards the black pieces (in particular the knight on e7) lack squares and have to maneuver around oddly. Secondly that move opens the h-file, so that when the queen turns up on h3 one cannot at least try h7-h6. I couldn't completely check the whole rest of the variations (I somehow feel black should not end up in such a horrible position as the end one, even after a mistake like 16.hxg6), but there is no real need (or so I suspect), because there is a much better move:
1. e4 d5 2. d4 dxe4 3. Nc3 Nf6 4. f3 exf3 5.Nxf3 c6 6. Bc4 Bf5 7. O-O e6 8. Ng5 Bg6 9. Ne2 Bd6 10. Nf4 Bxf4 11. Bxf4 Nd5 12.Bc1 Nd7 13.Bd3 N7f6 14.c4 Ne7 15.Be3 0–0 16.Bxg6 Nxg6!
http://www.france-echecs.com/diagramme/imgboard.phpfen=rxxqxrkx%2Fppxxxppp%2Fxxp...
Black's last move not only puts the knight on an active square where it supports e6-e5 (black should continue with that idea, even more so now, because white's bishop's pair has halved), blocks the b1-h7 diagonal (well, hxg6 obviously does that, too), keeps the h-file closed (i.e. h7-h6 is still in option if it is ever needed) and keeps all of black pieces coordinated, but also potentially creates threats on the kingside (yes, black can also do that in the BDG!).

The best I could find for white here is 17. Qc2, because 17. g4 is just bad due to 17...e5!. While after 17. Qf3 black has 2 attractive options, firstly 17... Nh4 18. Qf4 Nf5, which looks a bit weird, but seems quite good. Secondly there is 17.Qf3 e5 18. Rad1 Re8 19. dxe5 (19. Qh3 Qc8!) 19... Nxe5 20. Qg3 Qe7 21. Bd4 Rad8 22. Rfe1 c5, which as far as I can tell works (but it certainly looks a bit risky, on the other hand I couldn't actually see anything wrong with it).

The idea behind Qc2 is that it not only gets the queen of the d-file (in anticipation of e6-e5 so that doesn't lead to swapping queens), but it also exerts some potential pressure on h7 restricting the movements of the knight on g6 (which also keeps black from playing e6-e5 right away).

17... Qc7   
It's not completely clear how black should play, but I like this move - for once black is also aiming at the vicinity of the opponents king, now Nf6-g4 is one idea and should white play h2-h4, then g3 looks quite inviting for the queen.

Variation A
18. Rae1 
A move like 18. h4 just fails due to 18...Qg3 19. Rae1 Qxh4, that's the sort of thing 18.Rae1 is designed to prevent, however even if black does nothing something like 19.h4 is still not seriously good.
18... Ng4 19.Bf4 Qd7 20. Qd1
I tried to find something for white like 20. Rd1 f6 21. Nxh7, but it all doesn't seem to work: 21.Kxh7 22. Rf3 f5 23. Rh3+ Kg8 24. Bg5 Qd6 25. Rg3 e5 26. c5 Qe6 27. dxe5 Rae8 28. Qb3 Rf7 and black just wins.
20... e5 21. dxe5 Qxd1 22.Rxd1 N4xe5
The final position is clearly winning for black, but maybe white can improve earlier?

Variation B
Considering how things go wrong for white in variation A, I suppose that 18. Rad1 would have to be the mainline. It is less clear to me how black should play than in aforementioned variation, but after a bit of thought 18...c5 does make sense (which prepares once more e6-e5, which currently fails due to 18...e5? 19.Rxf6! gxf6 20.Ne4 with serious compensation for white).

What should white play? 
1) 19.dxc5 Ng4! is just bad for him, e.g. 20.Bf4 e5 21.Qe2 Nxf4.
2) 19.d5 exd5 just cannot be right, this just transforms white's space advantage on the king's side (due to the pawn on d4 controling the e5 square) into a meaningless isolated pawn on d5: 20.cxd5 Qe5 21.Bc1 Qd6!.
3) So maybe 19.Rf3, but then after 19...Rad8 I cannot see how white can avoid unpleasant exchanges usually based on the idea of ...cxd4 followed by ...e6-e5.

So, maybe the only serious options (which would of course appeal to a BDG fan) is 19.Rxf6 gxf6 20.Ne4 f5 21.Nxc5, but do those slight weaknesses on the black queenside really compensate a whole exchange? It does seem unlikely to me, but at least black now has a vague weakness. On the other hand white's pieces are hardly a well coordinated attacking force (in fact they are just standing on squares where they had to go to cope with threats by black), so I really think black should be fine, particularly if he can open some files for his rooks.

Mnb´s suggestion looks good. But 15.Rh3?! is not a good try. Maybe I will send the position to Rajmund but he has quite many corr. games in progress right now, so I´m not sure he would have time to look at it. 
Time ago I analysed 11.Rxf4 when h6 is met with Rh4 (then all pieces will eventually retreat and h6 is weak, probably with some sacrifice on h6). I found promising compensation. Rook lifting idea. 12.Bc1 Sun Variant is strange, a4-Ra3 idea is principially good, but is this enough? There is another alternative 12.Qf3!? instead of Bc1. But I´m far from conlusion. For the time being, I think the position is roughly balanced after 9..Bd6! But 11.Rxf4 should be important sideline worth further searching.

1. e4 d5 2. d4 dxe4 3. Nc3 Nf6 4. f3 exf3 5.Nxf3 c6 6. Bc4 Bf5 7. O-O e6 8. Ng5 Bg6 9. Ne2 Bd6 10. Nf4 Bxf4
Here the computer actually spotted something that would make every BDG fan happy: 10... Bf5? 11. Nh5 O-O 12. Nxf6+ gxf6 13. Rxf5 exf5 14. Nxh7, which is - I am sure - just the sort of thing they are looking for.
11. Bxf4 Nd5 12. Bc1
Rajmunde:"...resilient attempts of the alchemist seemed to bring a success in the end. They have brought into light a wholly new Alchemic formula consisting of some unusual even absurd moves like Bc1, a4 and Ra3. Thus, bringing a breath of fresh air into the Alchemy Variant again. The Sun Variant were found just a couple of moments after his birthday.  For astrologists (conjunction Sun – Sun) birthday are sometimes referred to as a „Return of the Sun“, bringing new beginnings…"

It's difficult to believe that such a long-winded maneuver can really make sense, particularly when the usual claim is that white is sacrificing material to gain a lead in development. But now white is slowing down his entire attack even though black has made no concessions except for swapping of his dark-squared bishop for a dangerous knight. I guess a4 + Ra1-a3 will make sure that all white pieces are somewhat actively aimed at the black kingside, but can that offer enough compensation? Somehow it's hard to believe. Of course it is a rather unspecific attack and it's easy for it to succeed if black does weaken himself (i.e. e.g. h7-h6), but MnB's suggestion of how to continue looks very promising for black to me - in those lines black just doesn't have any weakness and white just doesn't know what to aim at.

My thought on the alternative 12. Qd2 discussed above is that even after that I'd like to refrain from h7-h6 and would be more tempted by 12...Nd7 or 12... O-O. But in a way I find that I like 12.Qd2 better than 12.Bc1, at least it seems a bit more active. Somehow it seems like 12.Bc1 just desperately tries to hang on to the pair of bishops, just because they are supposed to be useful for attacks every now and then, but as a result of the bishop being on c1 white then has to resort to Ra1-a3 in order to get that rook into play. I suppose in principle there is nothing wrong with that rook lift and e.g. in the Budapest there are variations where this is a very strong idea, but I am not particularly convinced here.
12... Nd7
That seems like a very sensible suggestion to me, definitely the knight belongs here, are there any arguments for playing 0-0 first? Could some other move possibly be more useful soon than Nd7? Not as far as I can see, so Nd7 is quite intuitive.
http://www.france-echecs.com/diagramme/imgboard.phpfen=rxxqkxxr%2Fppxnxppp%2Fxxp...
13. a4
That certainly seems the obvious continuation in the spirit of the quote by Rajmunde that I gave above, but looking at Rajmunde's notes on the "sun variation", it may be that he might also consider 13. Bd3, but I don't really see much progress there after e.g. 13...N7f6 14. c4 Ne7 15. Be3, when white is certainly well developed and has a solid position, but he is just a pawn down.
13... O-O 14.Ra3 e5
White's compensation (whether it's sufficient or not is not what I am talking about here) is in part based on the fact that he has a slight space advantage, so it is in a way tempting to "equalise" in that respect, on the other hand this is opening the position up a bit when white has the bishop's pair - I am not sure whether that should really worry black, but it may encourage thoughts about 14... N7f6. On the other hand N7f6 may just be too passive? It's difficult to say, really.
15. Rh3 N5f6
http://www.france-echecs.com/diagramme/imgboard.phpfen=rxxqxrkx%2Fppxnxppp%2Fxxp...
I would tend to agree that white doesn't really have enough compensation here.15... Re8 16. Nxh7 f6 certainly looks more worrying for black to me - maybe it is actually fine, but that is just the sort of position where I would suspect tactical tricks everywhere (maybe I'm just too paranoid). My computer even suggests 15...Qb6, which looks quite interesting, but maybe it's unnecssarily ambitious.

Quite honestly maybe 15.Rh3 is just silly, there seems to be very little point to it at the moment. Maybe white should just wait and see whether he can do something useful with that rook later, but - really - it's very hard to suggest anything sensible for white at all. I would agree with MnB here that white's play just doesn't seem to be based on anything much.

Upon further reflection I'm not too sure about my 12.Qd2 suggestion.  I think Black can achieve =/+ in this variation.  Probably 12.Bc1 is best but I agree with MNb that White's a4 move appears fairly useless if Black refrains from ...h6.  The 9...Bd6, 10...Bxf4 plan once again makes the Ziegler Variation a good choice to play for advantage without risking too much.  Of course I won't be upset if someone disagrees and would actually like to see analysis that shows that White can get at least a dynamically (or undynamically!) equal game.  My repertoire doesn't require me to defend against the BDG and I'd like to occasionally use it as a surprise weapon.  The single time I've used it as such was a completely opening success and after many blunders I did manage to win the game.

As I wrote before, I can't get access to Rajmund's website (I have tried on two different locations), so I don't know when and where my analysis is independent - not that it is that spectacular. But I have come to the conclusion, that Black must avoid h7-h6. In that case the rook switch Ra1-a3-g3 is rather silly.
What's wrong with eg 12.Bc1 Nd7 13.a4 o-o 14.Ra3 e5 and with 14...N7f6 ? The best idea Fritz gives is to play Ng5-f3-e5 voluntarily. In that case Black will be very happy not to have played h7-h6 of course. In other cases I think White has some pressure, but not enough to compensate the pawn.

It seems to me, that 5...c6 6.Bc4 Bf5 7.o-o e6 8.Ng5 Bg6 9.Ne2 Bd6 10.Nf4 Bxf4 11.Bxf4 Nd5 12.Bc1/12.Bd2 Nd7 followed by c5, N7f6 or o-o =+ is an easier defense for Black, than 5...Bg4 followed by Bxf3 and Qxd4. Still the latter draws much more attention on this site!?