Latest Updates:
Hot Topic (More than 10 Replies) 6.g3 - How serious, & which Black reply? (Read 5681 times)
fling
God Member
*****
Offline


I Love ChessPublishing!

Posts: 1591
Joined: 01/21/11
Gender: Male
Re: 6.g3 - How serious, & which Black reply?
Reply #13 - 09/24/19 at 18:10:17
Post Tools
Pawnpusher wrote on 09/24/19 at 11:44:38:
I looked at it a few minutes using Allie 0.5 with a Leelenstein net, on a modest desktop. After 6g3 it likes 6..g6 7Bg2 Nxd4 8 Qxd4 Bg7 9e5 Ng4 10 f4 Bd7 11h3 dxe5 12 Qb4 Nh6 (pretty much evaluated Nf6 the same) 13 fxe5 Bc6 14 Bxc6 bxc6 with a little edge for white.


I took some time to look at the alternative 9...Nd7 with Lc0 and SF10. The PV is 9...Nd7 10. Qa4 0-0 11. exd6 Nb6 12. Qf4 exd6 13. 0-0 Be6 14. Rd1 Nc4 and after 15. Rd1 Rc8 16. Bxb7 Black has several options, 16...Ne5!?, 16...Rb8 and 16...Rc5, all of which look fine for Black.
  
Back to top
 
IP Logged
 
Pawnpusher
Senior Member
****
Offline


I Love ChessPublishing!

Posts: 273
Joined: 01/04/18
Gender: Male
Re: 6.g3 - How serious, & which Black reply?
Reply #12 - 09/24/19 at 11:44:38
Post Tools
I looked at it a few minutes using Allie 0.5 with a Leelenstein net, on a modest desktop. After 6g3 it likes 6..g6 7Bg2 Nxd4 8 Qxd4 Bg7 9e5 Ng4 10 f4 Bd7 11h3 dxe5 12 Qb4 Nh6 (pretty much evaluated Nf6 the same) 13 fxe5 Bc6 14 Bxc6 bxc6 with a little edge for white.
  
Back to top
 
IP Logged
 
MartinC
God Member
*****
Offline


I Love ChessPublishing!

Posts: 2073
Joined: 07/24/06
Re: 6.g3 - How serious, & which Black reply?
Reply #11 - 09/24/19 at 08:42:31
Post Tools
I have no idea Smiley The training process has definitely reproduced a load of major classical main lines, so I do tend to trust it a little bit. What it doesn't do is spend much time looking at all these 'side' lines, and I hadn't bothered walking down them.

Doing that, 6 Bb5 and h3 both seem to have pretty coherent, slightly unique, thinking behind them. Definitely worth a brief look as practical weapons?

6 Bb5 Bd7 7 o-o ^ 7.. e6 8 Nxc6 Bxc6 9 BxB+ PxB 10 Qf3, 7.. g6 Re1, Bf1 etc (pure classical dragon I think?),  7.. Nxd4 decent choice for black too.

6 h3 it thinks blacks best is definitely 6..e5 7 Nf3 etc. vs g6 it just goes Be3, Qd2,o-o-o, g4 etc.  Similar vs e6 of course. 6.. e5 7 Nf3 etc.

6 Nb3 it just seems happy with a Yugoslav setup vs g6 stuff. Maybe playing Be2/g4 more than h4,h5 and mate. I can't see immediately see any real issue.

6 a3 I don't quite understand - its seemingly almost using it as a waiting move - but it does think a slight edge. Classical dragon style vs g6.

6 Qd3 seems a little odd. vs g6 it seemingly wants to play the 9 o-o-o Yugoslav with a queen on d3.
  
Back to top
 
IP Logged
 
Jack Hughes
Junior Member
**
Offline


I Love ChessPublishing!

Posts: 72
Joined: 07/22/19
Re: 6.g3 - How serious, & which Black reply?
Reply #10 - 09/23/19 at 20:02:53
Post Tools
MartinC wrote on 09/23/19 at 16:22:41:
I think part of the classical thing is that (at least my version of) LC0 doesn't rate the dragon all that highly.

Perhaps unjustified, I have no idea.

It wouldn't at all surprise me for it not to rate the Dragon generally - the Yugoslav Attack is too promising. What is surprising is that LC0 doesn't seem to share the conventional wisdom that only the Yugoslav Attack offers white anything. In regards to the question of what is meant by such a statement (by analogy to the Rauzer against the Classical), it should be noted that people very often say that the Dragon would be the mainline of the Sicilian if only white couldn't play the Yugoslav Attack. I've always thought that this made sense given the fact that the Reversed Dragon is the mainline of the English Opening (although I guess if black insists on a Reversed Open Sicilian then the line 1. c4 e5 2. Nc3 Nf6 3. Nf3 Nc6 4. e3 d5 5. cxd5 Nxd5 6. Bb5, which I would call a Reversed Four Knights, is white's best option), but given that LC0 and A0 have shown a liking to those lines as white maybe I shouldn't be so sure!
  
Back to top
 
IP Logged
 
MartinC
God Member
*****
Offline


I Love ChessPublishing!

Posts: 2073
Joined: 07/24/06
Re: 6.g3 - How serious, & which Black reply?
Reply #9 - 09/23/19 at 16:22:41
Post Tools
I think part of the classical thing is that (at least my version of) LC0 doesn't rate the dragon all that highly.

Perhaps unjustified, I have no idea.
  
Back to top
 
IP Logged
 
Jack Hughes
Junior Member
**
Offline


I Love ChessPublishing!

Posts: 72
Joined: 07/22/19
Re: 6.g3 - How serious, & which Black reply?
Reply #8 - 09/23/19 at 11:54:19
Post Tools
Sorry about my delayed response. The quotes are getting a bit messy at this point so I'll just use bullet point form.
- Taking a look at those lines after 9. e5 I'm inclined to agree that 9... Ng4 isn't as much of a turn off as I first thought. I'd basically saw low-depth engine evaluations that flickered between 0 and even a slight edge to black, combined with the fact that this line was recommended by Gawain Jones in his Dragon repertoire, and overestimated it for black. In fact I'm not even sure that black fully equalises in the 11. 0-0 variation. I would add though that in the line 10. f4 Nh6 11. Bd2 Nf5 12. Qe4 0-0 13. 0-0-0 dxe5 14. fxe5 Qa5 that 15. Rhe1 seems like a better try, when Stockfish on my computer is still giving white +0.37 at depth 25 after 15... Be6 16. g4. At any rate, it might even be quite useful to have such a drawish alternative in 15. Bd5 against someone who has shown such good preparation!
- Without actually having any concrete lines to back up my claim I have always thought of the Fianchetto Variation against the Dragon as being rather toothless, and that if white goes down the traditional mainlines with Nde2 black should be absolutely fine. One piece of evidence I can give is that the website zipproth.de used to have a fully minimaxed tree of high depth engine evaluations for every position that has been played in at least 500 games analysed at very high depths (based on the Cerebellum/Brainfish opening book), and even gave black a slight edge after 6. g3 in the Dragon. But I guess if LC0 is saying the expected score is 57% then I should probably rethink that.
- I'm not sure I can speak in general terms about the meaning of the claim that the Rauzer is the only serious try against the Classical. For one thing, I don't think it can be unambiguously stated that even the Rauzer promises white an advantage (see for example Illingworth's analysis in Dismantling the Sicilian)! In terms of the English attack my understanding is that the move ...Nc6 is supposed to be seriously inferior for black in the Najdorf, since the knight is often exposed there (e.g. in the line 1. e4 c5 2. Nf3 d6 3. d4 cxd4 4. Nxd4 Nf6 5. Nc3 a6 6. f4 e5 7. Nb3 Be7 8. Be3 0-0 9. Qd2 Be6 10. 0-0-0 Nc6 11. g5 Nh5 12. Nd5, when capturing on d5 will lose a tempo compared to 10... Nbd7). Instead if black tries to go for queenside counterplay with a5-a4 then the knight is probably not placed any worse on c6 than d7, and so Nb8-c6 is surely a more useful move than a7-a6. I would thus summarise the debate as follows. White is trying to show that a5-a4 is simply inferior to a plan based on b7-b5 given that the former is harder to get going if white goes for a light-squared bind with Bb5, a4, Bxa7 etc. Black, meanwhile, will try to argue that the free tempo ...Nc6 is worth more than the adoption of an allegedly inferior plan. In the abstract neither side strikes me as obviously correct, and meanwhile both (superficial) engine analysis and practical results seem to me to be in white's corner.
- I'm inclined to agree with MNb that the Perunovic-Lupulescu line might be troubling for white. Black is scoring really well there, and if you click through the ICCF archive you'll see a few black wins where engine evaluations are in white's favour most of the way through before black crashes through with a kingside attack. For OTB chess I would be inclined to just view it as an unbalanced, roughly equal position, and whether one is willing to play it as white will likely correspond to how you feel about King's Indian mainlines as white. All that said, I think the 9. Qe2 line I mentioned earlier looks like great fun for white.
- I'm genuinely shocked to see LC0 give a white edge after so many sixth moves in the Classical (but not in the least for the Najdorf - you would expect this sort of thing when black plays a  "high-class waiting move" like ...a6 and the recent popularity of moves like 6. Bd3 and 6. Nb3 even at elite GM level testify to that). Should we all start looking into 6. Qd3 or 6. a3?
  
Back to top
 
IP Logged
 
Pawnpusher
Senior Member
****
Offline


I Love ChessPublishing!

Posts: 273
Joined: 01/04/18
Gender: Male
Re: 6.g3 - How serious, & which Black reply?
Reply #7 - 09/14/19 at 11:20:38
Post Tools
Yes using the Nibbler gui with Lco is a great tool for looking at all possible idea.
  
Back to top
 
IP Logged
 
MartinC
God Member
*****
Offline


I Love ChessPublishing!

Posts: 2073
Joined: 07/24/06
Re: 6.g3 - How serious, & which Black reply?
Reply #6 - 09/13/19 at 14:23:50
Post Tools
Very off topic then Smiley A '5 minute' LC0 perspective on my half decent (not massive) desktop. LC0 is, I think, good for that sort of meta questions as its very strong and utterly free of inbuilt human bias.

Vs the Classical it's roughly Bg5 58%, f3 and Bc4 at 57%, Bb5, h3, a3, Be3, g3, Nb3 and Qd3 at 55%,  Be3/a4/Nxc6/Nf3 at 53.5% and f4/h4 at ~51% etc.

The Nardojf is really quite amusing. Be3, Bd3, f3, Bg5, h3, Be2, f4, Nb3, a4, g3, Bc4, Rg1, Qf3 and Nf3 *all* more or less round to 55% expected score to white. Qd3/h4/a3/Qe2 do 53%
(There are a few ideas behind 6 h4, honest.).

I think it puts into perspective how overly narrow we sometimes get/have got.
  
Back to top
 
IP Logged
 
IsaVulpes
Senior Member
****
Offline


No.

Posts: 345
Joined: 12/09/07
Re: 6.g3 - How serious, & which Black reply?
Reply #5 - 09/13/19 at 09:57:22
Post Tools
Jack Hughes wrote on 09/11/19 at 20:13:34:
As black I would recommend 6... g6 7. Bg2 Nxd4 8. Qxd4 Bg7 9. e5 Ng4 10. f4 Nh6, when I'm really unable to find anything tempting for white. After 11. 0-0 0-0 the e5-pawn is going to drop for insufficient compensation, while the more ambitious 11. Bd2 gives black a pleasant choice between the relatively safe 11... Nf5 12. Qe2 0-0 and initiating complications with 11... 0-0 12. 0-0 Bg4 13. Rde1 Be6. In either case white doesn't seem to be even pressing for an objective advantage.

9. ..Ng4 is a good point, kinda forgot that's an option.

Still seems like a decent try to me, however:
- Presuming you mean 11.Bd2 Nf5 12.Qe4 0-0 (Qe2 not a legal move), the least I found was 13.0-0-0 dxe5 14.fxe5 Qa5 15.Qd5 Qxd5 16.Nxd5 Bxe5 17.Bb4 Kg7 18.Nxe7 Nxe7 19.Bxe7 Re8 and ok it's gonna be a draw, but there was no risk whatsoever, and Black had to solve some questions (eg 17...Kg7 isn't exactly the most natural move in the world to me, and more natural tries like 17...Re8 or 17...Bd6 leave Black in very unpleasant waters);
- While the complications after eg 11.Bd2 0-0 12.0-0-0 Bg4 13.Rde1 Be6 14.Bxb7 Rb8 15.Bd5 dxe5 16.fxe5 Nf5 17.Qd3 always seem to end in some sort of "I have an extra pawn, go prove something, I'm not seeing it" type positions.

IF that doesn't hold up to closer scrutiny, there's also always decent other choices for White on the path, like 6...g6 7.Nde2 and just going into the standard type positions, or  6...g6 7.Bg2 Nxd4 8.Qxd4 Bg7 9.0-0 0-0 10.a4 / similar, which get some kind of engine approval and are backed up database scores.

Quote:
In terms of a non-Rauzer weapon against the Classical I'm quite fond of 6. f3. In particular the mainline 6... e5 7. Nb3 Be7  8. Be3 0-0 9. Qd2 a5 10. Bb5 Na7 11. Bxa7!?, as played by Leela against Chess22k in TCEC Season 13, seems very interesting. Have you looked into this? What are your thoughts?

I loosely looked at this, as it's recommended in Shaw's Sicilian Mainlines, and I'm sorta/sometimes playing 6.f3 against the Najdorf as well (haven't actually faced the Najdorf in what feels like forever, there's been a few Dragons and a lot of 2. ..Nc6 recently), but I think back when I had a glance at it, there was still this "Bg5 is the only serious move" stuff spooking around in my mind, so I was too lazy to take it seriously.
It certainly is on some sort of "list" of things to look at, if I end up unhappy with the g3 line..
.. or perhaps I 'have to' look at f3 in any case, as Black can always sorta transpose into it, if he goes Nc6 somewhere in the Najdorf?
I always get lost in whether these lines are just the same position, or suddenly dead lost for Black, cause the extra tempo spent on ..a6 was necessary elsewhere  Lips Sealed

Remains the question: Why is this "Rauzer is the only serious try" something that gets repeated so often? Simply because it "promises" an advantage, while every other move is only "an attempt" at one (while eg vs the Najdorf, nothing promises one)?

Ie, is 6.f3/similar against the Classical *less* promising than against the Najdorf, or are they "the same", and there's just an even better option against the Classical (Bg5), while nothing as strong exists vs the Najdorf?

MNb wrote on 09/12/19 at 05:58:16:
@JackH: It's rather 6...e5 7.Nb3 Be7 8.Be3 Be6 (iso O-O) 9.Qd2 d5 or 9.Nd5 Bxd5 10.exd5 Nb4 that puts me off.

Shaw quotes this game http://www.chessgames.com/perl/chessgame?gid=1522621 amongst others, and doesn't seem too miffed about the whole variation.
Haven't done any analysis myself, so I'm not sure if he is omitting something, but it doesn't "feel" very problematic to me either..

Jack Hughes wrote on 09/12/19 at 10:47:19:
At any rate, we're getting off topic here (unless IsaVulpes wants to rename the thread 'Alternatives to the Rauzer' or something), and I don't my second of three thread contributions to be another derail, so perhaps we could continue this discussion in a new one.

Hah, it's no problem. I am happy with all discussion, good to see this forum pumping again in some fashion  Wink I guess this renaming is actually not too off, as it was one of the core questions I had upon thread inception anyway..
E: Okay, I can't rename it - either because people have posted here, or because it's too old, but that option simply doesn't exist anymore.
Feel free to open a new thread on discussing 6.f3 if you like, but I also don't mind if you keep that talk here.
  
Back to top
 
IP Logged
 
Jack Hughes
Junior Member
**
Offline


I Love ChessPublishing!

Posts: 72
Joined: 07/22/19
Re: 6.g3 - How serious, & which Black reply?
Reply #4 - 09/12/19 at 10:47:19
Post Tools
MNb wrote on 09/12/19 at 05:58:16:
@JackH: It's rather 6...e5 7.Nb3 Be7 8.Be3 Be6 (iso O-O) 9.Qd2 d5 or 9.Nd5 Bxd5 10.exd5 Nb4 that puts me off.

Taking a quick look at the database 9. Qe2 looks promising, with the point that 9... d5 with now run into trouble on the d-file after 10. exd5 Nxd5 11. 0-0-0. Instead black should play 9... 0-0, when white plays in typical English Attack style with moves like h4, g4, h5, Kb1, Nd5 etc. on the cards, with the additional possible idea of Qb5 to slow up black's queenside counterplay. I haven't done any proper engine analysis here, but a cursory click through the games played on ICCF in this line seem to suggest that white remains flexible between slowly building up on the kingside and entering into a maneuvering battle in the centre and queenside based upon the weak d5-square, while black's active prospects are a bit more vague and dependent on white's response.
At any rate, we're getting off topic here (unless IsaVulpes wants to rename the thread 'Alternatives to the Rauzer' or something), and I don't my second of three thread contributions to be another derail, so perhaps we could continue this discussion in a new one.
  
Back to top
 
IP Logged
 
MNb
God Member
*****
Offline


Rudolf Spielmann forever

Posts: 10756
Location: Moengo
Joined: 01/05/04
Gender: Male
Re: 6.g3 - How serious, & which Black reply?
Reply #3 - 09/12/19 at 05:58:16
Post Tools
@JackH: It's rather 6...e5 7.Nb3 Be7 8.Be3 Be6 (iso O-O) 9.Qd2 d5 or 9.Nd5 Bxd5 10.exd5 Nb4 that puts me off.
  

The book had the effect good books usually have: it made the stupids more stupid, the intelligent more intelligent and the other thousands of readers remained unchanged.
GC Lichtenberg
Back to top
 
IP Logged
 
Jack Hughes
Junior Member
**
Offline


I Love ChessPublishing!

Posts: 72
Joined: 07/22/19
Re: 6.g3 - How serious, & which Black reply?
Reply #2 - 09/11/19 at 20:13:34
Post Tools
Thanks for sharing this IsaVulpes. It's very interesting food for thought, and that MVL-Giri game is a treat. As black I would recommend 6... g6 7. Bg2 Nxd4 8. Qxd4 Bg7 9. e5 Ng4 10. f4 Nh6, when I'm really unable to find anything tempting for white. After 11. 0-0 0-0 the e5-pawn is going to drop for insufficient compensation, while the more ambitious 11. Bd2 gives black a pleasant choice between the relatively safe 11... Nf5 12. Qe2 0-0 and initiating complications with 11... 0-0 12. 0-0 Bg4 13. Rde1 Be6. In either case white doesn't seem to be even pressing for an objective advantage.
In terms of a non-Rauzer weapon against the Classical I'm quite fond of 6. f3. In particular the mainline 6... e5 7. Nb3 Be7  8. Be3 0-0 9. Qd2 a5 10. Bb5 Na7 11. Bxa7!?, as played by Leela against Chess22k in TCEC Season 13, seems very interesting. Have you looked into this? What are your thoughts?
  
Back to top
 
IP Logged
 
RdC
God Member
*****
Offline


I Love ChessPublishing!

Posts: 868
Joined: 05/17/08
Re: 6.g3 - How serious, & which Black reply?
Reply #1 - 09/11/19 at 07:36:38
Post Tools
IsaVulpes wrote on 09/11/19 at 04:49:39:
Hi,

6.g3 e6 looked the most logical to me, just going into a standard Scheveningen while having avoided the Keres, but as with the Classical I have no real idea about how good that is for either player


You could also look at 6. g3 against the Najdorf. In the Scheveningen, play builds slowly as Black sets up a Hedgehog formation and White prepares either to assault it or maintain pressure. There's an idea in main lines where White will play Bf1-e2-f3-g2, so playing directly Bf1-g2 looks a tempo gain. Without the threat of g2-g4-g5, Black can organise .. Qc7 .. b6 and .. Bb7 gaining time on the idea of the 1980s KvK matches of Bc8-d7-c8-b7.

In fact the attempts by both Karpov and Anand against Kasparov's Scheveningen are worth study as well as more recent attempts to improve on their play. 
  
Back to top
 
IP Logged
 
IsaVulpes
Senior Member
****
Offline


No.

Posts: 345
Joined: 12/09/07
6.g3 - How serious, & which Black reply?
09/11/19 at 04:49:39
Post Tools
Hi,

Following this game http://www.chessgames.com/perl/chessgame?gid=1967987 , I've gotten pretty interested in 6.g3 (from the White side), as I meet the Classical far too rarely to deal with learning the Rauzer, and I never really managed to warm up to playing the 6.Bc4 lines given in Modernized either.

The gambit is -as far I can tell- very legit, I've analyzed it a bit, and White always seemed to get a nagging edge at minimum - in the quoted game, MVL just played the most natural moves in the world and won effortlessly, with it being very unclear where Black should even improve.

However, Black is obviously not forced to play 6..Nxd4 7..g6; the Database spits out 6. ..g6/..Bg4/..e6/..e5 as more common, plus 6. ..Bd7/..Qb6 as additional options, and while smth like 6. ..Nxd4 7.Qxd4 e6 hasn't been played yet, it doesn't look illegal either.

Are there any other interesting lines to be aware of for White, if Black plays something that isn't ..Nxd4? I found eg this:
6.g3 g6 7.Bg2 Nxd4 8.Qxd4 Bg7 9.e5!?, in similar style to the MVL-Caru game
Play could continue
9. ..Nd7 10.Qa4, and now if
- 10..Bxe5 Bh6, with the Black King kinda stuck
- 10..dxe5 Be3 0-0 0-0-0, and Black looks terrible to me
- 10..0-0 11.exd6, when it gets a bit messy, but looks quite interesting.

6.g3 e6 looked the most logical to me, just going into a standard Scheveningen while having avoided the Keres, but as with the Classical I have no real idea about how good that is for either player;
Database spits out 7.Bg2 Bd7 8.Ndb5!? to be a tremendous weapon, but I don't really get what's going on there.. I'm also not sure how happy Black would be practically, with playing some Scheveningen instead of "their" opening.

After 6.g3 e5, I presume the simplest is to go Nde2 and play it in 6.h3 Najdorf style, just following up with h3-g4-Ng3-etc, when either the Black Knight on c6 would much rather be on d7, or Black never gets to do the standard ..h5 reply, which makes his position rather unpleasant to play.

6.g3 Bg4 is actually the second most common database choice, but I don't understand it at all. After 7.f3 Bd7, I'll go Be3+Qd2+0-0-0 & g4, landing in a standard English Attack (I didn't lose a tempo on g3-g4, as Black also played Bg4-Bd7), where Black not only missed out on all his active tries (such as a preemptive ..h5), but also went for an early ..Bd7, which is rarely any useful.

So my set of questions here would be
- How critical in general is 6.g3 judged to be? My vague recollection has always been "nothing but the Rauzer is serious against the Classical", but all of this looks decently promising for White to me, so am I missing some important line, or grossly misevaluating something?
- What's the generally preferred / most common approach for Black? Transposing into a Scheveningen with ..e6? Doing some Najdorf style ..e5? Direct ..g6? Would you consider playing 6. ..Nxd4 - or at least, would you have considered it, prior to seeing this MVL game (ie how likely would I be to actually get this fancy ..gambit?.. on the board)?
- Or does someone actually have some sort of 'refutation' for the MVL line, and it's not a serious try but just some one-off Rapid variation after all? I didn't find anything, but I'm not the greatest at doing so either.

Thanks!  Smiley
  
Back to top
 
IP Logged
 
Bookmarks: del.icio.us Digg Facebook Google Google+ Linked in reddit StumbleUpon Twitter Yahoo