Queens gambit declined three knights variation, with B6?

So I was playing a game and came to this position (as black) 

I'm pretty sure here, moves such as c6 are favorable. However with the move b6, it doesn't allow c6 (the c6 square is particularly weak) so that may be why.

Yeah c6 is most common although I don’t like the Slav much so I could play Bb4 or Be7

b6 is good, but later on. The problem is you'll be down tempi after both sides castle. Moves like c5, c6, Be7 and Bb4 are better. I would go for Be7 or Bb4 (Ragozin).

Yeah, b6 isn't necessarily bad but the way engines calculate evaluation includes a lot of tempo calculation so the tempo makes a decent difference

Ah ok- thanks for the help guys. Engines next suggested moves are bringing out their black squared bishop to chase my knight so I suppose that tempo-gaining makes sense. So if I wait until all my other pieces are developed and I’ve castled playing B6 directly before something like C5 might work?

i play the queen's gambit. both colors. b6 on move 4 is considered too early but it's also quite hard to exploit if you don't know how the typical queen's gambit plans for white proceed. you will probably get away with it for the most part if you play vs opponents rated below 2000. i personally can't even refute it directly.

i do like taking on d5 which is what white normally does when black commits to bb7. the idea is to force black to make a decision on what to recapture with on d5. if he takes with the pawn then he basically blocks his bishop that would be on b7, but if he takes with the knight it will get kicked with e4 and also black will have lost some control over the center as he doesn't have a pawn there anymore.

there are some positions where both these scenarios are still okay though. if you like this scheme with b6 and bb7, i suggest you play it only after completing kingside development, as this also allows you to see how white develops his pieces.

some guidelines, try to meet bg5 with nbd7 and then h6, so that after bxf6 you can take with the knight. i would even suggest playing 4..nbd7 here, preparing for it in advance, 

basically if white doesn't threaten anything immediate, you can complete the kingside development pretty easily. d5, e6, nf6, be7, 00. and then you may consider the queenside development. normally if you aim for bb7, then nbd7 and b6 would come first. and then you aim for either c5 or ne4 (there ar other ideas though) or both versus some standard queen's gambit scheme from white.

you could look up the tartakower variation of the queen's gambit declined. could also be named the makagonov-bondarevsky line. i'm sure there's lots of articles about it out there. it is the line featuring the queenside fianchetto that you like. i have played it myself.

Everyone is saying c6.  Yes, that would be a direct transposition to the Semi-Slav, but Black has legitimate options outside of that.

Orthodox - 4...Be7 5.Bg5 O-O 6.e3 Nbd7

Ragozin - 4...Bb4 and White has options - many books on the theory of this line, you would need one to play it right.

Vienna - 4...dxc4 5.Bg5 Bb4 - again, lots of theory - Everyman published a book on it a couple of years ago.

Tartakower/Lasker - 4...Be7 5.Bg5 h6 6.Bh4 O-O 7.e3 and now 7...b6 or 7...Ne4 (Tartakower and Lasker, respectively)

Cambridge Springs - 4...Nbd7 5.Bg5 c6 6.e3 Qa5

Personally, I would suggest starting with the Orthodox or Lasker.  They are easier than the rest.

Yes, I agree, personally find the chess.com obsession with c6 semi-slav a bit strange.  Also think 4...b6 isn't that bad, not +1 bad at least, software is frequently unreliable very early in opening.

I did some rudimentary research within my Mega Database. It appears that Black's position is rather unpleasant after either 4...b6 5.cxd5 Nxd5 or 5...exd5. In the former variation, 4...b6 obstructs the natural development of Black's queen to a5. In the latter, the early ...b6 allows White to exploit the weakened a4-e8 diagonal before Black could castle.

Here are a few games, for illustrating purposes:

After 4...b6 5.cxd5 Nxd5 black is deciding to play a semi-tarrasch position were hasn't managed to simplify the position with Bb4ch, Bd2 , BxB. The same type of position can arise after 1.d4 d5 2c4 e6 3Nc3 Nf6 4pxp Nxd5 5e4 NxN 6pxN c5 where white uses a tempo, a3, or Rb1 to prevent the exchange.

The OP can achieve the positions he wants more soundly by playing Be7, and castling getting king save, before playing b6. After 4...Be7 5Bf4 0.0 6e3 b6 there are lines can think about playing Ba6, instead of Bb7 to reduce white's attacking potential exchanging white square bishops, or for tactical reasons. For example, 7pxp Qxp 8Bxc7 Bb4ch 9Nd2 Ba6.

The easiest way (in my opinion) is to finish your king-side development first and castle.

Then, develop your queen knight to d7, as it invariably goes to that square, anyway. (You're going to be moving your c-pawn at some point, so playing it to c6 here would only block your own pawn).

Then you can play ...b6, to put your bishop on b7. (Saving the development of your queen bishop as your last minor piece.)

If White plays c4xd5 at this point, then you just recapture with your e-pawn. This leaves you with a pawn on d5.

"But wait!" you say. "If my bishop is going to b7, won't my pawn on d5 actually block the b7 bishop?"

Nope! Actually the opposite: your b7 bishop will support your central d-pawn, which is a big thorn in White's position.

White wants central control and expansion, but Black's d5 pawn denies it. It also supports an eventual leap of Black's knight from f6 to e4, where Black's knight will enjoy a powerful outpost.

And after your bishop is developed on b7, then you can continue with the normal plan: c7 to c5.

This is the point where you can start looking at engine suggestions, and database games, to get a good feel of what kind of play for Black comes next.

(The same goes for any other variations White may play, along the way.)

The QGD can be tricky when you're still learning it (and it's easy to go wrong if you get mixed up). But the more you play it and study it, the easier and simpler it'll get.

Hi guys, thanks for all the help your suggestions have been brilliant!

Thanks for the suggestions @king5minblitz119147, I’ll look through those moves and the tartakower variation. 

@Thrillerfan, thanks for all your lines to. I’ll make ensure to learn the orthodox and lasher quite well. Nice to know there are good alternatives to the Slav. 

@TwoMove and @laladeer thanks so much for explaining to me why my idea doesn’t work, input from better players who understand the game so much better than me is always really good. I would have struggled to get explanations that detailed (yet simple for me to understand ) elsewhere. Having master games from databases i don’t have is especially helpful, so thanks for putting the work in for that!

@HobbyChesser thanks to you also for explaining everything so simply and concisely, really breaks down the complicated concepts into simple ideas. The boards you included also really help me visualise everything. 

Everyone's been super helpful and I’ve got some good QGD homework to do now based on everyone’s suggestions! 

A move like 4...b6 is usually seen in a QID, not a QGD. The move tends to be better there, as not only it weakens the c6 square in a QGD (as others pointed out), but white may try to get some activity on that same diagonal with g3.