We need more amateurs to post their annotated games.

when did I started loosing on this one :(
1. You did not play 2.c4 That is whites best move after 1.d4. The move c3 is often better in open play to support the push d4. That has already accomplished here, hence is c3 not a good move because it prevents the knight to go to c3.
2. After c4 has white difficulty in getting space on the queen side. That makes the pawns on the a, b and c file weak, which you can see in the game.
3. Black is entering your queen side. Good strategy, because black has much more space there. You go with your DSB in several moves to h4. In the mean time can black build an attack on the queen side. That manoevring of the DSB costs way too much time. And what is the result?
4. You could have prepared Bd2 with Rb1, protecting that the black queen could capture the pawn base. (Always a very important pawn to protect.)
5. You exchange queens. When you have an open king side, move your pieces to it. You can not do it currently, because your king and queen side are too weak. You have however a threat of Qf4-f5#. If you would do it at move 23 (first time opportunity), then would that lose to Qxe3+ and Bxf4. So you had to prepare it. But at that time you are already lost I guess (after Qxb2 imo).
6. The major threat in the game from move 18 onwards are the pawns at the queen side, not your attack at the king side. You did not respond to that danger. Instead of that did you put your two rooks behind your g pawn (which would have a hard time passing the black square g7.

DSB and LSB: Dark Square Bishop and Light Square Bishop respectively.
Every move has its pro and its cons. When you play c4 you will not be able anymore to defend d4 with the c-pawn obviously. A fianchetto of the black DSB can have more effect after d4 and c4 then after d4 and c3. But generally speaking is the space advantage of white after c4 and the development of the knight to c3 behind the pawn a better position then the moves:
d4, c3, Nd2 or even d4, c3 Na3. Just put those three starting positions on the board and you will see the difference.

This is the stonewall game against computer Impossible, right?
If so, I actually like white's opening choice. The closed position against an engine is probably one of the best ideas, and white seems to have an advantage around move 12.
One possible idea would be Qd1-e1-h4 followed by g4-g5.

And hanging b2 and sacking on f7 are probably the two reasons he lost the game. Before b2 falls black's only queenside plans are to put a knight on b3 - which doesn't do much unless he can somehow prove that the a rook is out of the game, or for black to somehow prepare b5-b4. White meanwhile has quick kingside play.
when did I started loosing on this one :(
According to Houdini 1.5, white had the better position till move twenty three when you made the nasty mistake of trying to exchange queens. 23. Qg4! has the threat of mate on g5.

First round of the US Open, and already was paired way up. Blundered quite early too:
Game is Duncan Shepherd (1950) vs GM Chirila (2600), US Open 2013 Round 1.

Shepi13: interesting game!
Some questions and remarks:
if you want to avoid sidelines, why don't you play 7. Bb5?
8... 0-0 might be the mainline, but a6 is a move to be expected in the Benoni. If not on move 8, then is it a very common move on 9.
I was surprised to see the move Nh5. I agree that it is good for black to prevent the DSB on f4, but the DSB of black is normally saved to come later into play.
Is f5 a good move for black? I always play on the queen side with black, together with trying to make pressure on e4. f5 would give white the chance to open the king side of black.
11... Bf6 is the last move I expect in the Benoni. Amazing, I would without a moment thought have played 11... Qc7.
14. .. Bd4 shows that black can play on the king side in the Benoni as well. Learned new patterns today.
What about 17. Nf3 instead of 17. Qc2? The d4 and e4 pawns can be exchanged and you have field g5 and h4 covered. Nf3 prevents the sacrifice on h3 I suppose. If that is true, then looks Ne5 more accurate indeed.
Thank you for sharing this game.

In my opinion generally the h3 and Bd3 offers a solid advantage against the benoni if black does not go into the immediate b5 line.
Nh5 also surprised me.
Perhaps f5 is not ever a threat, but Bg5 also takes away the square from the knight on f6. Bf6 is a quite common idea, if the bishops are traded white's dark squares are very weak.
Nf3 should probably be about equal, but I would prefer black after the trade of the d pawn for the e pawn, and it is not a position I would like to have against a grandmaster.
Of course, I was fine in the game until I blundered with Rfc1. Rac1 and I am not sure how he would have tried to justify his sacrifice other than the perpetual, and drawing someone with a supplement of 1857 is not what a GM wants in the first round of his US Open.

@L2Gt:
I liked your awareness that you had to make a choice which pawn to return to white: c4 or b4 and you chose imo the best one to return.
From move 10 enz.. I do not understand why white opens the position when he has not developed yet his own queen side. He should have kept the position closed and starting to play 11. e4 and next e5. The benefit for white after e5 is that black does not have many pieces for the defense of his king and that it will not be easy to get his pieces there.
The main reason is that the knight from b8 is played to c6 (hence c5). Nc6 is playing to open the position and to attack on the queen side. I would prefer the piece to go to the king side via Nd7 - Re8 - Nf8 (and may be later Ng6). For sure when white would play e4-e5 chasing the last defender of the king away.
From move 10 onwards would I expect something like:
10. Bd2 a5 11. Nc1 Nd7 12. Nd3 Rb8 13. Bb3 and the situation is unclear - at least to me.
Your 16th move, Qb6, shows that white was opening too fast. It is a good counterattck.
At move 21 makes white a big mistake. I think that is the losing move. It looks to me that Rxa2 is the best move. Pawns on the second (seventh) rank can become way too dangerous. Furthermore prevents it Qxb2.
21. g3 shows that white will go for mate and that he does not want to be bothered by the bishop on f2. But that bishop on f2 is much smaller then a queen on b2.
24. Ng5 surprises. It looks dangerous for black, but he has a good combination at hand. There are at least three reasons why this move is bad:
1. it opens the diagonal a8-h1, which is controlled by the DSB of black.
2. Because white does not move the king, will the king stay on a white square. The king of white should stand on a black square, because he stands now inline of the DSB of black and both knights stand on a white field. That means that when they will be played they go to a black field and will give check on a white field. If therefor white puts his king on a black field, will black have less oppurtinities to attack the white king with those three pieces.
3. Easiest reason of all: If black moves to f2 will he come on a black field and capture a full bishop.
That is the combination played by black until move 28. On that move would it have been better for black to play differently.
Instead of the sacrifice 28... Bf3+ is 28... Qe2+ much better:
28... Qe2+ Kf4 29. Be3# or
28... Qe2+ 29. Kh3 Bg2+ 30. Kxg2 Bxg3+ 31 Kh3 Qf1+ 32. Kg4 Bxh5. All those moves are mandatory for white. The pressure of white is over, because the rook is gone and the field f7 is covered and white has hardly any pieces left.
Thanks for the analysis, very insightful; 2500 computer analysis is always helpful but sometimes an analysis from an 1800 human with reasoning included is even better haha.

@L2Gt:
I had some bad sentences in my post. I rewrote some of the text, because the quality of the language was too painful to see. :-)
From move 10 enz.. I do not understand why white opens the position when he has not developed yet his own queen side. He should have kept the position closed instead of starting to play 11. e4 and next e5 immediately. It needs preparation. The benefit for white after e5 is that black does not have many pieces for the defense of his king and that it will be easy for white and difficult for blackto get his pieces there.Nc6 is playing to open the position and to attack on the queen side. I would prefer the piece to go to the king side via Nd7 - Re8 - Nf8 (and may be later Ng6). For sure when white would play e4-e5 chasing the last defender of the king away.
24. Ng5 surprises. It looks dangerous for black, but he has a good combination at hand. That is the combination played by black until move 28. From move 28 would it have been better for black to play differently.Instead of the sacrifice 28... Bf3+ is 28... Qe2+ much better:
28... Qe2+ Kf4 29. Be3# or
28... Qe2+ 29. Kh3 Bg2+ 30. Kxg2 Bxg3+ 31 Kh3 Qf1+ 32. Kg4 Bxh4. All those moves are mandatory for white. The pressure of white is over, because the rook is gone and the field f7 is covered and white has hardly any pieces left.
The difference between the sacrifice of Bf3 on move 28 and Bg2 on move 29 is that black can put more pressure on the white king. 28... Bf3+ attracts an extra defender for the king, where Bg2+ pulls the white king out of safety into the open field.
when did I started loosing on this one :(
By move 6, I think black has already equalized. You have a bad dark-squared bishop, a backwards pawn of the e file, and your Knight can't go to its favorite C3 square.
Even after you do Nbd2, the squares that Knight can go to are limited. You have to spend several moves simply fixing a losing position and making your minor pieces active.