Scandinavian defence, which is best in your opinion d8, d6 or Qa5 ?

I prefer 1.e4 d5 2.exd5 Nf6....

Do you know the hardware specs? In fact strength of computer = speed x time. Even though it is blitz it is running on 90 cores PC, at 120 million nodes per seconds.

And do you have any reason or refutation? before telling everyone as stupid as usual? 

What is your better move for black move? Are you claiming that you can move better than Stockfish! I am laughing at you. Show the audience that a live human can beat current Stockfish not at 90 cores , even at 1 core pc.

Tell me with the reason, I like facts and proof! 

I would claim that if IM pfren had access to Stockfish, then yes, he would move better than Stockfish. Humans working with an engine are stronger than the engine on its own, especially when the engine only has a little time to think. Hence why correspondence chess is still a rich and competitive discipline.

To suggest that the Scandinavian is refuted based on a single blitz game is ridiculous.

This.

Ok, I show you how leela will play!

1. Nbd7 is responded by Nc4,  Qd8(only move, if Qc7, Qf3 and Bf4, worse), Ne3.

The same problem with awkward bishop.  and even worse.

Artificial intelligence dont care about space and mobility like traditional engines , they care to limit your piece activities.

If you play, h5, then the same Nc4! and kick Black queen away,

In this position, white is significantly better, developed all pieces . Well controlled centre.

Your h5?? just waste one tempo , delayed in development of pieces and give permanent weakness in king side,

Black has two problem now, Nd6+ thread, and also, as soon as Black do castling, there is f3, g4 attack  is coming

For black there are solid opening like

1.e4,e5,(multiple opening), e..g Berlin, (white dont have any favorable advantage)

or e4 c5 (Sicilian ), multiple counter play

Scandi and phliidor are just inferior to those opening , they are just passive and very difficult to draw. And  AI play white, it is more and more unplayable.

At your level, you should not come to this discussion, you should learn to stop pieces here and there first. 

You wont understand this level of positional play, Sorry.

You attacking me, re read your posts. I am just telling you that it is not your level ,sorry.

At your level,  just look at the board twice  before every move and prevent dropping pieces.

To discuss about the game changer or not?

First , people need to know what is the traditional move (Book moves)? What is the Novelty? Whether the Novelty is better than the book moves? 

However, in this game 7.... h4 is out of book (Novelty).(I know that immediately cos I know most of mainlines of opening ECO 500 up to 10 moves). That novelty leads to significant disadvantage to black player which is a surprise to me.

Note. There are games where Stockfish played out of book and suffer immediate dead within 12 moves. I have no interest in such loss as I know it was Stockfish' mistake. No need for theoretical discussion.

For example.

However in Scandi game Stockfish played book moves corretly up to move 6 until he was outplayed by 7.. h4 (Novelty).I have argument with IM pfren as he assumed that black loss was due to stockfish' mistake.

Rashid was right in the 1950's.

In 2019, Stockfish can play a game in 60 seconds that is objectively far stronger than any game he or anyone else in recorded history to that point had ever played in their lives.

He was talking about how worthless it was to analyze a game where players hadn't had time to think or calculate.  Really not the case for a player who can calculate trillions of moves per second.  This stuff was all sci-fi when that outdated quote still meant anything. 

Even in 60 secs, we know the strength of Stockfish, by looking at the number of nodes SF searched.

Around 1 billion nodes per move is a good sweet point for stockfish cos SF strength is a bell curve. (what I mean is although SF does a lot of mistakes in less than 1 billion nodes per moves, SF reached to a high standard at 1 billion nodes per move)

In cccc, 5+2 blitz, the speed was 120mnps, 8 secs per move hit 1 billion nodes per move, that hit the sweet point of SF.

Note, although 1 core SF hitting 1 billion nodes per move is much stronger than 45 cores SF hitting 1 billion nodes per move, I dont want that detail technical discussion (too much technical  jargons)

https://docs.google.com/spreadsheets/d/1ZAIuHR6n-5JTxKQc0XUSx1jyUrgVEcj8DNLKA7-urBw/edit#gid=413146908<br<

In comparison, 2800 human does much much more mistakes than  SF in 1 billion nodes per move., SF in 1 billion nodes is much much more reliable than a human.

h5 is interesting. My engine said Nbd7 draws 100%. But when you get away from the CCC garbage computer chess, h5 was played in game 47.1 Season 14 TCEC....result draw(which SF saw all the way). You said white would reply 8 Bf4, but Leela played 8 Nc4.

Now granted this is the same Leela version that in game 21.1 went from winning to losing. Leela as white's move plays Kf1 with an eval of +1.77 while SF eval went from +.18 to  -4.46. Leela then went to -6.39 eval....or an 8 point swing.

i like Qa5

Qd6 strikes me as the most interesting and possibly the best. Someone (maybe it was Sadler) wrote something interesting about the way Tiviakov has employed it and developed a repertoire around it with varying degrees of risk/safety depending on game situation, etc.

The "edge" that one side or the other receives doesn't depend on the level of play. Maybe likelihood of it being realized does.

Actually I'm sure white should be better somehow in the Qd6 Scandinavian but I'd love to hear what line you think is such a "large edge" and why you think it's only exploitable at GM level if that is the case.

That's not enough information. Are you claiming the Shirov stuff with f4 g4 is the problem, or Bf4, etc. Ne5 is a complex which includes some variety of variations within.

FWIW if I were facing the Qd6 Scandi I would play the Nf3 g3 stuff as Caruana has done. 

Lakdawala isn't a GM but anyhow AFAIK Kotronias had the most complete material on this opening. I'm neutral about how good of a practical opening it is for black, but it's probably not so bad and there are only a couple really critical lines. 

Wasn't it David Bronstein who championed 3... Qd6?  Maybe my recall is incorrect.

I think Qc6 personally

Mistakes in the comment section!

The Qd8 scandinavian does NOT give white a tempo. Quite the opposite! Black gainst a tempo in the main line and in exchange; white get to but his dark squared bishop wherever he likes. 

The mainline scandinavian:

As we can see here, the queen ended up on d8 either way! Almost all GMs tend to play this way as white; not only did he gain a total of 3 tempos on the queen, but he also managed to force damage to blacks pawn structure. 

Thus, black does not play the Qd8 scandinavian to give away a tempo, but to earn one!

Pros for black: Gains one tempo, Solid pawn structure, white blocked his c-pawn, the queen remains flexible and its very offbeat!

Cons for black: White get to put his c1-bishop on a more active square, good development lead.

Overall: This may be a bit to passive for some players, but it is rock solid! This is also a relatively unexplored line, so if you want to do your own research, this is the line for you. I think that black should aim for an improved caro kann structure by making use of the c8-bishop and putting it on either f5 or g4, trading the bishop, even for a knight is favourable for black because of the e6 and c6 structure.

none of the above

Nf6 move 2