quick shutdown to Caro-Kann advance? (for black)

BobbyPhisher960

Yeah poucin, we share the same suffering.

Firethorn15

At least the first couple of pages (especially the 5. a3 line, which I wasn't aware of before) were useful.

DeirdreSkye
drmrboss wrote:
Half_Ghan wrote:

What's wrong with 3... Bf5? You can get all sorts of interesting positions depending on how white proceeds.

You can play if you like Bf5. I dont play Bf5 because SF doesnt like it. 

      Of course, thinking on your own , assuming you can, is a well known terrible sin! 

froslap

poucin wrote:

engine slaves... cry.png

Seriously. It's disgusting. 😏

SmyslovFan

All of the top players use engines to help in their preparations, but Anish Giri is probably the one that is most closely tied to his engine. He has an excellent memory, and his lines are very well prepared. Every now and again, he gets a bad position out of the opening because his lines are a bit predictable. But that's rare. 

Players who are not as good as the very best in the world who rely on engines will get clobbered against stronger players in tournament chess. 

There's a difference between using an engine as a tool and being enslaved to the engine evals, arguing that a difference of a few centipawns is actually significant. Both ...c5 and Bf5 are playable. Find lines that you understand and that you can riff on. 

Farm_Hand

Yeah.

Using an engine to show that 3...c5 is playable / that the single Karjakin - Topolov game shown is not the end of the story, is fine.

Saying you don't play 3...Bf5 because the engine told you not to is... completely misunderstanding how engines are used.

Top players have engines too, and they don't just blindly follow them.

DeirdreSkye

    World's top players play 3...Bf5 but for an online patzer with an engine it's not good. He understands chess better than them.

 

 

drmrboss

Everyday stockfish programmers are attempting to upgrade stockfish knowledge. The one I really impress about SF knowledge is recent SF's patch in understanding of opposite color bishop evaluation. Today SF is 500 elo stronger than SF in 5 years ago. Today stockfish is not searching 100 times faster than SF from 5 years ago.   Those elo come from stockfish's better understanding of positonal evaluation and selective searching. In comparision with human, 2500 rated players are not 100 times faster in thinking than 2000 rated players. (but better in positional understanding and selective searching ). If stockfish doesnt prefer that particular move, it means, your coach doent prefer that move, it is up to you , you play your own or you follow your coach.

 

How good is Stockfish's knowledge now?, According to our latest tests, SF dev basic knowledge is rated 1600. (it means stockfish searching in 1 ply only (without searching millions) is as good  as 1600 rated player . If you ask me to play against 1600 player without any search,  i will definately lose to 1600 player. In term of pure basic knowlegde , SF is stronger than me. 

Human use by pattern recognition. In one journal,  a grandmaster has about 200,000 patterns in his brain. Similar to human training, Leela has millions of pattern recognition in her network. When tested her pure knowledge with 1 ply search, her rating is 2000. If you ask a grandmaster  to play with intution without search , (eg, in 10 secs ultra bullet) , i doubt a GM can beat 2000 rated player. 

If I doubt in stockfish's move, I check with Leela. Leela's pattern recognition is well beyond most human level now. 

Farm_Hand

GMs use engines too.

It's not like it's human vs machine.

It's some chess.com amateur with an engine saying to blindly follow SF vs professional players with an engine who don't do that.

drmrboss

Both SF and Leela prefer 3.....c5! grin.png 

null

nighteyes1234
Farm_Hand wrote:

GMs use engines too.

It's not like it's human vs machine.

It's some chess.com amateur with an engine saying to blindly follow SF vs professional players with an engine who don't do that.

 

Either human or computer coach is going to result in different levels of maturity. Myself being computer coached, I find it humorous that human coached players expect perfectness from computer coached...but noob level from human coached. Shows you which they think is superior!

 

Farm_Hand
drmrboss wrote:

Both SF and Leela prefer 3.....c5!  

And you convinced me in this topic that 3...c5 is a lot better than what that Karjakin Topolov game suggested.

But if I were a Bf5 player I wouldn't switch to c5 just because the engine prefers it.

Actually I did rely a lot on the engine when I first made my 1.e4 e5 reperitoire (for black). By following the engine I accidentally chose some pretty drawish boring stuff. After I realized I switched to lines it doesn't like as much, but they're still good, and give black a lot more play, which is what I wanted.

 

 

SmyslovFan

One thing to consider is that top GMs played both Bf5 and c5 long before engines were strong enough to influence their decisions.

Human chess players are pretty good, too. The top players of course aren't so interested in the engine evaluations after only three moves.

Farm_Hand
SmyslovFan wrote:

[humans] of course aren't so interested in the engine evaluations after only three moves.

Well, the dumb ones are.

and they use cell phone stockfish too...

BobbyPhisher960
DeirdreSkye wrote:
drmrboss wrote:
Half_Ghan wrote:

What's wrong with 3... Bf5? You can get all sorts of interesting positions depending on how white proceeds.

You can play if you like Bf5. I dont play Bf5 because SF doesnt like it. 

      Of course, thinking on your own , assuming you can, is a well known terrible sin! 

Hahaha, are you really arguing with a 2000 engine slave who won't ever make any major progress because of staring at stupid numbers? You disappointed me.

nighteyes1234
drmrboss wrote:

Both SF and Leela prefer 3.....c5!  grin.png

 

OK....

4 Nf3 lol.

null

depth: 5000 (give or take)

 

drmrboss
BobbyPhisher960 wrote:
DeirdreSkye wrote:
drmrboss wrote:
Half_Ghan wrote:

What's wrong with 3... Bf5? You can get all sorts of interesting positions depending on how white proceeds.

You can play if you like Bf5. I dont play Bf5 because SF doesnt like it. 

      Of course, thinking on your own , assuming you can, is a well known terrible sin! 

Hahaha, are you really arguing with a 2000 engine slave who won't ever make any major progress because of staring at stupid numbers? You disappointed me.

Chess is extremely massive! To play decent game , first you need to get an idea of

1. positional pattern evaluation with its benefits and risks

2. Thematic plan (or strategy ) to execute your benefits meanwhile preventing own weakness

( 3. .......Bf5 and 3. .....c5 have both benefits and risks, ) People who play Carokan than French is to develop that bad Queen bishop in French).  Black can develop that bisop quickly with bf5, but delay in   developing other king side bishop, king side kt and get weakness in Queen side after c5 push.

3. Move order (In some cases wrong move sequence will lead you fail, even if you play the same moves) 

4. Blunder check (tactics)

 

In previous times, engines are busy in step 3 and 4 cos they are doing unnecessary searching. (For example,   KNN vs K endgame, without prior knowledge, if SF search for 100 billion positions , SF will finally find out that the endgame is draw. However if we put hard coded knowledge into SF that this endgame is draw, SF will save 100 billion move search)

Today SF with hard coded knowledge prune searching  unnecessary moves and go deep into late middle game and early endgame straight away,(depth 40-50).   So, SF already find out the benefits and disadvantages of those positions from step 1 and step 2.

 

Of course chess is very drawish , black can play in many cranky positions and still get a draw .I think both c5 and Bf5 would lead to draw in best play by both sides.  SF saying more cranky position means, you have to sort out more problems before getting a draw. 

 

 

 

drmrboss

There are reasons, why I am impressed with Engines's stretegy. In this game, it is a very common London system wherease the game normally ended as draw or Queen side play. However in this game Leela played King side attack starting with 13. h4!! 

Her strategy was successful cos there is a blockadage of Queen side bishop with her own pawn in e6 that shut down whole queen side of black player. 

null

 

At position 38. It is seemingly opposite color bishop ending with draw. However, everything is locked with Zugwang for black. 

null

 

ilikewindmills
These massive engine lines confuse me. Anyway, I read everything and here's what I have to say:

e6 isn't great. I'd be better off just learning the French, which could definitely be a good idea, but maybe not now.

I'm going to keep trying c5, just looking at all these big lines. They leave me in a bit of a strange position, but I'm okay with that. White seems to have struggles as well, so it's not so bad.

I'm surprised there's been little mention of an f6 push, and I'd like a bit of information into when I'm supposed to do that (if I even am, I just assume it's right eventually).

I like these discussions and I hope they continue for some time.
froslap

ilikewindmills wrote:

These massive engine lines confuse me. Anyway, I read everything and here's what I have to say:

e6 isn't great. I'd be better off just learning the French, which could definitely be a good idea, but maybe not now.

I'm going to keep trying c5, just looking at all these big lines. They leave me in a bit of a strange position, but I'm okay with that. White seems to have struggles as well, so it's not so bad.

I'm surprised there's been little mention of an f6 push, and I'd like a bit of information into when I'm supposed to do that (if I even am, I just assume it's right eventually).

I like these discussions and I hope they continue for some time.

Sounds like a plan. But, one thing to consider, which I stated earlier and these ridiculous engline lines don't tell you, is that ...Bf5 gives you a French structure without a bad bishop. If you play ...c5, your Bishop will be locked in, like the French, but a tempo down. Instead of trying to dissect these engine evaluations, buy the Tal-Botvinnik 1960 book or Jovanka Houska's book on the Caro Kann. I bet you'll learn more from those two books than turning on your engine on move three. As you said in this quote: "and I'd like a bit of information into when I'm supposed to do that (if I even am, I just assume it's right eventually)."