An introduction to the greatest strategy game ever invented

[deleted_user0]

meh you guys are boring.

[Victor_swe]

Defenitly will check this out.

[Victor_swe]

Fair enough. You could join the Secret Hitler crew, though that game lacks a little in depth. TCG's would probably be right up your alley, the upcoming "Artifact" looks pretty promising.
Or you could wait for my awesome AoE card game.

Aoe card game?.
Plz tell me its done

[Goodspeed]

It was scrapped. It's now an equally unfinished digital board game.

[fightinfrenchman]

It was scrapped. It's now an equally unfinished digital board game.

Finish it, or I'll send you to the shadow realm

[Mr_Bramboy]

I guess I have to take advice from the reigning EPL champion.

[Goodspeed]

Yeah. That was all me btw

[Jam]

Interesting that the best strategy game is the one that barely has any rules.

[edeholland]

@Goodspeed What is up with your new avatar?

[harcha]

@Goodspeed What is up with your new avatar?

It must be a game that he is actively playing with someone else on the forum with the speed of 1 move/day

[Goodspeed]

It's a nod to computer Go, which has popularized the marked move. Just one of the many ways AI revolutionized the game. And you would think, it being the 5th move of the game, that humans would've figured out how strong it was in the thousands of years they had been playing. But no, the game's just that hard. Also humans dumb

[edeholland]

Didn't that AI ruin any other 5th move by "proving" that move is very strong?

[Goodspeed]

There's no way to prove it. Pros are still playing other 5th moves too. But yeah, a lot of what we thought we knew about fuseki (large scale opening theory) had to be thrown out the window after AlphaGo. I feel bad for all those people who wrote books that contain information we now know (well, kinda) to be false.

[Horsemen]

Why play a game that can be automated? Would rather play Civ where the AI is too dumb to win

[iNcog]

-- deleted post --

[Goodspeed]

Why play a game that can be automated? Would rather play Civ where the AI is too dumb to win

If Google had dedicated a team of its brightest minds to building an AI for it, it would be shitting on you

[Goodspeed]

There's no way to prove it. Pros are still playing other 5th moves too. But yeah, a lot of what we thought we knew about fuseki (large scale opening theory) had to be thrown out the window after AlphaGo. I feel bad for all those people who wrote books that contain information we now know (well, kinda) to be false.

Not necessarily false in the sense that it was true against human opponents when the meta was what it was. But objectively false as the AI has developed a new meta that is closer to the reality that is playing Go. Am I grasping at straws? yes, yes I am.

We "kinda" know it to be false because the AI prefers other moves, and the AI tends to be right (in any case, it's currently the primary authority on Go strategy). But it's impossible to know for sure. A good sign that the AIs are right is that pros have also switched from "old" patterns to mostly AI-inspired ones, but the AI could still be very far removed from perfect play.

The key to why the marked move in my avatar is suddenly so popular is that the AIs found moves in resulting variations that weren't thought of before, and this made the marked move much stronger than previously thought. The real innovation is in the continuation and it seems unreasonable to dismiss the possibility that, with an even better AI, additional moves can be found that would prove the current-day variations unviable, at which point the marked move could go out of fashion again. Hence the "kinda". You just never know.

[iNcog]

-- deleted post --

[minimoult21]

With all metas/strategy there is always a testing period. Something new comes up and seems to be super strong. Usually given enough time (AI accelerates this because they can run many simulations at a time) there is an answer that forms a new meta. To say a game as variant as GO is solved is ignorant. The AI has way more experience and perfect memory, making it better than any human. That doesn't mean the game is solved.

[Goodspeed]

Added a link to an old but excellent tutorial for beginners
http://playgo.to/iwtg/en/

Are any of you still playing and why not?

[Mr_Bramboy]

and why not?

I wasn't exposed to enough advertisement over the years.

[Goodspeed]

I take full responsibility for this. Will step up my game

[RefluxSemantic]

I was exposed to too much advertising.

[Jam]

I played my first game against a real person. Messed up the top left pretty bad, but didn't lose too badly. https://online-go.com/game/22128963

[Goodspeed]

You actually played quite well, especially considering the extremely fast time control. It would be pretty impressive if you were already beating 14 kyus (and I think you'd actually be close to winning if you had managed to save the bottom left corner). We have like 2 people at our club who are worse than that and they've been playing a hell of a lot longer than even me.
I recommend playing with byo-yomi so you have at least 20 seconds per move even after your time runs out.

I see you've met the "snapback". It's a classic beginner trap I used to fall into a lot, too.

Black seems safe, but white can capture his 3 stones with a counter intuitive move: (assume it's white's move)


https://senseis.xmp.net/?Snapback