The End of the Golden Era of American Chess
theatlantic.com
theatlantic.com
That said, let’s be clear. The golden era of American chess, and chess in general, is happening now and only getting better. I give credit to to open-source projects like LiChess and Lc0, tournaments like LiChess Titled where mere mortals get a chance to challenge Magnus, and the rise of chess streaming as a thing.
Heck, I’ll concede that the golden era of classical chess is over. But it’s been over for a while, pretty much since computers advanced opening theory and took all the wins out of classical. RIP classical chess. But for those willing to think on their feet, the golden era of blitz and bullet is here :-)
Competitive players can now prepare easily for individual opponents, which is cool and used to only be feasible for pros.
For pros, it means that everyone plays the same safe openings, because they’re all “booked up” thanks to engines. Also many super sharp tactical lines have been conclusively decided by engines so can no longer be played. End result is world championships like the last one of all draws, which was just pathetic.
Most of the GMs I’ve asked about this in the last year have expressed that they no longer love playing classical chess because it’s so boring. Hence the rise of legit Chess960 and rapid tournaments at the top level.
Yeah, this sucks. We've reached local maxima.
Something simple like a draw is 1/4 point for white, 3/4 point for black, might re-invigorate matches in classical format.
I do wonder though, in a drawn situation, can't one player push for a win if they are ahead on time, which one of the players is sure to be? Why would anyone accept a draw if they are ahead on time?
This is such a large amount of time, and the players get rest days with loads of time in between, that preparation becomes a huge factor. Players use top engines as well as databases (of every single move played by their opponent in tournament play) to cover every possible sharp line, leading to a lot of extremely safe (and correct) openings.
All of the above gets thrown out the window in the fast time controls (blitz and bullet). Players fall back on a mix of intuition and calculation and the games become highly dynamic, often with multiple opportunities for both sides (in a close match, rating-wise). Needless to say, it's a lot more fun to watch than a dozen draws in a row.
As an added point, the higher in rating you go, the more draws you see (against evenly matched opponents) at the classical time controls. This is because, while the rules of chess haven't changed with respect to what constitutes a draw, the number of draws by agreement goes way up. At the highest levels, there are many lines which can lead to a situation where neither player is willing to make even a slight concession to their position, thus they agree to a draw or a draw is forced by threefold repetition.
Other tournament time formats are "rapid" with 10 to 15 minutes per player, "blitz" of 3 to 5 minutes and "bullet" with 1 to 2 minutes.
I have no experience in the format, but I imagined that patterns would emerge (and Chess players are already good at spotting transpositions), and people would start studying a different form of opening theory.
I expect Chess960 to rise significantly in popularity and relevance as classical chess declines, at which point theory will become developed.
But that will take a few years, and even then, being booked up in 960 will only provide a small edge compared to classical.
AlphaZero and Lc0 have played many beautiful games against Stockfish. All three of these engines are far stronger than the strongest grandmasters. That implies that the drawishness of classical chess has to do with the conservatism of top grandmasters, who generally aren't willing to play the dynamic, seemingly risky style of an AlphaZero or Lc0. You have to really trust your own intuition and calculating abilities to play in that style.
Yeah, unfortunately, this is true. But inevitable as you pointed out, as it is with many games that become hyper-optimized due to the rise of computing power (and GPUs being turned into powerhouses as of late).
Fischer did get that part right; that classic tournament chess was dying and/or dead. FRC is a fun variant but the writing's on the wall; blitz/bullet and of course bughouse are all the rage today.
And of course, there's always Jerry to watch [0], who is a pleasure.
Bullet, lol---chess without thinking. It's the thinking I like about chess. The carefully-worked-out combinations, brilliant defensive plans, drawing swindles etc. If people like playing bullet, great, but it seems another game entirely.
People complaining here about the draws in the last world championship: the WCC matches are much too short these days, players won't take risks as there's no time to catch up if they backfire. But "shorter, faster" is supposed to cure everything. Anyway, Karpov and Kasparov had a lot of draws, 17 in a row in 1984, in virtually the pre-computer era.
No offense, only people who suck at bullet say it doesn’t involve thinking. It’s the same game, it involves developing and training your intuition. I could teach anyone how to get good, even you... if you were open to learn that is.
(And had the budget or could trade for coding hours :-)
Edit: Gee, lots of downvotes. Not sure why. Should I have said nothing? I thought the way they spoke to me was utterly unacceptable, as I calmly pointed out. My initial points were mostly obvious things everyone in the chess world knows, including the person I was responding to.
I'm hoping you want a reply here, or else I'm wasting my time ;-) I have found that letting others have the last word is both magnanimous and a great strategy for extracting myself from conversations with people I no longer wish to talk to. You seem like the kind of person who appreciates strategic thinking, which is why I thought you might enjoy that idea.
I'm entertaining getting into bullet chess, what kind of dev work do you need? Presumably for LiChess?
So maybe it's boring for the players, but still fun to watch. Anand's games were super exciting.
After a few months of practicing, though, I really started to enjoy it -- it really _feels_ like time dilation / thinking faster when you complete an entire chess game, to mate, in the range of 2 to 6 minutes. I'm also certain that this practice has helped me a lot with focus when doing technical work. [I'm still a lousy player, but that's beside the point, haha!]
p.s. re : American chess : Nakamura is, in my opinion, one of the greats -- and he [like Carlsen] has that quality about him that gets kids excited to play chess. [On a different note, it's great to have Hou as a role model for girls who want to play - I feel like she's up there with Polgar in the 'inspirational' category]
That was published just following the advent of the USSR which would go on to become the world's powerhouse of chess for many decades, arguably continuing to this day. It seems in the west we've gradually started to reverse that adult-based focus for many things, chess among them: it's a game for kids, that adults can play. We've spent an immense amount of energy, and money, getting chess in schools but I'm not sure we've seen anything like a proportional gain in outcome for it.
For instance this [2] is the home page for the US Chess Federation. They chose of all possible pictures, for their lead image, one of an under 10 year old girl playing, and the strong bias towards scholastic/youth chess is present throughout. The only reason I mention this is because I imagine like many when I was a kid I was reached an age (still a kid) where I was gradually repelled by things I perceived as 'kid things' and drawn to things I perceived as 'adult things.' In my case coding was something I perceived as an adult thing, and that attraction ended up playing a monumental role in shaping my life. I'm quite curious if things would have been different had my school been actively attempting to get kids involved in coding.
This article seems focused on a handful of luminaries, but I'd argue that a "golden era" of a thing is defined by it's blossoming into popularity and a new kind of explosive growth that thing hasn't previously seen, which (and maybe it's because I didn't grow up with Fischer) feels more like what's happening now.
I don't know if chess is more popular now than it's ever been in America (having read Pawn Sacrifice, I believe Fischer made chess fairly mainstream in America for a time), but it feels like more people are playing chess than ever before in the US, and that strikes me as more of a hallmark of a "golden era" than what this article refers to it as.
It has been turned into Steam game with a free demo : https://store.steampowered.com/app/883680/Four_Kings_One_War...
The AI for this game was coded in Rust, by me. Good luck!
What a legend! RIP
A few months ago, one would have thought that we had the best algorithm and that progress would only be slowly incremental for now on... until the approach adopted to best humans at the game of go got applied to chess. Now you have these programs like LeelaChess able to best Stockfish (a program that had been improved over the last fifteen years) after a few months of training. There's definitely a lot more to discover in this game.
That is how I understand it as well. Blue Gene used heuristics such as rook = 5 and knight = 3 and there must be many more complex heuristics or direct approaches that you can use once your machine learning methods get enough computing power.
Each player sets up their pieces in any way they want on the back two rows. The number of possible of possible starting positions for each player is 64,864,800. This means that the starting game position possibility is 4.2e15.
Next is the branching factor. Chess is about 35 possible moves each play for each player. Arimaa, each player makes two successive moves. This makes for a very large branching factor... 17,281 possible moves each turn ( http://arimaa.janzert.com/bf_study/ ). Go has a branching factor of 250.
https://en.wikipedia.org/wiki/Game_complexity has a chart on it... Go has a larger board size and space state complexity, but Arimaa has a larger game tree size and branching factor.
That 35 and 17,281 thing... 35^8 ~= 17,000^3. If you have a computer program that can search all the possible moves 8 deep in chess (Deep Blue went 6-8 typically), it could only look 3 deep into Arimaa (a similar search space is 5 deep in Go).