Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Salt Fish
Sep 11, 2003

Cybernetic Crumb
chess.com has the video feed from the arena, but they also paywall the VOD, so you have to spend 5 bucks to see what happened earlier in the day. Very annoying Danny I know you're a goon please fix.

Adbot
ADBOT LOVES YOU

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Helianthus Annuus posted:

(skipping over 1000 humiliating games) have a look at this game from tonight.

code:
1. e4 1... Nf6 2. e5 2... Ng8 3. d4 3... Nc6 4. Nf3 4... d6 5. Nc3 5... dxe5 6. d5 6... Nb4 7. a3 7... Na6 8. Bxa6 8... bxa6 9. Nxe5 9... Bf5 10. Bf4 10... Nf6 11. g4 11... Bc8 12. g5 12... Nd7 13. Nc6 13... e5 14. Nxd8 14... exf4 15. Nc6 15... Nb8 16. Qe2+ 16... Be7 17. Qxe7# 1-0
check out move 13 for a funny queen trap!



It was very polite of black to begin setting up the board.

Salt Fish
Sep 11, 2003

Cybernetic Crumb

busalover posted:

It's also my first time, and I'm not sure I'll be able to follow all games. Watching chess, especially high-level chess, you're constantly trying to deconstruct the moves, trying to figure out what their overall strategy is... as a newcomer, it's pretty taxing. My brain is already smoking after three days, pretty sure I'm gonna skip a whole bunch and tune back in at the end.

Just imagine that you understand it. It's not like there's a quiz at the end. Anyway, that's my strategy for most chess.

Salt Fish
Sep 11, 2003

Cybernetic Crumb


I watched a youtube video about the rules of chess in May of this year after watching pogchamps. 2130 puzzles later I hit 2000 puzzle rating on lichess. I might starting playing chess some day! Maybe!!

Salt Fish
Sep 11, 2003

Cybernetic Crumb

jng2058 posted:

Welp, that's pretty much ball game. I don't think anyone thought Nepo could come back from two down, but now three down, with two of the losses from his own blunders? No chance. Gotta feel for the guy, no one wants to go out like this. :sigh:

"The prize fund is €2 million split 60% vs 40% between winner and loser"

Salt Fish
Sep 11, 2003

Cybernetic Crumb
One of the coolest unanswered questions is if the opening position is won for white. I think its pretty neat that nobody knows that yet.

Salt Fish
Sep 11, 2003

Cybernetic Crumb

jiggerypokery posted:

If you had chess.com's database you could probably get a long way pinging off moves actual players have made in most positions with just position lookup.

Definitely I can recommend this entire video:

https://www.youtube.com/watch?v=DpXy041BIlA&t=2147s

But the timestamped highlight is where he programs exactly this engine, consuming 21 billion positions from lichess. There are some problems with this approach that are pretty interesting. The most significant problem is that around half of all chess positions have only ever occurred exactly 1 time.

Salt Fish fucked around with this message at 03:00 on Dec 14, 2021

Salt Fish
Sep 11, 2003

Cybernetic Crumb
I would love to hear and discuss why chess is so infested with crypto, but out of respect for what is an excellent thread I am going to ask that over here in the crypto thread:

https://forums.somethingawful.com/showthread.php?threadid=3838405

Salt Fish
Sep 11, 2003

Cybernetic Crumb
After watching what happened to Wesley So I don't think the opening made a difference there.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Here is a fun paper by tom7, the guy that did the bad chess algorithms youtube video. He calculates the longest possible game of chess as having 17,697 moves according to FIDE rules. It presents the oddities of FIDE rules, and weird corner cases. For example, did you know that this is officially a drawn position due to insufficient material?



How about this one?



http://tom7.org/chess/longest.pdf

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Hand Knit posted:

I'm pretty sure the word "gambit" comes from chess and made its way out into the wider world. Myself, I find myself saying "intermezzo" instead of "intermediate" pretty often.

Whoa, that's surprising but sure enough:

"gambit (n.)
"chess opening in which a pawn or piece is risked for advantage later," 1650s, gambett, from Italian gambetto, literally "a tripping up" (as a trick in wrestling), from gamba "leg," from Late Latin gamba (see gambol (n.)). Applied to chess openings in Spanish in 1561 by Ruy Lopez, who traced it to the Italian word, but the form in Spanish generally was gambito, which led to French gambit, which has influenced the English spelling of the word. Broader sense of "opening move meant to gain advantage" in English is recorded from 1855."

https://www.etymonline.com/word/gambit

"Ruy Lopez invented the word gambit" is some s tier chess trivia.

Salt Fish fucked around with this message at 23:40 on Jan 5, 2022

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Dias posted:

Me stumbling into a three move tactic: :v:
Computer tells me it actually was a blunder from move one: :smith:

Suboptimal line and still won. You're showboating.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Are there collections of puzzles, either online or in books, that focus on specific openings? For example, a collection of puzzles that are positions in say, the QGD and the goal of the puzzle is equality or simply progressing into a main line or executing known trap? I was thinking sort of like flash cards for opening prep, or puzzles focused on more ordinary progression of a game.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Yasser was a 2nd to Korchnoi in his first WC match againt Karpov and he has a bunch of crazy stories about it. He's in a documentary called Closing Gambit which was pretty interesting if not a little bit weird stylistically, Like a cheap history channel special.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
All the parts of chess that are fun and entertaining take place outside of the computer's optimized line.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
The wayward pawn attack is very strong.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
I spent like 10m looking for a move that would blunder a mate in 1 for black

edit wait I thought we were talking about the chess.com puzzle of the day what the hell is chessle

Salt Fish fucked around with this message at 00:01 on Apr 2, 2022

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Really struggling with this kind of puzzle. I looked at the solution, and then looked at it on the analysis board for like 20 minutes lol... It's just very challenging for me to visualize. Black to move and win a piece.



The ramp up in difficulty starting around 2050 into 2100 is severe compared to earlier ratings. I'm starting to stall out after a year of puzzles.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Did you calculate the variations where the rooks come off the board? Ultimately what separates one set of captures from the other? essentially, the only reason that it wins a piece one way is that the king participates in the captures, so in this case one of the captures being a check works against white

edit: I still can't visualize it knowing the answer. I think its just practice, I've only been playing chess 1 year, I'll check back next year.

Salt Fish fucked around with this message at 03:17 on Apr 5, 2022

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Oh I should mention that in the puzzle the computer plays bxg7 intermezzo before recapturing the queen. That isn't important.

Salt Fish
Sep 11, 2003

Cybernetic Crumb

poonchasta posted:

Thank you, all.

I have platinum membership on chess.com for unlimited analysis and puzzles. Is it worth it to upgrade to diamond for a ELO 600 ish player? The insights feature that analyzes your games and tells you what you suck at seems like a good feature.

You can get a free trial of diamond here: lichess.org

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Redmark posted:

Occasionally when I analyze my games, the engine will call some opening book moves inaccuracies.
Here's an example from today: https://lichess.org/sdlNh2Ih, where it doesn't like 5. Be2.

I'm aware that engine evaluations shouldn't be taken literally for human play, especially for openings like the King's Indian where it doesn't like Black in general. But my question is, do these evaluations represent the "actual" state of computer play, or is it an artifact of this particular run? I'm not sure exactly how Lichess does its analysis; IIRC they use donated CPU time for decently deep runs. Does that even matter for openings, or does Stockfish use opening books? In this example it wants to play 5. h3 which is a real line but the 4th most common in Lichess's masters database (with the best results for White apparently).

I believe this is because be2 in this position is evaluated at +0.5 at the specific depth the engine runs the analysis. However, if you run this on the lichess stockfish out to depth 33 the computer realizes that the position is actually +0.7. Since before be2 the evaluation was +0.7 and you end up with a swing of -0.2, that qualifies as an inaccuracy. Lichess checks for inaccuracies by comparing two calculations called "winning chances" and determines the proportion of change between them, in this case 10% change is an inaccuracy, 20% change is a mistake and 30% change is a blunder, with special cases hard coded in such as if you miss a mate or blunder a mate:

https://github.com/lichess-org/lila/blob/master/modules/analyse/src/main/Advice.scala

This move is a few centipawns away from being a mistake because the engine doesn't go to enough depth to realize the true evaluation.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
https://www.imdb.com/title/tt7494416/

Closing Gambit is a documentary about Korchnoi vs Karpov. It features a bunch of GMs from that era and was pretty good! Main issue is that its very "history channel" in its presentation but I can live with that.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
lichess stockfish gives it +5 after liquidating into a queen rook endgame 2 pawns up

https://lichess.org/analysis/rn1qk3/pp2n1p1/1b6/2p2p2/1PrP4/2P1Q3/P4PPP/R1B1R1K1_w_-_-_0_1

Salt Fish
Sep 11, 2003

Cybernetic Crumb
I think Nakamura is full of poo poo - he plays 200 games a day 7 days a week. "I'm a streamer now, not a serious player". Yeah right!

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Here is the first (free) puzzle:

https://www.nytimes.com/2022/06/13/crosswords/chess/chess-replay-you-versus-frumkin.html

I hate their piece graphics so here is the same position:
https://lichess.org/analysis/r1b1qr1k/5ppp/2p1p3/p1PpP3/P2N1PQB/2P1R3/6PP/3n2K1_w_q_-_0_1

Salt Fish
Sep 11, 2003

Cybernetic Crumb
What's the longest a player ever held a piece during a tournament before making a move?

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Naroditsky gave a 20 minute talk about the Magnus situation and his thoughts about it. Unfortunately he did it on the same twitch VOD that he later deleted because he had to perform a dance after losing a bet to the Botez sisters.

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Chromatics posted:

what's the summary of his take on it?

I don't know I wasn't able to watch it unfortunately.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
In fairness there are a lot of rapid and blitz events. If you want to watch short time controls you can find top players on twitch pretty much every single day. There's something pretty cool about seeing just how far humans can push perfection in playing 1 game of chess with maximum concentration and time.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
I agree that chess tournaments should not use engine evaluation until after the game is done.

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Sub Rosa posted:

You are going to have chatrooms full of people yelling BLUNDER at every slight inaccuracy anyway though.

True, but I can close the chat. I can't get rid of the eval bar on the stream though, but I wish I could. Like in the Rapid Chess Championship happening at this very minute the eval bar is front and center the entire time both for the viewers and the commentators.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
https://www.youtube.com/watch?v=4jit4VkyFtI

I watched about 50 eric rosen videos but I am only posting this one.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
That ones tough! I see puzzles like this once in a while, it can be annoying, but the rating of the puzzle is based on the puzzle rankings of the people that play it. It could literally come down to "1400 rated puzzle players are more likely to move the rook because they just moved it".

In terms of that specific move my understanding is basically that white's only hope is promoting, and you want both pieces working to defend against it. Moving the bishop blocks your own rook and opens up nice squares for white. If you start with the idea that the rook belongs on f8 and the bishop belongs on f6, then only 1 move order is possible. From my time doing lichess puzzles to me that is like a 1900-2000 puzzle idea? Something like this.

edit: here is the analysis board if anyone wants to play with it
https://lichess.org/analysis/5rk1/3n1qb1/3pQ3/pPpPp2R/P7/1P4PK/7P/5R2_b_-_-_0_1?color=black#1

Salt Fish
Sep 11, 2003

Cybernetic Crumb
stage 4

https://www.youtube.com/watch?v=SkNJDDt2jtg

Salt Fish
Sep 11, 2003

Cybernetic Crumb
I'm a huge dummy and know nothing about chess but if the director of FIDE tweets this that probably isn't a good sign:

https://twitter.com/EmilSutovsky/status/1566876951146647553

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Hand Knit posted:

What was Nakamura's reason for thinking the game was dirty?

In true Hikaru fashion he says that he doesn't think Hans cheated, he just wants to review all the circumstantial evidence with chat... (yeah right lol)

But in the after game interview he claimed to have studied a key position on some very late move and when pressed he claimed that Magnus had that position before which turned out not to be true.

As well Hans was caught cheating on chess.com during money events and admitted it and was banned for a time.

Its all pretty dumb, but its true that his interview was weird and even Alejandro was very surprised at some of the answers. Hans claimed to have studied a specific position during prep and then gave an example move which was a blunder.

Salt Fish
Sep 11, 2003

Cybernetic Crumb

Omnikin posted:

vis a vis the Super GM being unable to explain his position and/or provide analysis/sidelines on a game he just played, it's absolutely bonkers. These games are so engrained in high level players' brains that it shouldn't even be a discussion, just a flat out retelling. When even a lower level guy like a Levy can do three blindfolded games (against cans, mind you) simultaneously and recall position/moves/everything about all three games at the same time, you'd expect a much better player on the heels of the most important and incredible win of his career to be able to spout out endless analysis of the game

I know that noone else is doubting this, so I'm just yelling at the void. But I want to post in what will be a huge scandal as time rolls on (either cheating OR not)

In fairness I can imagine being in a post-game interview in front of camera, in a studio, with hot lights on, and you know thousands of people are watching, and you know people think you cheated... the host is asking super pointed questions on purpose. Maybe you don't give the greatest interview of all time in that situation.

Salt Fish
Sep 11, 2003

Cybernetic Crumb
Also you will never find confirmation of cheating on chess.com. They have a specific policy to never release names or other details.

Adbot
ADBOT LOVES YOU

Salt Fish
Sep 11, 2003

Cybernetic Crumb
...Hans?

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply