|
that's a really good explanation -- that white is never getting past +1 or whatever with perfect play from black and the advantage probably needs to be bigger than that to win, i can dig that. also congrats on your title!
|
# ? Dec 15, 2021 20:47 |
|
|
# ? Jun 3, 2024 21:42 |
|
Still working on IM, but I've had CCE and CCM for a while. At this point it looks like it is a long rating grind just to get to a high enough rating that will allow me into events where all draws will be enough for an IM norm. Or someone will randomly flag or read b6 as d6 or so on. I actually have 2 IM norms (and one SIM norm even) which is all you need, but there is a game requirement I didn't meet in the two events I overscored in. And all of my lines that would score against people just playing Stockfish's top move haven't worked since NNUE was released. I think I will stop once I make IM. Although there are some possibilities that title requirements will change in the future. I think I would auto-promote to IM in some of the proposals.
|
# ? Dec 15, 2021 23:10 |
|
Do engines use opening databases or do they calculate from move 1? Are there any tournaments for 960 chess and does it even make a difference to them?Sub Rosa posted:Or someone will randomly flag or read b6 as d6 or so on. How can that happen? I assume you don't send moves in handwriting.
|
# ? Dec 16, 2021 00:21 |
|
They have databases for openings and endgames (literally brute force crunching all possibilities for certain ever-increasing number of pieces left), the midgame is impractical to handle in such a way so it just does computer magic to it. Then there's Alpha Zero, which is pure computer magic.
|
# ? Dec 16, 2021 14:52 |
|
Doctor Malaver posted:Do engines use opening databases or do they calculate from move 1? Are there any tournaments for 960 chess and does it even make a difference to them? Without a tablebase, engines can misevaluate endgames with 7 pieces. Especially when the draw or win is super far out. This is partially due to how engines select moves - they frequently ignore entire ideas because searching the entire space isn't practical. Openings have a book that was created through human work with engine assistance. Opening theory continues to develop though, and while there are sound openings there continues to be more sound continuations found. Khorne fucked around with this message at 20:16 on Dec 16, 2021 |
# ? Dec 16, 2021 16:00 |
|
Hah, they did the Karpov-as-final-boss thing for real with a mystery chess player in Spain's Got Talent: https://www.youtube.com/watch?v=dSHm2didVVM&hd=1
|
# ? Dec 18, 2021 19:14 |
|
How have I never heard of Rey Enigma? Just, like, some random IM-quality player wondering around in a morph suit anonymously playing chess. Extremely cool. e: apparently some people think it might be Vallejo Pons
|
# ? Dec 18, 2021 19:32 |
|
Hand Knit posted:How have I never heard of Rey Enigma? Just, like, some random IM-quality player wondering around in a morph suit anonymously playing chess. Extremely cool. Or should I say... an enigma.
|
# ? Dec 18, 2021 19:41 |
|
Doctor Malaver posted:Do engines use opening databases or do they calculate from move 1? Are there any tournaments for 960 chess and does it even make a difference to them? Engines in general do calculate form move 1. It's very common for people to provide opening books for engines to choose their moves from instead, but the strength of those moves over modern NNUE engines are possibly dubious. But then some people will generate opening books from like NNUE at depth 99. Openings are traditionally where engines are the weakest due to the horizon effect (in theory) but I think it is more just something that has been repeated for years than anything very true about about the current crop of neural net engines. There are 960 events on ICCF. I think my 960 elo is higher than my standard elo. There likely are positions in 960 that are a forced win for white. Also maybe some that a forced loss for white? But you don't get norms or titles for 960 and so I haven't played it in a long time. I don't know how NNUE works with it. Maybe it would still be weak in the opening because the Neural Nets wouldn't have been generated with those positions in mind. Of course being that it is centaur chess, on the human side it's common to check the database, and you can have a 960 database as well as regular correspondence database. So in a way the database is your opening book. You can submit moves by typing in notation. But the misread would more happen from reading the engine line wrong, and then just not realizing it is actually a bad move when you see it on the board because you are just trusting the engine knows best or what have you. Or another way it may happen would be you know you have been doing analysis on a game that has the most recent move Ra8, and you know you want to move Ra1 in response. But actually you have two games pending a move where the last move was Ra8 and you submit in the wrong game.
|
# ? Dec 19, 2021 05:03 |
|
I've been working hard to learn to play chess the last several weeks, and if someone came out and said that it's a game designed to torture you while pretending to be fun, I would believe them. When I lose, which I do a lot, it often feels like my opponents are pretty much never wasting a move. Like I have no breathing room before my defense collapses and I lose a major piece. Or, I make a dumb move early on my opponent spends most of the game cleaning up. When I win, I never get to enjoy it because the evaluation pops up and I made 6 blunders and only pulled off a win because my opponent made 7. Or what I thought was a textbook rook-king endgame would have been shattered if my opponent had noticed their bishop was in the position to take out my rook. Or I feel like I'm slowly tightening the noose around my opponent and end up winning on time, only to pull up the evaluation and see that I actually had a dreadful position and was saved by the clock. Or when I try to follow basic principles and avoid an exchange, then have the analysis go "yes, in this position you should exchange queens while down on material, idiot." I have always been terrible at strategy games, so that part isn't new, but being able to see exactly how badly I play and where my mistakes were really turbocharges the potential for self-flagellation.
|
# ? Dec 19, 2021 20:28 |
|
please don't pay much attention to the engine analysis
|
# ? Dec 19, 2021 20:40 |
|
Chess is one of those games where unfortunately the computer knows how to do everything perfectly and can tell you exactly what you did wrong. Suboptimal play is normal for any game you play casually but a robot doesn't pop up after your Halo match to tell you every way you hosed up. Go in with the knowledge that engines rate you at a level you won't hit without basically making it a side job.
|
# ? Dec 19, 2021 21:06 |
|
Baronash posted:When I win, I never get to enjoy it because the evaluation pops up and I made 6 blunders and only pulled off a win because my opponent made 7. This is literally what chess is about, you win when you make fewer and/or less bad mistakes than your opponent. As you get better the kinds of mistakes you and your opponent make will change but the engine will still call you an idiot for them
|
# ? Dec 19, 2021 22:32 |
|
Sub Rosa posted:Engines in general do calculate form move 1. It's very common for people to provide opening books for engines to choose their moves from instead, but the strength of those moves over modern NNUE engines are possibly dubious. But then some people will generate opening books from like NNUE at depth 99. Openings are traditionally where engines are the weakest due to the horizon effect (in theory) but I think it is more just something that has been repeated for years than anything very true about about the current crop of neural net engines. Thanks for the explanation. In all honesty, it doesn't sound like that sport has much left in it since those NNUEs appeared. Human chess likely suffers from the similar effect on the top level and slow time controls. The creative mid-game space between opening prep and end game technique is getting ever narrower.
|
# ? Dec 19, 2021 23:11 |
|
yeah, the computers were calling both nepo and carlsen idiots during a bunch of games in their match. impossible not to have the computer call you out for suboptimal play. just keep practicing with a focus on developing the discipline to spot and avoid blunders and memorize some tactics, and soon you can be rueful about all the folks who seem to have infinite time to study every opening while you're at work instead
|
# ? Dec 19, 2021 23:16 |
|
IMO if you blunder and you don't quickly see the problem when the engine points it out it wasn't a real blunder for your current rating. On the other hand one thing that actually does feel uniquely frustrating about chess compared to other games is how crushing small errors can be. If you're dominating a game of soccer or tennis or Starcraft there's no equivalent to miscalculating one square and hanging a queen.
|
# ? Dec 20, 2021 00:35 |
|
Dias posted:Hah, they did the Karpov-as-final-boss thing for real with a mystery chess player in Spain's Got Talent: He's posted an analysis of the game on his own channel (https://www.youtube.com/c/ReyEnigma, the auto-translation captions seems to do a decent job), and maybe the most impressive part of it to me is that in the split second after he realized he had blundered the rook, he still had the presence of mind to give it away instead of exchanging it so that Karpov wouldn't immediately have a passed pawn. The move made no sense when I first saw it but ended up being incredibly important.
|
# ? Dec 20, 2021 02:27 |
|
Redmark posted:IMO if you blunder and you don't quickly see the problem when the engine points it out it wasn't a real blunder for your current rating. This is honestly the beauty of chess. Just how many real skill levels there are. And how high the ceiling is.
|
# ? Dec 20, 2021 04:18 |
|
Redmark posted:IMO if you blunder and you don't quickly see the problem when the engine points it out it wasn't a real blunder for your current rating. In Starcraft there is, actually. I remember playing terran, walling in, and watching zerglings squeeze into my base because I built one building just a bit off. That's an immediate loss, with the added bonus of probably ruining it for my teammates too. And it happened many times. .
|
# ? Dec 20, 2021 07:32 |
|
Redmark posted:IMO if you blunder and you don't quickly see the problem when the engine points it out it wasn't a real blunder for your current rating. https://www.youtube.com/watch?v=4KSV1_FCmOI fart simpson fucked around with this message at 09:03 on Dec 20, 2021 |
# ? Dec 20, 2021 09:00 |
|
The pain of daily chess: I've been waiting to analyze this game for three weeks. Hoping my opponent concedes and doesn't make me play out this endgame for another two weeks
|
# ? Dec 20, 2021 10:36 |
|
i hate when people deliberately use time to frustrate someone from a losing position. dirty flagging is fine, the clock is a piece etc, but just making someone stew over the remaining time to annoy them is really lovely question about daily chess: the idea is that opening books are okay and computer analysis is not. so with that in mind, chess.com’s openings section, which shows the most popular replies to a position without showing any analysis of advantage, is okay? or no?
|
# ? Dec 20, 2021 10:40 |
|
It's even worse on chess.com with stupid "holiday time" so even after their time runs out they still get to come back and play.
|
# ? Dec 20, 2021 11:30 |
|
jesus WEP posted:i hate when people deliberately use time to frustrate someone from a losing position. dirty flagging is fine, the clock is a piece etc, but just making someone stew over the remaining time to annoy them is really lovely Yes, their openings are fine. It's even accessible with one click from inside daily games, along with an analysis board (no engine but you can just move the pieces around) I don't think my opponent is a sore loser so much as lazy, we were getting in one move a day even when they had an advantageous position.
|
# ? Dec 20, 2021 12:09 |
|
jesus WEP posted:chess.com’s openings section, which shows the most popular replies to a position without showing any analysis of advantage, is okay? or no? website that doesn't allow cheating has built in tool to allow cheating
|
# ? Dec 20, 2021 14:17 |
|
didn't realise you could access the openings db from within the matches, that's cool
|
# ? Dec 20, 2021 14:37 |
|
Turns out learning tactics is cool + good! I had the shittiest day of opponents playing 90%+ accuracy yesterday so it's good to win matches convincingly for a change.
|
# ? Dec 20, 2021 14:50 |
|
err nvm
|
# ? Dec 20, 2021 19:32 |
|
I try not to rely on the analysis board TOO much as it becomes a habit/crutch, but I find it's a good learning experience to see how complicated positions can pan out after a bunch of exchanges. And then you get to make an almost computer-like, seemingly pointless king move because you realise that after the simplification he could potentially be in the line of fire. The opening book for daily games has been my main tool for learning all the different openings. I check out what opening it says we're playing and go look up the theory/main lines on simplifychess or whatever and pick one that appeals. E: also you can report people for "poor sportsmanship" on chess.com if they are stalling out a clearly lost position - that offence is specifically mentioned in their help section although it's probably the least serious of the offences there and unlikely to result in action unless they're doing it in every game they play. Maugrim fucked around with this message at 19:40 on Dec 20, 2021 |
# ? Dec 20, 2021 19:33 |
|
Redmark posted:IMO if you blunder and you don't quickly see the problem when the engine points it out it wasn't a real blunder for your current rating. I sort of needed this and didn't think about it this way. I'll see a bit ol' ?? and not understand that 9 moves down the (perfect) line this loses my queen or some poo poo. Does lichess have an opening book that I'm just missing? Also, is there any way to view centipawn advantage as opposed to "this % of games went white/draw/black"? Hand Knit posted:Also, the fact that the opening database shows results and not evals is precisely why it’s permissible. I was looking more to study openings for advantages as either side. Only one I know cold is the Ruy Lopez and I want to expand that. Looks like it doesn't shop up on the Android app for some drat reason, guess I'll see if it does in mobile Chrome. Thanks! D34THROW fucked around with this message at 21:53 on Dec 20, 2021 |
# ? Dec 20, 2021 20:30 |
|
D34THROW posted:I sort of needed this and didn't think about it this way. I'll see a bit ol' ?? and not understand that 9 moves down the (perfect) line this loses my queen or some poo poo. Lichess has their opening explorer under I think the rightmost main tab (tools?). Also, the fact that the opening database shows results and not evals is precisely why it’s permissible.
|
# ? Dec 20, 2021 21:33 |
|
Haven't seen a post about this, but Agadmator's Christmas arena starting in 20 minutes has a pretty insane (crypto) prizepool with 1 bitcoin for first place: https://lichess.org/tournament/WMEZdyxr. 2 hour arena for a ~$70k prize pool Nakamura, Firouzja, Zhigalko, and Naroditsky are usernames I recognize. According to Reddit, the following are also partipating: Bortnyk, Minh Le, Giri, Andreikin, Dubov, Hansen. Magnus still hasn't joined, but he's no stranger to Lichess arenas and he was playing bullet last night. The money might make this a pretty fun watch, especially towards the end. edit: Magnus is in Pengwin fucked around with this message at 19:45 on Dec 21, 2021 |
# ? Dec 21, 2021 19:42 |
|
Pengwin posted:edit: Magnus is in The servers, however, are not.
|
# ? Dec 21, 2021 20:10 |
|
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
|
# ? Dec 21, 2021 20:20 |
|
lol some rando just berzerked against Hikaru. It did not go well
|
# ? Dec 21, 2021 20:40 |
|
After watching what happened to Wesley So I don't think the opening made a difference there.
|
# ? Dec 21, 2021 20:42 |
|
Salt Fish posted: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: It's not specifically a chess thing, crypto companies have a lot of money coming into them and they're spending it on ads all over the place
|
# ? Dec 21, 2021 20:50 |
|
Salt Fish posted: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:
|
# ? Dec 21, 2021 20:52 |
|
this tournament loving rules haha, hikaru vs rosen was pure content
|
# ? Dec 21, 2021 21:08 |
|
|
# ? Jun 3, 2024 21:42 |
|
is there a link to the current tourney? this one seems to link to the previous one or something https://lichess.org/tournament/WMEZdyxr
|
# ? Dec 21, 2021 21:16 |