Hikaru + Magnus vs Stockfish
HTML-код
- Опубликовано: 25 сен 2024
- 👕 MERCH ► streamlabs.com...
♟️ LEARN CHESS & PLAY WITH ME ► go.chess.com/h...
🎁 GIVE CHESS ► www.chess.com/...
📽️ GMHIKARU MAIN RUclips CHANNEL ► www.youtube.co...
🎞️ MORE GMHIKARU RUclips ► www.youtube.co...
💜 TWITCH ► / gmhikaru
💖 INSTAGRAM ► / gmhikaru
🐦 TWITTER ► / gmhikaru
✨ TIKTOK ► / gmhikaru
💛 DISCORD ► / discord
💙 FACEBOOK ► / gmhikaru
💪 GFUEL ► gfuel.ly/2LS5Rct and use CODE HIKARU
💚 SUPPORT ► streamlabs.com...
🤣 REDDIT ► / hikarunakamura
━━━━━━━━━━━━━
🎥 Edit by: / dosechess
📧 Business inquiries only: TeamHikaru@WMEAgency.com or teamgmhikaru@gmail.com
━━━━━━━━━━━━━
Music in this video:
Song:
Artist:
━━━━━━━━━━━━━
#gmhikaru #clips #twitch #chess
Stockfish is like: “1. e4?? Blunder, mate in 30.”
178 likes and no replies ? Lemme fix that real quick
178 likes and no replies at all ? Lemme fix that real quick
@@thatguymaty you fixed nothing
Cuz it’s not that funny
@thatguymaty oh my God you're so funny
Asking if a human could best stockfish is like "Could you beat a gorilla in a fight"
I actually like this analogy. Like yea maybe I could survive (= to a draw) but no shot I win
@@jonboybluegaming if you survive to a gorilla it means you kill him lol no other chances unless he fell in love with you but idk if it’s a draw..
@@enacoid93 lmao
@@enacoid93 use a gun
Well you'd actually have better odds against a gorilla. You could get lucky and poke its eye out and the gorilla would submit.
It would be pretty darn cool to play with Magnus against Stockfish and see if you could draw.
That'd be the close equivalent to Levy and Botez trying to beat Hikaru together. It's not going to happen. ELOs can't be added like that.
@@MechPaul btw its Elo not ELO. Its not an abbreviation
@@rykehuss3435 actually☝️🤓
@@terer9270 "actually☝🤓" - 🤓
@Rykehuss " "actually☝️🤓"☝️🤓" ☝️🤓
I would absolutely love to see Hikaru and Magnus working together, discussing moves against stockfish, knowing they can't win, but trying to make it a very good game
I think their problem would be instantly and fundamentally obvious, they would probably not understand computer moves until later in the game I assume, maybe if they changed the way they see chess it would be different story idk
I think they would probably be able to predict the computer's moves to a point, but it would be interesting to see at what points they diverged.
@@shootingblueyes Yeah they could see pretty deep for sure, especially in a long game. I think they can pull of a "lucky" draw here and there. But the engine sees perfectly 30 deep and humans can't do that...
@@DomenG33Kengines don't see perfectly 30 moves deep, because they make mistakes, and they don't calculate all the lines. Stockfish 16 is incredibly good at pruning, so that it can cut down an insane amount of the position's it needs to calculate, but even then it can miss moves, or miss evaluate the position. Different computers see different lines and evaluate positions differently at the same depth.
If it would consider ALL moves it wouldn't get past depth 10. lol@@luisyfonsy711
The roast to Alexandra Botez 💀🤣😭
Hikaru's chance of beating SF 15 is actually higher than Alexandra Botez's chance of beating HIkaru in classical, assuming SF 15 classical rating is 3600. But thats only in theory, in practice both Hikaru's and Alexandra's win chance is effectively 0%. Draw chance a bit higher than that, but not much. Still starts with 0.
repent to God
I didn't really think it was a roast or any sort of disrespect. It's just true. The whole point of the ELO system, in the first place, was to calculate the chance of one individual beating another, and Alex is just a well-known figure in the rating range that illustrates Hikaru's point.
I think the levi comparison would've be better.
Levi has a much stronger chance of beating Hikaru than Hikaru has of beating Stockfish.
Room full of Super GMs vs Stockfish would be an incredible match
classical time controls too
Use Full Super GMs, full Computers and everything- Stockfish will still win..its that much bettah
@@AryanDevkota500 yes but still would be an incredible match
As Hikaru explained, it will be something like a group of 2000 rated players playing against Hikaru, would it be an incredible match, I very much doubt so.
@@AryanDevkota500 Its not about who would win, but rather the match itself. Obviously stockfish would win, but HOW it wins, and HOW it actually crushes top masters of the game is what would be interesting.
especially with these players collaborating together and explaining their ideas and moves outloud, so people can see their thought process, and then analyzing later where they made mistakes allowing for stockfish to beat them.
Hikaru playing Stockfish is like him feeling what it is like to be 400 elo again.
It ain't that much but I get your point.
@@kaivalyakhairnar305 It's that much. Give Hikaru 1 hour and stockfish 1 minute and stockfish will still beat him
@@majestic-domination cuz stockfish plays in less than a sec
@@majestic-domination but Hikaru isnt bad at chess , it's just that Stockfish is much much better , 400 elo is simply being bad at chess
@@advaitvibhute72the point is hikaru is bad at chess in the eyes of stockfish
The one human who can beat Stockfish - Hans Niemann
Hans Nieman be like-
Hold my vibrator...
Chess speaks for itself
@@adnanebelfaquir vibrates for itself*
No wtf 😂
It would be a vibrating game.
New mythical chess player unlocked: Alexander Botez
Rather average player, but a mythical skin.
@@maxzweistein8951 bro💀💀
@@maxzweistein8951 💀💀💀
@@maxzweistein8951 expensive account but cardboard rank
Just imagine Magnus and hikaru in one stream teaming up against stockfish.... Could you ask for a better content than that
Edit("I didn't expect that this will get 1.8k likes")
Edit("now it's 2,6k I can't believe it")
Hello same name and i would rot to see that happen
They'd get destroyed unfortunately. Make 2 1800 players team up vs 1 2800 player, it won't help.
@@blackjack6406 ya thats true but it would be entertaining
@themagnuscarlsen please give the people what they want
In chess boxing and robot wins
When AlphaGo first came out and its strength was still not entirely clear they had five top Go players team up to play against it, discussing moves, playing out variations etc. They lost ofc, but it was still an interesting format to see players collaborate to find the best move. I'd love to see Hikaru and Magnus do something like this
It would make for good content, but it wouldn't help their odds against Stockfish. There have been several attempts at combining multiple chess engines to beat Stockfish/Lc0- all you get is an average performance from the ensemble.
@@A_Swarm_of_Waspcrabs Well yea ofc combining multiple chess engines wouldn't really improve their performance, however combining two similarly rated players can definitely improve their performance compared to one, since they can discuss ideas and prevent inaccuracies as long as one of them sees it. Not that they would win against stockfish, but they would still be stronger than just one of them.
The simul game you're talking about was long after AlphaGo (under the pseudonym Master) destroyed 50 pros in a row online. It's strength was already clear to everyone in the go community, and many players began copying its moves and ideas, such as the early 3-3 invasion of a star point
I'd love to watch Hikaru and Magnus pair up against stockfish. That would be a good stream. Just hearing their discussion and analysis of the moves.
i heard stockfish doesn't wanna play against those losers
That's actually impossible. Because they couldn't agree on the moves. There would need to be a team of GMs to play and vote for best moves, like in Kasparov vs world.
I think Kasparov won that one BTW
@@aspirin02 well i think the problem is that you wouldn't gain much from pairing as they would almost definitely see the same moves 98% of the time, especially when discussing
Hikaru and Magnus trying to draw stockfish together would be epic, please make it happen
@ronalddoctor8622 the video that will attract millions of views on any site of their choosing with ads in place? is that what your asking to be funded? youtube is pre-funded. twitch is pre-funded. This is how the internet works... the only extra funding would be flights if magnus and hikaru are far from one another, but I presume they would be logical enough to try this when they are near one another.
In my experience playing stockfish it likes to make these moves that I don’t understand. It is only 8-10 moves later that I realize the weird move he made completely destroys my position and my plan.
I was expecting this to be about a match with Hikaru and Magnus on one side vs Stockfish. That could be fun to watch, as they communicate their ideas to eachother. Two heads should be better than one, right?
Yes, better than 1 head but pales in comparison to a high quality algorithm on an ok CPU...
2 heads compared to millions of heads...not a chance
Some mathematical background: the numbers in the ELO rating system are designed in such a way that you are 10 times better than your opponenet for every 400 rating difference. In other words, if you are facing an opponent rated 400 points higher, you are expected to get 1 point in 10 games (draws being worth half a point). We add another 400 points on top of that for a total of 800 difference, that becomes an expected 1 point in 100 games. Hikaru says stockfish is 1000 rating higher, that means that for Hikaru to get 1 point, they would need to play on average 316 times.
But there would be a point where the difference is high enough that the denominator becomes infinity no? 1/350 is almost as good as 0 points
@@Harishiv18 I mean... it's 1 in 350, not 1 in 10 billion.
@Lightn0x I think saying stockfish is 800 elo above Hikaru is being super generous too. Stockfish will basically beat Hikaru / not lose every single time.
If Hikaru played a match against stockfish for every millisecond of his life he wouldn't win a single game even if he lived to 1000 years old. My point is that these bots are out of our league and have been for ages. (You probably already know this though lol)
@Shisiiii if Hikaru were to play 316 matches against stockfish, stockfish would result as the winner every time. (Using the math)
That's a weird system, why 400 instead of like 100, 1000, or maybe even 10
You and Magnus with 30 minutes, Stockfish with 5.
Go.
stockfish plays moves in 0,1 sec lmao.
@@chanddd2 no, it needs to calculate the best move
@@aayushpatel5208 Barley tbh it will go depth 20 in a few seconds on a old iphone lol
not even close
5 seconds will be an overkill. We tested SF vs fruit, 2nd best chess engine of 2008 and which was on par with fritz that beat Kramnik. Sf on 8 seconds / game won vs fruit having classical TC with like 500 elo gap, winning 92 games / 100.
I'm remembering the Kasparov vs the world game in 1999
It might actually be an entertaining game of correspondence chess to have "Every 2700+ Classical Player with 1 Day to Agree on a Move", vs "stockfish at the highest depth it can reach in 10 seconds".
Not sure how many players would agree to such a farcical game but I would hope they can at least hold a draw
Considering Kasparov won that game vs. 50,000 people, I'm pretty sure Stockfish could trivially smoke all current GMs.
On my computer, the latest version of stockfish from github reached a depth of 31 before stopping during depth 32, from the start pos. In 10 seconds. On one thread. After a few arbitrary moves, it reached depth 29. Maybe the humans might stand a chance of winning a single game if stockfish had 10 milliseconds per move. Who knows.
Its basically asking a toddler to beat up the strongest gorilla 1 on 1
But Stockfish literally doesn't care either.
Honestly Magnus and Hikaru against stockfish would truly be peak content
They would still lose every game. It is not like having two grandmasters magically doubles their elo rating. They are still only as strong as their strongest player (in other words Magnus).
@@pauldavis5665 content isn't always about winning dude it would be super entertaining to see both of them talk about the positions and joke around and stuff and occasionally win. Also Magnus had beaten stockfish before I think Hikaru also has in the past
@@tristanbryan3067 okay dude
@@tristanbryan3067if you have some weaker version or bad computer and flag it. But in pure skill and understanding of the game stockfish eats both of them. Even consolidation between top20 supergms would be nothing. Machine for a few years right now are far beyond human reach
I want to see a centaur chess tournament, Hikaru + Stockfish vs Magnus + Stockfish
Assuming Stockfish is 3600 Elo and using Hikaru's peak FIDE rating of 2816 his odds of drawing are actually more like 1 in 200
To win it's about 1 in 26,400 but at that point the difference between that and zero is academic
Asking if a human can beat stockfish is like asking if a human can calculate faster than a calculator
Manuel Akanji can!
@@huseyintok02he cannot
Ask him to do 63737228838013783-2838474728229 and a calculator will be faster
@@francy3643 Yes he can!🕶 You can save those random numbers for another day, lad!
@@huseyintok02That's the point. A calculator doesn't care if they are "random numbers", it just gives the result
I would love to see him trying to draw a stockfish
It would be so cool if Hikaru and Magnus teamed up against Stockfish, and it would probably be the most watched chess video of all time
Damn, this stockfish guy must be one smart kiddo
Stockfish : " you foolish humans don't ever think of beating me even in your dream"
It’s like asking pro CS players to beat five spinbot aimbots. It’s just an incapability
didnt he beat a major engine by tricking it into a closed pawn game then getting it to trade its knights and bishops to take his rooks which are worthless in that setting? Seemed very clever.
When you make the first move against Stockfish and it's already found a way to win.
Hikaru is very good at keeping a level headed cold objective analysis of the situation
"I literally understand nothing about chess, compared to stockfish''
Another ''I literally..." tshirt.??
Asking if hikaru and magnus can beat stockfish is like asking if the strongest man can beat a bear or a gorilla. Stockfish isnt human, it is on whole different league
Gms assembling to beat stockfish sounds like power rangers combining powers to become the megazord
and get sqaushed, unfortunately for humanity
Bro is going ham on Botez 💀
hikaru ceiling stockfish
Hikaru and Magnus doesn't need stock fish to win, the stock fish needs them
I mean stockfish is so strong that he beat the 4th dimensonial chesster, chatgpt.
They should take a number of the top players and make them team up against Stockfish in a correspondence game, just like they did with Kasparov vs the world.
Why not stockfish and top players vs exhaustive chess search algorithm? When we have the computational power.
lol, he didn't have to do Alexandra Botez so dirty 🤣
He play stockfish with taking away a minor piece. If he still loses, take away a Queen at the start of the game, etc. To see up to where he can.
An aquarium fish beats two men, watch without registration and SMS
Casually roasting Alex mid conversation
love the part where magnus came on stream and they both teamed against stockfish
How many moves would Hikaru or Magnus survive in a duel with stockfish? According to engine tests, it appears to be around 70-75
According to stockfish tests 💀
Botez catching strays
Botez catching strays in the introduction.
So you're saying there's a chance
Hans would do with 100% accuracy 😂
I know hes talking real shit but everytime he talks it sounds sarcastic lmao
I think he's trying to explain that he is worse at chess than Stockfish.
Thumbnail is such a good idea.
Imagine magnus and hikaru having to debate about each move playing for one side against some challenging opponent. Would give even better insight into their processes when they must discuss with each other towards common goal
When humans have a 70 point gap in IQ, it's hard to communicate. Being on the higher end of that equation, we can understand how it would be hard to communicate with a person with a 50 or 60 IQ level. But we can't really understand how it must feel for a 210 IQ to dumb it down for us because we lack that perspective. Likewise, we lack the perspective of how far ahead of us AI is. We can only use tools like chess ratings to show the gap and extrapolate how that equates to real world results. So when Hikaru says he doesn't understand chess compared to Stockfish, he's not being hyperbolic.
Exactly it's crazy. Magnus or Hikaru basically can never win against stockfish no matter how many games they play even if the number of games played is beyond our comprehension
Magnaru Carlsemura
Didn’t stockfish 15.1 NNUE break into 4000? That would make a 1225 elo difference or something. Or the difference between Hikaru and a 1550. That’s far below national masters level, and more like the guys at your local club who do it competitively though. Yeah that’s a massive ass difference.
Also why do people think having 2 super GM’s together will work? If you have 2 1550 rated players, will they beat hikaru? Hell nah.
They asked bro if he could be stockfish and he started going off on alexandra 😭🙏
Hikaru should play Stockfish until he draws a game.
You can't even get a draw imao. Unless stockfish is down a piece. I know magnus Carlsen bot can beat stockfish if the engine is down a full bishop
Magnus bot is way more stronger than actual Magnus.
Infact it looks like a mini version of stockfish.
See Real Magnus vs Kid magnus bot on Magnus Carlsen's official channel. That bot wins.
What's all the fuss about? Just take this stock fish out of the water and it literally won't be able to breathe. Brilliant move right there
The problem comparing to stockfish is no human can calculate the millions of possible combos that exist after each love. Stockfish set to high depth cannot be beaten.
It's crazy how Hikaru sees Stockfish like we see Hikaru
take a drink everytime he says stockfish.
Bro really said Botez doesn't even know chess to him.
Me, rocky of chess, know who is hikaru and magnus but no idea who is stockfish
well if the super GM says it, I don't know what else to say
What if you could spend 10 hours on each move and compare with Magnus and other GM's. Would you still lose?
If that's the scenario, then they all might stand a chance
Even if they have a full week per move they lose
We infact will lose.....but many IMs,FMs will definitely win with that time I guess
I think a draw would be the outcome
If stockfish plays one of the best openings chances are it will win, otherwise theres a possibility for a draw
Ok but could Hikaru beat stockfish?
In a fight?
Yes, he'd rip it's head off and it it for lunch.
No
@@cheshirecat111 no
Does human have 3 eyes?
No
Does human really have 3 eyes?
No
The person who's aking questions must be from Ohio💀💀💀💀
@@happygood18but is he really from Ohio?
Two worlds collide.
The difference between a World Champion and stockfish in rating is higher than the difference between a WFM and a GM.
I despise these ads.
stockfish be the guy to ask hikaru how the horse moves
10 seconds into the video man is already destroying Feelings and Lives 😂
Stockfish:e4
Human:e5
Stockfish: Blunder. Mate in 40
Chess is a game of engines
" i hate chess , i hate chess ,i hate chess"
Bobby fischer
I think in the future, every move would be a gambit lwading to either white or black winning
Alexandra catching mad strays out here
If you assemble all GM 2700+, they won’t beat stockfish. At least a draw but beating it, just God can..
How to say the exact same thing in 200 different ways for a minute and 17 seconds
Would be fun to see a challenge, Hikaru 20min vs stockfish 1second no ponder and see if he can manage to draw one game out of 10.
Magstocknaru Carlfishemura
Stock fish is a computer, you can’t beat it
Hikaru already drew against and that's probably where the limits of humans are against todays robots at chess.
“Stawkfish”
Botez family was no harmed in this video
Andrew Tang: You underestimate my power
At a 1000-rating difference, that would mean Hikaru should win against Stockfish once out of ~333 games.
However, we know that would be impossible, since Stockfish got its 3600-3800 rating by playing against other extremely powerful chess engines, meaning that Stockfish's actual ELO, much like any of the chess engine's ELO, is on a whole different level and doesn't directly compare to a human ELO rating.
Saying stockfish is better than human is like saying excel is better than human.
i loose to stockfish level 4 regularly LOL
Hikaru throws the best side disses on youtube. Hes gotten Giri Levi botez, still waiting on the Carlson one
Now, now. You could have a brain aneurysm and Botez could beat vegetable Hikaru. But stockfish?
Stockfish in a nutshell: 1.e4 e5 2.Nf3 Nc6 3.Bb5 “Blunder. I have a forced checkmate in 1937492738 moves.”
PLEEEEAAASEEEE YOU AND MAGNUS AGAINST STOCKFISH OR KOMODO IS ONE OF THE MOST EXCITING MATCHES FANS ARE DREAMING WITH
Bro roasted alexander botez in 69 differenr languages 💀
The thing about rating in chess is that it is absolutely non-linear, and more arguably exponential. The difference between a 400-500 is microscopic, compared to the difference of a 2400-2500. Put that into the context of Stockfish being 1000 rating higher, when Hikaru peaked 2816... beating Stockfish is unfeasible.
What is the least handicap should Hikaru get to win against Stockfish from the start of the game?
A Bishop? A Rook? Two Knights? A Queen?
Could there theoretically be a first move for white that, if played perfectly against any possible followup, would definitely lead to a win? If so, an algorithm that has been through all of these followups (practically endless though), would be able to just win as white no matter what. Just imagine two AIs going up against eachother. White moves. Black goes "ggwp".
Actually he is right, because alex vs hikaru, after some Hikaru gets tired and blunder like after playing for hours, while Stockfish won't
if stockfish loses ever it’s because stockfish broke
Daniel, give me coffee