he played well if he only learned 30 days before the game, but i really don't get what his plan was here... like if his 'algorithm' was actually 'ready' was he going to... memorize all of chess??
Exactly. I still don't get what he was on about. Was he thinking he was onto something with this revolutionary chess engine? Was he gonna memorise every move ever? He took the same approach as his rubix cube video but that just wouldn't work
@@senior.toucan6370 Unlike learning all of the chess solving a Rubik's cube is largely intuitive, the algorithms given aren't just a solution in themselves, they are more of moves you can make to, for instance, swap a few pieces. Trying to memorize chess is completely pointless, there is no reason to do so because it is also an intuitive game, you look at your position and calculate it. I completely agree
I think the reason why people give him so much hate for this video is it’s fairly insulting to chess. Like, another one of his 30-day challenges was “learn a backflip” it wasn’t “beat Simone Biles on a floor routine.” But when it came to chess he begins by saying he can design software to solve the game as if there haven’t been actual programmers and chess players refining exactly that sort of software for decades. Then he pays however much money for the greatest player alive, maybe ever, to play against him and be nice to him afterwards so he can feel like he wasn’t being a total clown the whole time. The entire thing just reeks of cringy, entitled arrogance.
Yeah, he treated Chess as if it’s some sort of game that an amateur can beat the top player in the world with just a bit of practice and maybe some good luck which is what infuriated a lot of the chess community. And to a person who knows very little about chess watching that you might believe that. After all there are some games where that happens. Think poker for example, there are pros who lose to amateurs all the time. Sure the pros come out on top overall but because the game has a element of luck to it inherently it’s not an absurd thing to do. If you said you’re going to beat Lebron 1 on 1 after 30 days of basketball training though it would be highly absurd.
Very well said. I would add that when he finally noticed his mistake he tried to cover it with random made up BS about some "algorithm" not being ready to feel somewhat less of the fool that he made of himself. I honestly and whole hartedly despise people like that.
So summary: Max: "I will beat Magnus at chess" Max: "Fuck, chess is hard, imma make a computer" Max:"Fuck, I couldnt make a computer" Max: *hangs all his pieces, loses the game* Max: *Incomprehensible mumbling to Magnus*
Max Deutsch is proficient at reaching the point of the Dunning-Kruger curve where you know nothing, but are confident because you don't even know how much you don't know.
I think we witnessed Max Deutsch travel from the Peak of Stupidity on the Dunning-Kruger curve when he brought up his revolutionary algorithm ("chess players surely haven't thought of that yet!"), to the Valley of Despair when he lost his first piece. It was a very teaching journey.
Excuse me sir but don't you mean the Diane Kruger curve? I'm *pretty sure* I'm right considering I've taken multiple psychology 101 courses at numerous online universities.
@@StarryNightGazing under 20 seconds is not complete amateur level. That's basically equivalent to being a fide master. And you're actually not gonna be competing at the top unless you're solving in under 5 seconds at least.
@@user-svqmbiv Not a fide master,you can get under 20 seconds with only a few months of practice.Rubix cube is much easier than chess,as it requires only memorizing and fast reaction time.Yes you can argue that F2L does require some thinking, but it is still absurd to compare it with chess
@@user-svqmbiv lol sorry but you kinda have no idea what you're talking about. A 20 seconds average means you're in the ~70th percentile which is not at all at the equivalent level of national Master which is in the 98.5+ percentile. And the last world championship was won with an average of 6.74.
on the article his "algorithm" said magnus queen attack on h4 was a bad move, but i checked on stockfish and it was the best move so i already know this guy is a bum
the problem with his "algorithm" plan is, even if it worked and it was the best chess computer of all time, he still has to memorize every line for every opening that magnus will do against him. to make it "believable" lets say he memorized 30 moves in for all possible black openings against his opening, and after 30 moves hes up +5.00. magnus will still destroy an 1100 down +5.00 blindfolded pretty easily while singing macarena.
Yeah, even if you had a tablebase for the starting position, there's still such a thing as the human limit. You can't memorize every possible branch, there's a point where you have to take over. Even if your memory could give you a winning position, it doesn't matter. Magnus is too good to get into an irreversible scenario from your memory benchmark. Maybe he'd lose against another grand master with the same superhuman ability, but unless there's forced mate on the board he'll beat an 1100 every time. In chess your memory can only get you so far, hell even raw calculation can only get you so far. Why is it that the earliest computers still lost against the best players, even though they could calculate better? It's because they didn't understand strategy, so if there was no winning tactic, they would effectively make random moves that would land them in a *strategically* losing position. Which is something that can not be understood if you're inexperienced.
Yes, totally true. I think that if you give a 1100 to play a position which is M10 against one like Magnus there are like 95% odds that Magnus will save it and eventually win.
Agreed. Also because the guy’s behavior makes me think that there is no “algorithm ”. He should have either published his code, or tried the challenge again. I’m sure there are a bunch of GMs who would love to see the results. Hell even Magnus might give him another chance, just to see the results.
@@lordinquisitor6651 i found all about this clown when i stumbled upon his blog... there is no algorithm indeed ... he is effectively trying to reinvent the wheel while knowing nothing about the wheel (chess) and throwing around some nonsense like his computer is still choking ? Trying to come up with the algorithm ?? At that point i went from thinking he just dont know what he's talking about to: This guy is trying and failing to fool everyone with some nonsense ''algorithm'' and i have bo respect for him.
@@aminebrahmi8034 You’re most likely right. He tried his best, saw that he’s a buffoon, then decided to tell everyone about his algorithm, so that he can claim „if my algorithm had worked I would have most likely won“… that’s like 14 year old me faking being sick at a test, with the simple goal of being able to tell everyone „if I hadn’t been sick, I’d have aced that test“. The difference is, this guy is an adult, 14 year old me was a kid (but still a better chess player than this idiot)
Could also be maybe Magnus was actually interested? I mean he could get paid a lot by other channels as well, we haven't seen him appear somewhere else. Just what i felt, think he got paid? Yes, surely. Paid a lot, so he took part in this challenge. I don't think Magnus cares that much about money.
@@thelast1iq694 I mean he obviously had nothing to lose. And when he had the time for that nonsense and was in the mood for it, he maybe even did it without taking that much money. And like Levy mentionend the whole thing at least was good for a good laugh and a lot of p.r. for chess.
@@brsbrs9275I know the motto is 'any publicity is good publicity' but that isn't really true in this case. Even if people know his name now and who he is, they're not gonna buy stuff from him just because of that. They won't go 'oh, its the guy who trained for a month and tried to beat Magnus Carlsen, I should buy whatever he's advertising.'
UPDATE: Magnus has now said that he won't be defending the world championship title. Sources say that this may be in correlation to the fact that Max was not participating in the FIDE world championship event. Magnus Carlsen himself has yet to deny these stories. Over to you, John, for the weather.
@@user-rk7rl7tm5w Man I bet Magnus gets ridiculous money offers for appearances from wealthy families with chess-obsessed kids. He probably did accept because he was genuinely curious what a self proclaimed expert-learner can accomplish given a month. Or Magnus got a taste of Max’s ego and wanted to humble him. If he did it just for the money I’d imagine he would’ve played more aggressively for a win. Imagine if he just went into deep Sicilian theory lol
I love the part of the original video where Max says something like: "Since it's very complicated to learn chess and I have 1 month, I plan to learn chess like a computer" Yeah, like that's easier XD
So I believe what he wanted to do is to analyze with a computer all of Magnuses games and see where he can win. And then just memorize everything he possibly can in a month. So he would become like an anti-Magnus chess player as much as possible. However I guess he didn't know there were engines for chess, so he tried to create one himself from scratch in a month. And found out it's hard to calculate every move without any AI or anything. And so he just learned some openings and generally how to play the game, and that's about it. And all the "My computer still calculates" was just complete bs for drama and to seem like he did something there but just was a little too slow. I don't think there was any chance of him actually succeeding in becoming anti-Magnus machine in a month, but he could've done better maybe if he knew exactly the lines he was playing.
It's disrespectful, let's be honest. You can learn new skills in a month. But becoming the best in a game which is played by people who dedicated their life to it... It's delusional.
It would have been more interesting if he actually trained for an entire month under a strong chess coach and then faced Magnus. Instead of just preparing a few minutes prior to the game and not preparing at all.
@@dekippiesip yeh but hiring a coach isn't "extreme learning" for that you need to invent an algorithm that already exists but also fail at inventing it and then I guess memorizing 10 moves of the Ruy Lopez
The thing is, I believe Max knew he has na chance of beating any even semi-professional player, but he knows how to make content. I would also say he wasn't necesserily disrespectful to Magnus or any profesional, he's a content creator, if anything Magnus got a chance to make some money. Or I can be totally wrong, the guy is a lunatic and thinks 30 days of not knowing what You're doing can profit more than dedicating Your entire life to something, idk.
The people were more offended at the wall Street journal's comments, stating that by move 5 Max was winning. That was an exaggeration, just because he was playing with white and was slightly better doesn't mean it is "winning", but since it was technically correct, the non-chess fans were likely to be astounded by this and to be interested in the story. The main reason that video got so much attention IMO is clickbait.
As an extreme learner he must have memorized a lot of openings. He knew he was playing white and what he will start with. So very easy to get the first few moves right. Then Magnus plays something that is not a main line. That means white has a slight upper hand, but has no book theory to follow. Now he just plays like a normal chess player and gets totally crushed by Magnus.
I don't hate Max for thinking he could beat Magnus. I hate him for spinning the story into that his method works and he actually have a winning advantage at any point at all during the match.
That, together with him saying the 'first 5 moves' were good, truly shows how little he actually understands chess. And that's even admitting Magnus wasn't technically wrong saying maybe the first 10 moves were 'decent'. Not that Max would actually know, which is precisely my point. There also never was an actual attack by white.
@@DeepfriedBeans4492 not even just to a super GM. Playing a decent opening kind of expected at even an average chess level. You really can’t win in the opening. You can only not lose.
during my very first week of learning chess i genuinely remember thinking 'if all my pieces are defended at all times i don't understand how i can lose', thought i'd managed to solve the game and couldn't understand why i was so bad. max's video fully gives me that energy
My version is :: if i keep making moves with my knight forward and backwards without disturbing the initial setup, how would anybody invade. If they come too close I'll kill them, easy.
That play style is one of the most satisfying to played against as attacking maniac who loves to sacrifice mostly just gotta find the overloaded piece.
most likely engine developers. Like the people in charge of making the algorithms and stuff. I mean he thinking he can come up with something better just like that
To me that's a telltale sign that he's probably not a great programmer either. Any good programmer (with their ego in check) always checks for existing solutions before starting from scratch.
@@evanbelcher That may be so but Alpha Zero was created when we already have umpteen chess engines to choose from. Maybe if he was a genius Google programmer his approach may have worked.
I wonder why he didn't go play black. Black typically picks which opening they're playing, ex. Spanish, Italian, Sicilian, Scandanavian, etc against e4 or some kind of Indian system, Queens Gambit, etc against d4. He could have chosen black and pick a few variations to do a deep study rather than having to prepare for every possible opening in e4.
I mean, if they’re willing to pay Magnus, and if they want views, having the shitter play against him is perfect. They don’t want to legit test how good he can become in a month, they want views
Probably untrue. 200,000 hours (the minimum to fit your comment) would be almost 23 years nonstop, and Magnus is only 31 years old. The man’s gotta sleep 😂 (I realize that this is overly particular and misses the point, lol)
@@dozydude1159 No doubt Magnus has much more than that. He's been playing/studying chess for multiple hours per day since he was a child. 10,000 would be if he had played an hour per day since he was 8, so just multiply that by whatever you think his daily chess time is. Given Magnus' insane work ethic, an average of 6 hours a day seems reasonable, which would put him over 50,000 hours.
@@dozydude1159 Probably is way more than 20,000 hours. He started chess at 8 and pretty much devoted his entire life to it for 20 more years. After he left school at around 14-15, he was spending 7-8 hours+ per day on chess. 7305 days in 20 years and let's say an average of 5-6 hours a day. Easily 30-40k
chess engine developer here. The thing he was trying to make didn't do any search at all (pretty much relied on "instinct"), and even if it was strong, it would involve having to do over 50k multiplications (with decimal values mind), additions, and possibly even sigmoids all in his head.
on the article his "algorithm" said magnus queen attack on h4 was a bad move, but i checked on stockfish and it was the best move so i already know this guy is a bum
I think all the hate for Max arises from him being like, "I'm gonna try really hard for 30 days and then beat literally the best person in chess when there are thousands of people who have been trying to do the same thing for years." That's where this ambition turns to insult. Had he just decided to try and beat a titled player, there would be less hate. You can't just go out and be like, "I'm gonna be better than all you nerds who play this game in 30 days." Look at the other challenges on his list. None of them involve literally being the best in the world at anything. "Play a 5 minute blues solo". "Solve a Rubik's Cube in under 20 seconds." Like, come on. Max deserved to get flak for conflating these tasks with beating the best player in chess.
To be fair to him, he knew this and did it on purpose - after 11 hard-but-feasible challenges, he decided to push for one which, well, let me quote his own words: "Unlike my previous challenges, this one was near impossible." He comes off as delusional in the highly-edited video about it, not in his actual writing.
@@edanmaor he was delusional. Near impossible is way too optimistic. He chances to win were as hight as if tried to beat the 100m world record but bunny hoping...
To give a tiny amount of lee-way to Max, if you don't play Chess it can be really hard to understand *just how good* Carlsen is. Like, sure you can figure out quickly that the skill cap is high and that there's years of practice for the top players, but it's hard to realize without being exposed to the theory of the game just how difficult it really is. I know that when I first started, my thoughts were similar to "I bet I'd be good if I just stopped hanging pieces." If you genuinely don't know how good someone is at a given skill, it can be easy to convince yourself that "If I just put in some work and learn some tricks and strategies, surely I can at least put up a fight." And looking at the game, it seems like that's kind-of what happened. He learned some tricks and strategies, did decently to not hang pieces in the opening, but didn't really grasp how many other factors go into being a world-class player.
You can see in the code/log output that was displayed, that Max fed the data to a machine learning model (86% Train accuracy...). After hearing that his algorithm "looks at the data without going any deeper" I can only assume that he used simple image recognition to train a model on 8x8 board positions. This is a bad joke. You can just google how chess engines work, to learn that his approach is not sufficient. Even when using machine learning (like AlphaZero or the latest Stockfish) it takes very complicated setups to get a good engine. And of course it is an engine, you can't memorize the algorithm because it is incredibly complex and not humanly understandable. Just from a programming perspective alone he made a fool of himself, even if you ignore the insult to the chess world and Magnus :)
@@NeosSimp agree 100%. this guy is 700 rated. He failed to see a simple capture order with a game that was 22 mins, longer than a regular 10 min rapid game. these mistakes are only made by 1100 in bullet or 3 min blitz (Rare)
@@real.snatch as an 1100 who loves playing 3 minute, that only happens if you're in actual real pressure, as in 7 seconds left on the clock and not even in the endgame yet
The issue with this was the guy’s end goal. If he had said he would take a month to learn then enter an ~1000 elo tournament then it would’ve been fine with everyone. It’s when he starts acting like he can beat the world chess champion who’s worked his entire life to be the best that it gets kind of insulting.
Everybody likes to talk about how disrespectful this was to Magnus, but the real kicker for me is how disrespectful his "algorithm" idea is to the many many people who have collectively spent thousands of hours improving chess-programs: the Stockfish and Leela maintainers and all the people who came before them in that field. Like, becoming familiar with the state of the art (which is kind of what Deutsch did in many of the other challenges) is one thing; attempting to independently innovate the state of the art, starting from nothing, in a 50-year-old field, is just incredible hubris.
I think he may have been trying to find a new way of learning that he thought was faster. There is a difference between trying to insult someone and ignorance. I do have to admit it was a weird video and it didn't really make sense to go about it that way, I find it funny how they tried to make him seem like some genius who can solve a Rubik's cube and all that. Solving a Rubik's cube is a lot like chess, it is more of learning how to do it than having to be a genius to do it.
I don't think his approach was just disrespectful to Magnus, I think it was disrespectful to everyone who has tried to play chess for more than 30 days.
@@Shmared solving a Rubix cube and learning the rules of chess aren't that far apart, but playing chess well, even above the av. player and solving a rubix cube are so far apart. Comparing the number of GMs and speed cubers, chess is alot harder.
@@sebastianoparenti3536 But also compare the skill level, how many people can solve a cube in less than 6 seconds is less than there are GMs. I would still say that chess is a more dedicated sport, as it is older and more popular. The reason I thought your comment was weird is that you compared ALL speedcubers to only the top 2000 or so chess players, a more accurate comparison would be the top 2000 chess players with the top 2000 speedcubers. I mean I could make the argument that there are more chess players than speedcubers so therefore chess is not as difficult, but it just doesn't work like that.
@@Shmared out of 800 millions current chess players there are 1800 something GMs. The probability of becoming a chess GM is 1800/80000000. There have been more than a billion chess players over the period of human history. You are more likely to win a lotto ticket than becoming a GM. Now compare these numbers to Rubik's cube solvers. Lol I myself solved the Rubik's cube in 10 seconds. Solving a Rubik's cube does not take much effort once the algorithm is known. Kids get training to learn those algorithms. Chess does not work like that because you have to play your moves according to what your opponent has played. Rubik's cube does not have 2 humans working against each other. What makes chess so difficult to master? Even for the machines ....? There are more than 10^120 combinations possible in a game of chess. In chess this number is progressive, for example at move 5 there 10 million possible combinations and this goes on. For a comparison, there are 10^83 atoms in this universe. That's why it is impossible to develop a chess engine. It is literally impossible to memorize or record every chess move. We can't even build an AI that can solve chess. Chess isn't a "Skill". Your comparison of 2000 GM to 2000 Rubik's player is just stupid. LOL Chess grandmasters are highly gifted people. You need a super human memory and super human pattern recognition. You can't become a super GM just like that.
The first time I saw Max's video I was waiting for the catch to come with the final challenge. Like, he couldn't possibly be trying to beat Magnus straight up after a month. It's gotta be "up a queen" or something like that, but then the sentence just... ended
Magnus: I agreed to this challenge because I'm genuinely curious about what He's able to do in a month Levi: Aha! What I'm hearing is you got paid a lot Bruh 😂😂😂
As not only a chess player but also a speedcuber I find it hilarious how he just sat on the floor and made random moves while the cube was in a solved state. I did actually watch his Rubik's cube video and I have to admit he did pretty well, but not exactly world-class like he was trying to accomplish against Magnus.
Very nice catch! And indeed 100% true, he was not actually solving the Rubik's cube, he was pretending he did. I guess that was also part of the whole suggestion he had going on 'super algorithm to crack chess', 'smart guy because solving rubik's cube', the whole thing was a total joke indeed.
@@PHeMoX I would like to point out that most speedcubers do this, if they are waiting for something and they have a cube on them they do this all the time. Not to say they don't solve it, I just thought it was funny they included this as opposed to him actually solving it. It was probably just a misunderstanding because most people wouldn't pick up on it.
@@Shmared (fellow speedcuber) I read his blog back in the day, when I was a 40-second solver, and he actually inspired me quite a bit. He managed to get down to sub-20 iirc in a month, which is no easy task. And his blog posts on the path he took to get there were super interesting. To be fair to him, he had 12 challenges (1 a month for a year), and all of them were fairly doable, except the last one (beat Magnus at chess), which he knew ahead of time would almost certainly not really be possible.
@@edanmaor Yeah I think they sound like reasonable goals... except the Magnus one lol. He just didn't take the right approach really. Learning videos are very inspirational and most of them are really good in my opinion, so I can understand from your perspective. If you do like learning content I recommend Mike Boyde, he makes some cool content.
Though it's hard to get a good impression of him from this video, he might still be smart, to be fair. But being smart doesn't make you a good chess player in a month. And it certainly doesn't make you able to beat a smarter person who has played chesd all his life.
@@sigurdh.s8320 Yeah I agree being smart has a broad definition with lots of differnt meanings. What I meant is, that a smart person would acknowledge his lack of understanding about a certain subject and would seek help from people with more knowledge. His approach to learning chess was just ridiculous. He would´ve put up a bigger fight if he only played online chess for a month.
@@sigurdh.s8320 Given that he is not even able to recognise his own incompetence at chess and the sheer difficulty of beating Magnus, i would say he is not smart at all.
Including the past tests that are from different subjects, schools and nations. Magnus could easily have switched from main line Ruy Lopez and purposefully played a bad move as I am sure this dude cannot win against Magnus even with many points in advantage.
Nah, bruh wasn't even thinking about how "okay" he was gonna be he probably just made a slight facial expression when he realized Just how fucked Max was
"I move my knight here, *points with finger* and you twitched a little bit" Telling this to a person who could literally play 20 parallel matches blindfolded xD
“Keeping him a little bit nervous for Atleast 8 moves was a minor victory.” Oh boy 😂 to think you made the world champion who’s 1700 elo points higher than you nervous is hilarious.
@@Nnnn88888 What? Kiliab said that Max is about 1700 elo points below Magnus, so about 1100. I said he’s more like 2300 points below Magnus, so like 500 rated. Don’t know what you’re calculating lol.
I appreciate an enthusiastic learner. But him saying "Keeping Magnus a little bit nervous for at least 8 moves was a minor victory." was pretty much a giveaway that he hadn't been learning any chess in the past 30 days...
Guy says he has nothing memorized 3 hours before the game then plays 10 moves into the main line Ruy Lopez. Sure. Didn't do any studying ahead of time. Pure instinct right there.
I wouldn't mind Max Deutsch's nonsensical statement about algorithms and other stuff, but the WSJ article about his game with Magnus was just the worst. A bunch of sensationalist, fake news stuff from them that made it look like Magnus was nervous or something, when really, Max was just getting dominated. Also Max saying, ''I ran out of time'' when he just got checkmated makes me feel like, ''Does this guy know which reality he's in?''
It's because it was all for marketing. The marketing would be worse if people were more aware of what actually happened here, and the reality of the process of learning chess. The whole purpose was to mistify everything, of the process, of the chess, and of the match.
Max Deutsch knows so little about chess that he didnt realize that Carlsen is in a complete different metaverse of chess. Its called the Dunning Kruger effect and basically means that you're overconfident in your knowledge simply because you have no idea how little you know
His approach was unbelievably wrong. From what I can tell from their fancy graphics which is just for show, is that he trained a neural network to predict the next move on the basis of millions of chess games - a very simplistic approach. Throwing only a neural network at chess is not enough to get a strong engine/bot. That's why it took years to build engines like AlphaZero, Stockfish. On top of making an arguably weak computer "algorithm", the fact that he thought he could somehow make the same predictions as the computer is baffling. How exactly did he expect to memorize or replicate his algorithm. Even if the algorithm worked, how exactly did he plan on getting it into his head. Brute force? I've never been able to figure his plan no matter how much I watch the video. In summary, he touched his nose by putting his hand back around his head.
I'm sure he thought something ridiculous like "64 squares, about 64 parameters to memorize." This is what happens when you take one class on machine learning and think every problem is now trivial to solve.
@@frankjohnson123 yup, because Magnus was at least paid I find Deutsch's Dunning-Kruger based confidence in his "algorithm" and his ML and Software Development skills much more offensive :D. Also, as a German I am mildly miffed about his last name being Deutsch :D.
He wrote a 4 layer deep MLP with no preprocessing or anything. Basically what a first-year undergrad would write after 10 minutes of introduction to machine learning. Unsurprisingly, it didn't work.
The most impressive part of this is that this guy managed to play an OTB game with Carlsen in a month. That's the achievement, and with all this media coverage of it I doubt much of the money came out of this Extreme Learner's pocket. I respect him for this hustle to play Carlsen OTB
If this game had been in your Guess the ELO series, you probably would have assumed that Magnus was a max of around 1500 or so just because he was playing with a noob.
except he might have very well had prior experience. The guy is a hack. Lied about "the algorithm". Seems likely he was lying about being new to chess as well.
@@matthewtelles4809 there is no algorithm, obviously. You're telling me the guy programmed a chess engine all by himself? One that's specifically tailored to teach him how to beat GMs? He's obviously full of shit. Clearly lying.
What would have been good is to have Max play against Magnus, then a GM, then an IM, then an fm, then 1800-2000, then 1600-1800, then 1400-1600 and so on I am pretty sure he still would have lost every single one But Then he might have probably seen how much different chess is than any other games
I think you would be correct. In fact, I have to wonder if he would beat people below 1400 straight away. I'm inclined to say he would lose most if not all games, dropping in online elo rating to below 500 before ever gaining points.
@@PHeMoX that is harsh . I started learning chess about a month ago and not really learning just having fun and I'm 770 elo . I think he could beat a 1000
would love to see that, think the ELO rating works out to beating someone 200 points lower 9 times out of 10, imagine a line of 10 people after Magnus each of whom can beat the next one 90% of the time and on the very end is Max lmao
“I could tell you thought it was going to be fine.” Of course it would be fine. If Magnus was blindfolded, drunk, and getting tasered it still would have been completely fine
A Rubiks cube really isn't difficult to solve. It has a far bigger reputation of difficulty than it deserves. You can learn to solve it in a day. Its just repeating the same puzzle again and again.
Before the algorithm was readily available online it gained a reputation of being very hard to the point where a lot of people in there 40's or 50's think you are some sort of magician for looking it up and commiting it to muscle memory
the most agregious thing they do in the original wsj video is that they never show the actual game. i was waiting for a video covering this for quite a long time. thank you, levy!
Bruh, assuming he actually is 1100, the odds of him beating Magnus "Magnum Dong" Carlsen is literally less than 1 in a billion. If he played a game against Magnus every day until he won, it would take him ~40,000 lifetimes to beat him, statistically speaking. Seriously. Look it up on the Wismuth calculator. Normies have no Earthly idea how high the skill ceiling goes in chess.
There’s a part in the video Levy didn’t include where he explains that algorithm and he says ‘I will ultimately turn my brain into very slow and slightly less sophisticated chess computer’ while Stockfish calculates 120 million positions in one second. He should try to become first violin in Vienna orchestra in 30 days next
The part where Magnus says why he agreed to the challenge and Levy responding with "so you got paid a lot" actually made me bust out laughing so hard it woke my wife and scared my cat lol
Sorry Levy, but the guy claims that he made Magnus nervous for the first 8 moves... Magnus was maybe only nervous that he wasn't delivering a fool's mate. Playing the main line of an opening, isn't making the chess world champion nervous. Magnus has seen that position and all of its variations thousands of times by this point in his life... How would one be nervous about something they are so intimately familiar with...
I actually really like the idea of a normal person spending a month to get as good as possible at something like chess and documenting the process. If done right, it can provide great insights into what it takes to be a grandmaster, how people train for it, what the difference is between a good and a great player et cetera. In other words: it can demistify the vague concept of grandmaster to regular people. However, this guy didn't do it right at all. He should have talked to chess pros, take lessons, copy their preparations and in general learn from the people who are already good at chess. Instead he thought he was smarter than everyone and could trick the system. So he worked alone, came up with some nonsense to tell the viewers and taught the public absolutely nothing about chess I would totally watch a real documentary about taking on the chess world champion, but this is utter garbage.
Michelle Khare cried when she reached 1000 after 8 months of training. This guy did nothing for 30 days and lost against the strongest player on the planet.
"Keeping him a little bit nervous for at least 8 moves was a minor victory.", that's like saying that you make Hussein Bolt worried because you're just 2 metres behind him after a second of race
I remember this story and, whilst I'm trying really hard to avoid being mean here, the only word I deem fitting to classify this entire ordeal is "ridiculous". Really no idea of what he was trying to accomplish
Everyone are criticizing only his plan to beat Magnus, but don't forget his plan-B was to build an algorithm equivalent (or better) than the ones built by big teams of dozens of experts, big budgets and years of work, in less than a month. This might have been even crazier than his plan-A.
I respectfully disagree with Gotham admiring max for trying. Max is a fraud where most of his stunts he tries todo in 30 days are fake or he already new how to do them. He most likely was originally going to play a Magnus Carlson chess app and just memorize the apps move. But when the wsj noticed him and offered him to actually play Magnus he ended up over his head. He made up some weird algorithm that wasn’t ready yet as a prepared excuse for why he was about todo so bad. He honestly deserves the hate.
I can kinda see why this guy gets so many haters. Sorta insulting for somebody to just show up and think they can easily accomplish what others work their whole lives for.
I played a game of hand and brain with some friends a week ago and I've exclusively played Caro against e5 for probably a year. We got into a Spanish and I was lost immediately.
Grüße zurück :D Den Irl stuff hätte ich gerne, wäre lustig. Aber nicht auf dem Hauptkanal für mich. Das wäre mir zu casual im Vergleich zu den Analysen
Isn't this a little like me saying "I will learn to write a program that rivals Photoshop from scratch in 30 days"? (or build an auto engine, or fly an airliner, etc.) Anyone with the remotest understanding of a particular field will know it's not possible. But my problem isn't with Max, it's the excruciating production values of that WSJ video.
My favorite part of that video is that the guy believed he could not just beat Magnus, but also create an algorithm that beats stockfish with brutte force
Found this purely entertaining. Levy struggling not to angry whilst asking us not to get angry 😂 all jokes aside love this channel and all the content and learnt alot.
Won't blame anyone for hating him after listening to the amount of nonsense coming out his mouth. He brought it upon himself. P.S. keeping Magnus what? 13:45
I actually read Max's entire blog where day by day he detailed his process of designing this algorithm. He didn't memorize moves at all. Instead, the model he sought to train is one which evaluates any individual move: e.g., given the positions of all the pieces on the board, if i move my pawn from h4 to h5, is that a positive move? Such a model would have thousands of parameters (real numbers) that he would have to memorize and then add and multiply in his head in order to use. The mechanics of using such a model would then be to assess the board position, and for any move he wanted to consider playing, run it through this process, which by his own admission would takes hours or days depending on his speed of calculation. And again, that's just to give a rough answer to the question "is this move good?" and not even "what is the best move?". As it happened, he blamed the slow speed of model convergence as the limiting factor. But in reality even if he had the best chess engine in the world, and it only required 10,000 parameters (a very low estimate), he would still be hopelessly lost, both because in reality a) I doubt he could memorize that many digits or b) actually perform the mental math on that memorized matrix, and c) even if he could do those things, it would take him far longer than any over the board chess game ever lasts. It was a terrible plan from conception to execution. you can read more here: medium.com/@maxdeutsch/m2m-day-342-is-this-even-humanly-possible-c6c5910d5ab7 or watch Max explain his algo here: ua-cam.com/video/ql3KtXtX-Mk/v-deo.html&ab_channel=MaxDeutsch
"Keeping him a little bit nervous for at least 8 moves was a minor victory."
that's pure comedy right there.
I am an 800 and I can keep it relatively equal against the max engines for 8 moves . Like did he expect he is gonna get mated with e4 😂
Exactly
Making magnus nervous, that guy is very funny indeed
Magnus was terrified of the main line ruy lopez
@@astfgl1265 haahah my thoughts exactly
Levy: "Notice how Magnus did all the damage without developing most of his pieces."
600's watching: "Maybe I am Magnus Carlsen?"
600's make the damage to themselves
🤣
LOL true
I'm a 700 and that is literally not true. Everyone I play develop their pieces and attack, and normally blundering 8 or 9 times
I just have the get the damage part down and I'll be a grandmaster in no time.
he played well if he only learned 30 days before the game, but i really don't get what his plan was here... like if his 'algorithm' was actually 'ready' was he going to... memorize all of chess??
Exactly. I still don't get what he was on about. Was he thinking he was onto something with this revolutionary chess engine? Was he gonna memorise every move ever? He took the same approach as his rubix cube video but that just wouldn't work
@@senior.toucan6370 Unlike learning all of the chess solving a Rubik's cube is largely intuitive, the algorithms given aren't just a solution in themselves, they are more of moves you can make to, for instance, swap a few pieces. Trying to memorize chess is completely pointless, there is no reason to do so because it is also an intuitive game, you look at your position and calculate it. I completely agree
Just overly cocky, and ignorant to how much depth chess has.
@@senior.toucan6370 yeah that didn’t make sense to me either. It was a fun watch IMO
@@ZigZagPower you play chess too?!
I think the reason why people give him so much hate for this video is it’s fairly insulting to chess. Like, another one of his 30-day challenges was “learn a backflip” it wasn’t “beat Simone Biles on a floor routine.” But when it came to chess he begins by saying he can design software to solve the game as if there haven’t been actual programmers and chess players refining exactly that sort of software for decades. Then he pays however much money for the greatest player alive, maybe ever, to play against him and be nice to him afterwards so he can feel like he wasn’t being a total clown the whole time. The entire thing just reeks of cringy, entitled arrogance.
Exactly. Well said
Yeah, he treated Chess as if it’s some sort of game that an amateur can beat the top player in the world with just a bit of practice and maybe some good luck which is what infuriated a lot of the chess community. And to a person who knows very little about chess watching that you might believe that. After all there are some games where that happens. Think poker for example, there are pros who lose to amateurs all the time. Sure the pros come out on top overall but because the game has a element of luck to it inherently it’s not an absurd thing to do. If you said you’re going to beat Lebron 1 on 1 after 30 days of basketball training though it would be highly absurd.
Exactly. His next challeng should be traning a month to beat Jon Jones on a MMA fight
Hi guys my name is Max and my challenge for today is to create a computer program from scratch to beat AlphaZero at a game of chess.
Very well said. I would add that when he finally noticed his mistake he tried to cover it with random made up BS about some "algorithm" not being ready to feel somewhat less of the fool that he made of himself. I honestly and whole hartedly despise people like that.
So summary:
Max: "I will beat Magnus at chess"
Max: "Fuck, chess is hard, imma make a computer"
Max:"Fuck, I couldnt make a computer"
Max: *hangs all his pieces, loses the game*
Max: *Incomprehensible mumbling to Magnus*
You also forgot the part where Max faked solving a Rubik's cube to appear smart, but yes, pretty much exactly like that.
Don't forget that he made Magnus nervous for the first few moves
@@PHeMoX Anyone can solve one if they just look up a strategy
thats exactly what happened
😂😂😂😂
“I would consider myself an extreme learner” how are we supposed to not make fun of that Gotham?
There's a difference between learning an existing skill, and being competitive in a very short period of time.
It beats sitting on your ass and watching youtube
@@dragonbane44 I mean, maybe but at least we are not alleging to be extreme watchers xD
I thought the combination of words actually existed xd
I lost it at the rubiks cube
Dude's basically screaming "i'm smartie"
Max Deutsch is proficient at reaching the point of the Dunning-Kruger curve where you know nothing, but are confident because you don't even know how much you don't know.
It could be
I think we witnessed Max Deutsch travel from the Peak of Stupidity on the Dunning-Kruger curve when he brought up his revolutionary algorithm ("chess players surely haven't thought of that yet!"), to the Valley of Despair when he lost his first piece. It was a very teaching journey.
@@pilazpilaz Not what Dunning Kruger is
Hahahha great comment
Excuse me sir but don't you mean the Diane Kruger curve? I'm *pretty sure* I'm right considering I've taken multiple psychology 101 courses at numerous online universities.
Only problem with Max was his goals. Solving a Rubix Cube under a time limit is not equal to beating Magnus Carlsen
Iirc his objective was something like under 20 seconds which is complete amateur level. World class cubers average at 6 seconds.
@@StarryNightGazing under 20 seconds is not complete amateur level. That's basically equivalent to being a fide master. And you're actually not gonna be competing at the top unless you're solving in under 5 seconds at least.
@@user-svqmbiv Not a fide master,you can get under 20 seconds with only a few months of practice.Rubix cube is much easier than chess,as it requires only memorizing and fast reaction time.Yes you can argue that F2L does require some thinking, but it is still absurd to compare it with chess
@@user-svqmbiv lol sorry but you kinda have no idea what you're talking about. A 20 seconds average means you're in the ~70th percentile which is not at all at the equivalent level of national Master which is in the 98.5+ percentile. And the last world championship was won with an average of 6.74.
And writing the spelling as 'rubix' rather than 'rubik's' is also very not equal to human brain IQ
plot twist: Carlsen is also an extreme learner who has been extreme learning chess since age 6
Magnus actually finished the algorithm. That is the difference.
hahah
😂😂😂 best comment
kasparov octopus knight moment
“i had him nervous for the first 8 moves” lmao magnus absolutely shaking in his boots watching this guy go into main line ruy lopez
Magnus was all covered in ice cold sweat, all his soul trembling in terror.
Hmm I remember seeing a similar comment on Danya’s video…
pretty sure i saw this exact comment on daniel naroditskys post lol
on the article his "algorithm" said magnus queen attack on h4 was a bad move, but i checked on stockfish and it was the best move so i already know this guy is a bum
@@real.snatch You do know that the extreme learner has a positive lifetime record against Stockfish, +10, -0?
Max: I decided to practice MMA
Max: I will challange Khabib
Max at the hospital: I kept Khabib nervous for the first 10 sec
😂😂😂😂
"He is a self-proclaimed 1100" fucking killed me lmao
He is a self proclaimed chess world champion. Lmao
Lmfao I have tears
hes a reality 600 700 tops
dude is like a 700 at best.
1100s I've played against would have crushed him
Didn’t max deutsch end up defeating Magnus forcing him to retire from chess and give up all of his champion titles?
Lol
Dude! The video just came out, don’t spoil it.
Link to the video ua-cam.com/video/dQw4w9WgXcQ/v-deo.html
magnus switched to tetris?
@@omooba6382 i know that link too well
the problem with his "algorithm" plan is, even if it worked and it was the best chess computer of all time, he still has to memorize every line for every opening that magnus will do against him. to make it "believable" lets say he memorized 30 moves in for all possible black openings against his opening, and after 30 moves hes up +5.00. magnus will still destroy an 1100 down +5.00 blindfolded pretty easily while singing macarena.
Yeah, even if you had a tablebase for the starting position, there's still such a thing as the human limit. You can't memorize every possible branch, there's a point where you have to take over. Even if your memory could give you a winning position, it doesn't matter. Magnus is too good to get into an irreversible scenario from your memory benchmark. Maybe he'd lose against another grand master with the same superhuman ability, but unless there's forced mate on the board he'll beat an 1100 every time.
In chess your memory can only get you so far, hell even raw calculation can only get you so far. Why is it that the earliest computers still lost against the best players, even though they could calculate better? It's because they didn't understand strategy, so if there was no winning tactic, they would effectively make random moves that would land them in a *strategically* losing position. Which is something that can not be understood if you're inexperienced.
But he's an EXTREME LEARNER, THO.
Not only that the algorithm won't spot tactics
Very graphic
Yes, totally true. I think that if you give a 1100 to play a position which is M10 against one like Magnus there are like 95% odds that Magnus will save it and eventually win.
The fact that he thought he made Magnus nervous within 10 moves is enough reason to mock the life out of this man xDD my God.
Seriously
Yep. He deserves every ounce of mocking he's received, and then some.
Agreed. Also because the guy’s behavior makes me think that there is no “algorithm ”. He should have either published his code, or tried the challenge again. I’m sure there are a bunch of GMs who would love to see the results. Hell even Magnus might give him another chance, just to see the results.
@@lordinquisitor6651 i found all about this clown when i stumbled upon his blog... there is no algorithm indeed ... he is effectively trying to reinvent the wheel while knowing nothing about the wheel (chess) and throwing around some nonsense like his computer is still choking ? Trying to come up with the algorithm ?? At that point i went from thinking he just dont know what he's talking about to: This guy is trying and failing to fool everyone with some nonsense ''algorithm'' and i have bo respect for him.
@@aminebrahmi8034 You’re most likely right. He tried his best, saw that he’s a buffoon, then decided to tell everyone about his algorithm, so that he can claim „if my algorithm had worked I would have most likely won“… that’s like 14 year old me faking being sick at a test, with the simple goal of being able to tell everyone „if I hadn’t been sick, I’d have aced that test“. The difference is, this guy is an adult, 14 year old me was a kid (but still a better chess player than this idiot)
"Alright, so I heard you got paid a lot, very understandable."
No idea why I laughed so damn hard
Cause its true
Could also be maybe Magnus was actually interested?
I mean he could get paid a lot by other channels as well, we haven't seen him appear somewhere else.
Just what i felt, think he got paid? Yes, surely. Paid a lot, so he took part in this challenge. I don't think Magnus cares that much about money.
@@thelast1iq694 I mean he obviously had nothing to lose. And when he had the time for that nonsense and was in the mood for it, he maybe even did it without taking that much money. And like Levy mentionend the whole thing at least was good for a good laugh and a lot of p.r. for chess.
I'm sure they payed him a decent amount of money
- ua-cam.com/video/MQR7-ZZdR0k/v-deo.html
Only cultured people know Carlsen afterwards lost to Max Deutsch in a game reminiscent of the Evergreen Game
Here's a link ua-cam.com/video/dQw4w9WgXcQ/v-deo.html
@@omooba6382 Why did I not look at the last three letters. XcQ. XcQ. Come on, I should know this...
69th like.
And at this position everyone should checking Agadmator videos.
@@omooba6382 truly it's a great lesson on never giving up
yup.
As an “extreme learner”, the guy sure “learned” something with this.
“yes”
Yes. Getting Magnus on your channel is valuable! Otherwise the content is worthless.
Ye tbh doesn’t seem like he learned anything 😂
Yes he did. 10M views on UA-cam and international advertising on TV worldwide. An we are asking did he learned something?
@@brsbrs9275I know the motto is 'any publicity is good publicity' but that isn't really true in this case. Even if people know his name now and who he is, they're not gonna buy stuff from him just because of that. They won't go 'oh, its the guy who trained for a month and tried to beat Magnus Carlsen, I should buy whatever he's advertising.'
BREAKING NEWS: Magnus has said that he likely won't pursue his championship title unless he goes against Max Deutsch in the championship.
XD
Haha
underrated comment
So Max has to win the candidates. If I play in the candidates, he might get two wins.
UPDATE: Magnus has now said that he won't be defending the world championship title. Sources say that this may be in correlation to the fact that Max was not participating in the FIDE world championship event. Magnus Carlsen himself has yet to deny these stories. Over to you, John, for the weather.
Magnus before the game: "I'm genuinely curious to see what he can do in a month."
Magnus after the game: "Well, that was a complete waste of my time."
He gained some money so I guess it wasn't completely a waste of time
@@randomrandomgo that was def the only reason he went
Magnus after the Nepo match: "Actually Max was not that bad after all. If Ian was 2780 something, I'd give Max at least 2650."
@@user-rk7rl7tm5w Man I bet Magnus gets ridiculous money offers for appearances from wealthy families with chess-obsessed kids. He probably did accept because he was genuinely curious what a self proclaimed expert-learner can accomplish given a month. Or Magnus got a taste of Max’s ego and wanted to humble him.
If he did it just for the money I’d imagine he would’ve played more aggressively for a win. Imagine if he just went into deep Sicilian theory lol
I love the part of the original video where Max says something like:
"Since it's very complicated to learn chess and I have 1 month, I plan to learn chess like a computer"
Yeah, like that's easier XD
Dewa Kipas did it
he must master the gambit and tempo
You know, computers. Notoriously easy to duplicate, especially 3600 chess engines.
So I believe what he wanted to do is to analyze with a computer all of Magnuses games and see where he can win. And then just memorize everything he possibly can in a month. So he would become like an anti-Magnus chess player as much as possible. However I guess he didn't know there were engines for chess, so he tried to create one himself from scratch in a month. And found out it's hard to calculate every move without any AI or anything. And so he just learned some openings and generally how to play the game, and that's about it. And all the "My computer still calculates" was just complete bs for drama and to seem like he did something there but just was a little too slow. I don't think there was any chance of him actually succeeding in becoming anti-Magnus machine in a month, but he could've done better maybe if he knew exactly the lines he was playing.
All the GMs in the word lke "oh shit how did I not think of that??"
This Magnus guy played pretty well. He must be at least 1600.
Idk, I’m 1600 and I could totally beat him. He’s like 1300 MAX
I got that reference
FUNNY ORIGINAL JOKE
@@Sympanet ok, tiny Johnson.
He should maybe try out for competitive chess or smth.
It's disrespectful, let's be honest. You can learn new skills in a month. But becoming the best in a game which is played by people who dedicated their life to it... It's delusional.
It would have been more interesting if he actually trained for an entire month under a strong chess coach and then faced Magnus. Instead of just preparing a few minutes prior to the game and not preparing at all.
@@dekippiesip yeh but hiring a coach isn't "extreme learning" for that you need to invent an algorithm that already exists but also fail at inventing it and then I guess memorizing 10 moves of the Ruy Lopez
Incredibly disrespectful… what a shame
His 40 pull-up challenge is the absolute worst though. Extreme learned my ass
The thing is, I believe Max knew he has na chance of beating any even semi-professional player, but he knows how to make content. I would also say he wasn't necesserily disrespectful to Magnus or any profesional, he's a content creator, if anything Magnus got a chance to make some money. Or I can be totally wrong, the guy is a lunatic and thinks 30 days of not knowing what You're doing can profit more than dedicating Your entire life to something, idk.
The people were more offended at the wall Street journal's comments, stating that by move 5 Max was winning. That was an exaggeration, just because he was playing with white and was slightly better doesn't mean it is "winning", but since it was technically correct, the non-chess fans were likely to be astounded by this and to be interested in the story. The main reason that video got so much attention IMO is clickbait.
e4, and Max has a favorable position against Magnus Carlsen, unbelievable!
As an extreme learner he must have memorized a lot of openings. He knew he was playing white and what he will start with. So very easy to get the first few moves right.
Then Magnus plays something that is not a main line. That means white has a slight upper hand, but has no book theory to follow. Now he just plays like a normal chess player and gets totally crushed by Magnus.
I don't hate Max for thinking he could beat Magnus. I hate him for spinning the story into that his method works and he actually have a winning advantage at any point at all during the match.
He thought he put magnus under pressure for 8 moves . Dude if the game started with magnus down a his pair of rocks he still wins
Bruh, u HATE him? Touch grass dude
I lost it at the part where he says "Keeping him nervous".
That, together with him saying the 'first 5 moves' were good, truly shows how little he actually understands chess. And that's even admitting Magnus wasn't technically wrong saying maybe the first 10 moves were 'decent'. Not that Max would actually know, which is precisely my point. There also never was an actual attack by white.
Yeah lmao, Magnus was not ever remotely uncertain about anything in this game.
Yeah that was arguably the worst part for me, decent moves in the opening means absolutely nothing to a super gm
@@DeepfriedBeans4492 not even just to a super GM. Playing a decent opening kind of expected at even an average chess level. You really can’t win in the opening. You can only not lose.
@@PHeMoX chess is not about creating an attack. It is actually mostly about baiting moves and trading pieces at the right time at the highest level.
during my very first week of learning chess i genuinely remember thinking 'if all my pieces are defended at all times i don't understand how i can lose', thought i'd managed to solve the game and couldn't understand why i was so bad. max's video fully gives me that energy
Good thing the idea to defend all the pieces never crossed my mind!
It's like this simple trick to win any poker tournament: just never go all in! If you are never all in, you can't lose, by definition! :-)
@@OlaRozenfeld Thank you very much! I may share my tournaments' prize with you.
My version is :: if i keep making moves with my knight forward and backwards without disturbing the initial setup, how would anybody invade. If they come too close I'll kill them, easy.
That play style is one of the most satisfying to played against as attacking maniac who loves to sacrifice mostly just gotta find the overloaded piece.
Why did you play this game?
Magnus: Did you know they can make extra long checks if you can't fit all the 0s on a normal one?
Best comment
I like that this wasn't just insulting to chess players, but also to computer engineers
most likely engine developers. Like the people in charge of making the algorithms and stuff. I mean he thinking he can come up with something better just like that
And many engine programmers are also chess players 😂
I don’t think he realizes that chess engines already exist lol
No no you don't understand he made an algorithm… I mean the computer is coming up with the algorithm? Wait… what?
To me that's a telltale sign that he's probably not a great programmer either. Any good programmer (with their ego in check) always checks for existing solutions before starting from scratch.
@@younis24de that’s what “machine learning” means. You can check it on Wikipedia. Max had no clue how to apply it though.
@@evanbelcher That’s right! Why make your own code when you can use prebuilt, pre-tested libraries?
@@evanbelcher That may be so but Alpha Zero was created when we already have umpteen chess engines to choose from. Maybe if he was a genius Google programmer his approach may have worked.
My dude learned an opening and proceeded to challenge Magnus💀
XDD
Yeah lmao
I wonder why he didn't go play black. Black typically picks which opening they're playing, ex. Spanish, Italian, Sicilian, Scandanavian, etc against e4 or some kind of Indian system, Queens Gambit, etc against d4. He could have chosen black and pick a few variations to do a deep study rather than having to prepare for every possible opening in e4.
I mean, if they’re willing to pay Magnus, and if they want views, having the shitter play against him is perfect. They don’t want to legit test how good he can become in a month, they want views
@@Rayden440 because his a tool?
Even if he played chess 24/7 for a whole year, he would still be hundreds of thousands of hours of chess behind Magnus...
Probably untrue. 200,000 hours (the minimum to fit your comment) would be almost 23 years nonstop, and Magnus is only 31 years old. The man’s gotta sleep 😂
(I realize that this is overly particular and misses the point, lol)
@@dozydude1159 It’s probably more than that, though.
@@dozydude1159 No doubt Magnus has much more than that. He's been playing/studying chess for multiple hours per day since he was a child. 10,000 would be if he had played an hour per day since he was 8, so just multiply that by whatever you think his daily chess time is. Given Magnus' insane work ethic, an average of 6 hours a day seems reasonable, which would put him over 50,000 hours.
@@dozydude1159 Probably is way more than 20,000 hours. He started chess at 8 and pretty much devoted his entire life to it for 20 more years. After he left school at around 14-15, he was spending 7-8 hours+ per day on chess. 7305 days in 20 years and let's say an average of 5-6 hours a day. Easily 30-40k
@@dozydude1159 idk which csgo pro has 60000 hours bruh s1mple has 16k and hes the goat
chess engine developer here.
The thing he was trying to make didn't do any search at all (pretty much relied on "instinct"), and even if it was strong, it would involve having to do over 50k multiplications (with decimal values mind), additions, and possibly even sigmoids all in his head.
"3 hours before the match starts and my algorithm hasn't finished yet, cutting it a bit fine"
yeah good luck with that buddy lmao
Yeah but have you considered the fact that he’s an extreme learner? Checkmate, nerd.
on the article his "algorithm" said magnus queen attack on h4 was a bad move, but i checked on stockfish and it was the best move so i already know this guy is a bum
hes a sigmoid alright
I think all the hate for Max arises from him being like, "I'm gonna try really hard for 30 days and then beat literally the best person in chess when there are thousands of people who have been trying to do the same thing for years." That's where this ambition turns to insult. Had he just decided to try and beat a titled player, there would be less hate. You can't just go out and be like, "I'm gonna be better than all you nerds who play this game in 30 days."
Look at the other challenges on his list. None of them involve literally being the best in the world at anything. "Play a 5 minute blues solo". "Solve a Rubik's Cube in under 20 seconds." Like, come on. Max deserved to get flak for conflating these tasks with beating the best player in chess.
To be fair to him, he knew this and did it on purpose - after 11 hard-but-feasible challenges, he decided to push for one which, well, let me quote his own words: "Unlike my previous challenges, this one was near impossible." He comes off as delusional in the highly-edited video about it, not in his actual writing.
@@edanmaor he was delusional. Near impossible is way too optimistic. He chances to win were as hight as if tried to beat the 100m world record but bunny hoping...
To give a tiny amount of lee-way to Max, if you don't play Chess it can be really hard to understand *just how good* Carlsen is. Like, sure you can figure out quickly that the skill cap is high and that there's years of practice for the top players, but it's hard to realize without being exposed to the theory of the game just how difficult it really is. I know that when I first started, my thoughts were similar to "I bet I'd be good if I just stopped hanging pieces." If you genuinely don't know how good someone is at a given skill, it can be easy to convince yourself that "If I just put in some work and learn some tricks and strategies, surely I can at least put up a fight." And looking at the game, it seems like that's kind-of what happened. He learned some tricks and strategies, did decently to not hang pieces in the opening, but didn't really grasp how many other factors go into being a world-class player.
Yeah, he deserved all the hate he got. Maybe even more....
Even a titled player would be entirely unreasonable (but certainly less infuriating). Even an FM or NM is miles above the average player.
You can see in the code/log output that was displayed, that Max fed the data to a machine learning model (86% Train accuracy...). After hearing that his algorithm "looks at the data without going any deeper" I can only assume that he used simple image recognition to train a model on 8x8 board positions.
This is a bad joke. You can just google how chess engines work, to learn that his approach is not sufficient. Even when using machine learning (like AlphaZero or the latest Stockfish) it takes very complicated setups to get a good engine. And of course it is an engine, you can't memorize the algorithm because it is incredibly complex and not humanly understandable.
Just from a programming perspective alone he made a fool of himself, even if you ignore the insult to the chess world and Magnus :)
“I plan to memorise the chess positions objectively using my algorithm” what, all 415 quintillion of them lol
Nah, he's only gonna memorize the GOOD ones, tho. *thonking*
1100 is a bit generous. A guess the elo of this game without knowing who played would have been pretty cool.
@@NeosSimp agree 100%. this guy is 700 rated. He failed to see a simple capture order with a game that was 22 mins, longer than a regular 10 min rapid game. these mistakes are only made by 1100 in bullet or 3 min blitz (Rare)
@@real.snatch as an 1100 who loves playing 3 minute, that only happens if you're in actual real pressure, as in 7 seconds left on the clock and not even in the endgame yet
I love how he really thought that chess is so niche that a world champion isn't even good and probably still uses books instead of computers
Hehe, yes. He seemingly thought something in that direction. Very weird approach.
The issue with this was the guy’s end goal. If he had said he would take a month to learn then enter an ~1000 elo tournament then it would’ve been fine with everyone. It’s when he starts acting like he can beat the world chess champion who’s worked his entire life to be the best that it gets kind of insulting.
Everybody likes to talk about how disrespectful this was to Magnus, but the real kicker for me is how disrespectful his "algorithm" idea is to the many many people who have collectively spent thousands of hours improving chess-programs: the Stockfish and Leela maintainers and all the people who came before them in that field. Like, becoming familiar with the state of the art (which is kind of what Deutsch did in many of the other challenges) is one thing; attempting to independently innovate the state of the art, starting from nothing, in a 50-year-old field, is just incredible hubris.
I think he may have been trying to find a new way of learning that he thought was faster. There is a difference between trying to insult someone and ignorance. I do have to admit it was a weird video and it didn't really make sense to go about it that way, I find it funny how they tried to make him seem like some genius who can solve a Rubik's cube and all that. Solving a Rubik's cube is a lot like chess, it is more of learning how to do it than having to be a genius to do it.
I don't think his approach was just disrespectful to Magnus, I think it was disrespectful to everyone who has tried to play chess for more than 30 days.
@@Shmared solving a Rubix cube and learning the rules of chess aren't that far apart, but playing chess well, even above the av. player and solving a rubix cube are so far apart. Comparing the number of GMs and speed cubers, chess is alot harder.
@@sebastianoparenti3536 But also compare the skill level, how many people can solve a cube in less than 6 seconds is less than there are GMs. I would still say that chess is a more dedicated sport, as it is older and more popular. The reason I thought your comment was weird is that you compared ALL speedcubers to only the top 2000 or so chess players, a more accurate comparison would be the top 2000 chess players with the top 2000 speedcubers. I mean I could make the argument that there are more chess players than speedcubers so therefore chess is not as difficult, but it just doesn't work like that.
@@Shmared out of 800 millions current chess players there are 1800 something GMs.
The probability of becoming a chess GM is 1800/80000000.
There have been more than a billion chess players over the period of human history. You are more likely to win a lotto ticket than becoming a GM.
Now compare these numbers to Rubik's cube solvers. Lol
I myself solved the Rubik's cube in 10 seconds. Solving a Rubik's cube does not take much effort once the algorithm is known. Kids get training to learn those algorithms. Chess does not work like that because you have to play your moves according to what your opponent has played. Rubik's cube does not have 2 humans working against each other.
What makes chess so difficult to master? Even for the machines ....?
There are more than 10^120 combinations possible in a game of chess. In chess this number is progressive, for example at move 5 there 10 million possible combinations and this goes on. For a comparison, there are 10^83 atoms in this universe.
That's why it is impossible to develop a chess engine. It is literally impossible to memorize or record every chess move. We can't even build an AI that can solve chess. Chess isn't a "Skill".
Your comparison of 2000 GM to 2000 Rubik's player is just stupid. LOL
Chess grandmasters are highly gifted people. You need a super human memory and super human pattern recognition. You can't become a super GM just like that.
Give his chess algorithm time, he will be past triple digit rating in no time
The first time I saw Max's video I was waiting for the catch to come with the final challenge. Like, he couldn't possibly be trying to beat Magnus straight up after a month. It's gotta be "up a queen" or something like that, but then the sentence just... ended
Not losing against Magnus starting up a queen would be difficult enough.
Good point - That would have been better content and less disrespectful. Who doesn't want to see Magnus devour someone without a queen.
It is out of character for magnus to agree to play this game, they must have paid magnus a lot of money, good for him
I don’t think so. He played with Ludwig once before, it’s not insane for him to play a silly exhibition match like this one.
Hey. The man needed a new, snazzy watch. Of course he would take the challenge on.
Magnus: I agreed to this challenge because I'm genuinely curious about what He's able to do in a month
Levi: Aha! What I'm hearing is you got paid a lot
Bruh 😂😂😂
“keeping him a little bit nervous for eight moves” … 🤣🤣🤣 Yes, the GOAT was nervous because you played 8 competent moves in the opening.
Magnus scared of Ruy Lopez Theory 😆😆
Magnus be like: who cares about the game, ez win and of course EZ STONKS 🕶
EL MAGNITO EZ Clap
As not only a chess player but also a speedcuber I find it hilarious how he just sat on the floor and made random moves while the cube was in a solved state. I did actually watch his Rubik's cube video and I have to admit he did pretty well, but not exactly world-class like he was trying to accomplish against Magnus.
Very nice catch! And indeed 100% true, he was not actually solving the Rubik's cube, he was pretending he did. I guess that was also part of the whole suggestion he had going on 'super algorithm to crack chess', 'smart guy because solving rubik's cube', the whole thing was a total joke indeed.
@@PHeMoX I would like to point out that most speedcubers do this, if they are waiting for something and they have a cube on them they do this all the time. Not to say they don't solve it, I just thought it was funny they included this as opposed to him actually solving it. It was probably just a misunderstanding because most people wouldn't pick up on it.
@@Shmared (fellow speedcuber) I read his blog back in the day, when I was a 40-second solver, and he actually inspired me quite a bit. He managed to get down to sub-20 iirc in a month, which is no easy task. And his blog posts on the path he took to get there were super interesting.
To be fair to him, he had 12 challenges (1 a month for a year), and all of them were fairly doable, except the last one (beat Magnus at chess), which he knew ahead of time would almost certainly not really be possible.
@@edanmaor Yeah I think they sound like reasonable goals... except the Magnus one lol. He just didn't take the right approach really. Learning videos are very inspirational and most of them are really good in my opinion, so I can understand from your perspective. If you do like learning content I recommend Mike Boyde, he makes some cool content.
I assumed he was just practicing PLLs or something, I do that a lot when I'm waiting
Magnus could crush him drunk with just pawns and maybe a knight even if he trained for a year.
It's funny because it's true
And this idiot thinks he can beat him
Na that is too much at least give magnus 2 rooks and 8 pawns
@@spacerocks9740 Knights are better than rooks, you know...
@@itsdifficulttocreateaperfe9850 against bigenners probably yeah
Levy: let's not be mean to this guy
Also levy: proced to destroy the hell out of him
This guy is the living proof that being able to learn/memorize stuff fast and being smart are completely different things.
Though it's hard to get a good impression of him from this video, he might still be smart, to be fair. But being smart doesn't make you a good chess player in a month. And it certainly doesn't make you able to beat a smarter person who has played chesd all his life.
@@sigurdh.s8320 Yeah I agree being smart has a broad definition with lots of differnt meanings. What I meant is, that a smart person would acknowledge his lack of understanding about a certain subject and would seek help from people with more knowledge. His approach to learning chess was just ridiculous. He would´ve put up a bigger fight if he only played online chess for a month.
@@sigurdh.s8320 Given that he is not even able to recognise his own incompetence at chess and the sheer difficulty of beating Magnus, i would say he is not smart at all.
its the equivalent of instead studying before a test, you try to remember every question from every test from years before
Including the past tests that are from different subjects, schools and nations. Magnus could easily have switched from main line Ruy Lopez and purposefully played a bad move as I am sure this dude cannot win against Magnus even with many points in advantage.
"That's when you knew it would be okay"
My guy... He knew he was gonna be okay before he walked in the building.
Nah, bruh wasn't even thinking about how "okay" he was gonna be he probably just made a slight facial expression when he realized Just how fucked Max was
This dude discovered chess engines.
Lmao. I died at the part when he was talking about " chess algorithms" as something new revolutionary
"I move my knight here, *points with finger* and you twitched a little bit"
Telling this to a person who could literally play 20 parallel matches blindfolded xD
“Keeping him a little bit nervous for Atleast 8 moves was a minor victory.”
Oh boy 😂 to think you made the world champion who’s 1700 elo points higher than you nervous is hilarious.
1700 points? Try more like 2300 lol
@@thegamingbadger5940 Magnus is about 2800 FIDE rating. 2800-1100=1700. Lol.
@@Nnnn88888 What? Kiliab said that Max is about 1700 elo points below Magnus, so about 1100. I said he’s more like 2300 points below Magnus, so like 500 rated. Don’t know what you’re calculating lol.
@@thegamingbadger5940 I guess the 'self-proclaimed 1100' could be incorrect. I thought you were unable to do basic math, sorry.
I appreciate an enthusiastic learner. But him saying "Keeping Magnus a little bit nervous for at least 8 moves was a minor victory." was pretty much a giveaway that he hadn't been learning any chess in the past 30 days...
Guy says he has nothing memorized 3 hours before the game then plays 10 moves into the main line Ruy Lopez. Sure. Didn't do any studying ahead of time. Pure instinct right there.
I wouldn't mind Max Deutsch's nonsensical statement about algorithms and other stuff, but the WSJ article about his game with Magnus was just the worst. A bunch of sensationalist, fake news stuff from them that made it look like Magnus was nervous or something, when really, Max was just getting dominated.
Also Max saying, ''I ran out of time'' when he just got checkmated makes me feel like, ''Does this guy know which reality he's in?''
It's because it was all for marketing. The marketing would be worse if people were more aware of what actually happened here, and the reality of the process of learning chess. The whole purpose was to mistify everything, of the process, of the chess, and of the match.
The article literally stated Max was winning, or in the lead after 8 ish moves. That was painful to read.
@@sigurdh.s8320 Yeah, according to WSJ, Max playing just 8 opening moves was enough to make Magnus nervous. 🤦♂
Max Deutsch knows so little about chess that he didnt realize that Carlsen is in a complete different metaverse of chess. Its called the Dunning Kruger effect and basically means that you're overconfident in your knowledge simply because you have no idea how little you know
His approach was unbelievably wrong. From what I can tell from their fancy graphics which is just for show, is that he trained a neural network to predict the next move on the basis of millions of chess games - a very simplistic approach. Throwing only a neural network at chess is not enough to get a strong engine/bot. That's why it took years to build engines like AlphaZero, Stockfish.
On top of making an arguably weak computer "algorithm", the fact that he thought he could somehow make the same predictions as the computer is baffling. How exactly did he expect to memorize or replicate his algorithm. Even if the algorithm worked, how exactly did he plan on getting it into his head. Brute force? I've never been able to figure his plan no matter how much I watch the video.
In summary, he touched his nose by putting his hand back around his head.
I don't think he even thought about it honestly.
I'm sure he thought something ridiculous like "64 squares, about 64 parameters to memorize." This is what happens when you take one class on machine learning and think every problem is now trivial to solve.
@@frankjohnson123 yup, because Magnus was at least paid I find Deutsch's Dunning-Kruger based confidence in his "algorithm" and his ML and Software Development skills much more offensive :D. Also, as a German I am mildly miffed about his last name being Deutsch :D.
I could even say that the prediction maybe only embedded supervised learning scheme not a reinforcement learning one, haha
He wrote a 4 layer deep MLP with no preprocessing or anything. Basically what a first-year undergrad would write after 10 minutes of introduction to machine learning. Unsurprisingly, it didn't work.
This game was SO hilarious to watch - i can't believe you made a video about it. You are a Legend.
I was hoping Levi included the clip of him saying, "i was actually winning"
@@BenjaminWeeb that would have hit the spot.. it was the worst part of the original video..
@@BenjaminWeeb goshh lmaoooo
I breath for 30 days and challenge a whale in deep diving
The most impressive part of this is that this guy managed to play an OTB game with Carlsen in a month. That's the achievement, and with all this media coverage of it I doubt much of the money came out of this Extreme Learner's pocket. I respect him for this hustle to play Carlsen OTB
If this game had been in your Guess the ELO series, you probably would have assumed that Magnus was a max of around 1500 or so just because he was playing with a noob.
Holy shit, this game should really have been published for GTE
Magnus said he won't defend his title next year unless Max Deutsch is the challenger
Max didn't play bad for someone with 30 days of experience. He tried to follow the basic principles (develop, control the center), at least.
except he might have very well had prior experience.
The guy is a hack. Lied about "the algorithm". Seems likely he was lying about being new to chess as well.
@@railey2343 How did he lie about the algorithm?
@@matthewtelles4809 it would be completed if he hadn't lied
@@matthewtelles4809 there is no algorithm, obviously. You're telling me the guy programmed a chess engine all by himself? One that's specifically tailored to teach him how to beat GMs?
He's obviously full of shit. Clearly lying.
@@spark5010 I mean sure, but that isn't really an argument as maybe it failed afterwards.
What would have been good is to have Max play against Magnus, then a GM, then an IM, then an fm, then 1800-2000, then 1600-1800, then 1400-1600 and so on
I am pretty sure he still would have lost every single one
But Then he might have probably seen how much different chess is than any other games
I think you would be correct. In fact, I have to wonder if he would beat people below 1400 straight away. I'm inclined to say he would lose most if not all games, dropping in online elo rating to below 500 before ever gaining points.
@@PHeMoX that is harsh . I started learning chess about a month ago and not really learning just having fun and I'm 770 elo . I think he could beat a 1000
would love to see that, think the ELO rating works out to beating someone 200 points lower 9 times out of 10, imagine a line of 10 people after Magnus each of whom can beat the next one 90% of the time and on the very end is Max lmao
@@mastery4667 ah yes you're right
@@olivetree7430 1000 is actually pretty decent. He could beat a 1000 if he played like 20 game match, with two draws and 17 losses.
This is the problem when your parents tell you are “something special”
He's "special" alright. Unquestionably "Special" Olympics material without a doubt lmfao.
@@bloxorzwizard7931 Paralympic atheletes are amazingly talented. This guy is no chess athelete.
“I could tell you thought it was going to be fine.” Of course it would be fine. If Magnus was blindfolded, drunk, and getting tasered it still would have been completely fine
It's a little know fact that Magnus retired from chess after Max beat him in a rematch on April 1st, 2021.
A Rubiks cube really isn't difficult to solve. It has a far bigger reputation of difficulty than it deserves. You can learn to solve it in a day. Its just repeating the same puzzle again and again.
Before the algorithm was readily available online it gained a reputation of being very hard to the point where a lot of people in there 40's or 50's think you are some sort of magician for looking it up and commiting it to muscle memory
@@seanwalsh5989 I used to bring it to school and everyone assumed I was a genius.
Yeah, im as average as it comes and even i could learn how to do it in like 3 hours xD so anyone can.
the most agregious thing they do in the original wsj video is that they never show the actual game. i was waiting for a video covering this for quite a long time. thank you, levy!
“Keeping him nervous for 8 moves” bro you were playing a mainline position….
Bruh, assuming he actually is 1100, the odds of him beating Magnus "Magnum Dong" Carlsen is literally less than 1 in a billion. If he played a game against Magnus every day until he won, it would take him ~40,000 lifetimes to beat him, statistically speaking.
Seriously. Look it up on the Wismuth calculator. Normies have no Earthly idea how high the skill ceiling goes in chess.
There’s a part in the video Levy didn’t include where he explains that algorithm and he says
‘I will ultimately turn my brain into very slow and slightly less sophisticated chess computer’ while Stockfish calculates 120 million positions in one second.
He should try to become first violin in Vienna orchestra in 30 days next
The part where Magnus says why he agreed to the challenge and Levy responding with "so you got paid a lot" actually made me bust out laughing so hard it woke my wife and scared my cat lol
Sorry Levy, but the guy claims that he made Magnus nervous for the first 8 moves... Magnus was maybe only nervous that he wasn't delivering a fool's mate. Playing the main line of an opening, isn't making the chess world champion nervous. Magnus has seen that position and all of its variations thousands of times by this point in his life... How would one be nervous about something they are so intimately familiar with...
Diary entrance.
'Had White against Magnus. Magnus was extremely nervous until the move 1. g4!?'
I actually really like the idea of a normal person spending a month to get as good as possible at something like chess and documenting the process. If done right, it can provide great insights into what it takes to be a grandmaster, how people train for it, what the difference is between a good and a great player et cetera. In other words: it can demistify the vague concept of grandmaster to regular people.
However, this guy didn't do it right at all. He should have talked to chess pros, take lessons, copy their preparations and in general learn from the people who are already good at chess. Instead he thought he was smarter than everyone and could trick the system. So he worked alone, came up with some nonsense to tell the viewers and taught the public absolutely nothing about chess
I would totally watch a real documentary about taking on the chess world champion, but this is utter garbage.
Exactly. This Max guy is a charlatan and the definition of the Dunning-Kruger effect.
Michelle Khare cried when she reached 1000 after 8 months of training. This guy did nothing for 30 days and lost against the strongest player on the planet.
"Keeping him a little bit nervous for at least 8 moves was a minor victory.", that's like saying that you make Hussein Bolt worried because you're just 2 metres behind him after a second of race
I remember this story and, whilst I'm trying really hard to avoid being mean here, the only word I deem fitting to classify this entire ordeal is "ridiculous". Really no idea of what he was trying to accomplish
What really takes the cake is the fact that he admits that 3 hours before the match he hasn't even memorized anything yet...
Real final exam energy
So much for Max the 'extreme learner'. He had learned exactly nothing about chess in 30 days.
4:54 The level of sarcasm Levy put into this sentence is immeasurable.
hilarious😂😂😂
Everyone are criticizing only his plan to beat Magnus, but don't forget his plan-B was to build an algorithm equivalent (or better) than the ones built by big teams of dozens of experts, big budgets and years of work, in less than a month. This might have been even crazier than his plan-A.
I respectfully disagree with Gotham admiring max for trying. Max is a fraud where most of his stunts he tries todo in 30 days are fake or he already new how to do them. He most likely was originally going to play a Magnus Carlson chess app and just memorize the apps move. But when the wsj noticed him and offered him to actually play Magnus he ended up over his head. He made up some weird algorithm that wasn’t ready yet as a prepared excuse for why he was about todo so bad. He honestly deserves the hate.
Honestly he's lucky to have met and played the world champ. But it is funny that he just started the game with the final boss as the challenger
I can kinda see why this guy gets so many haters. Sorta insulting for somebody to just show up and think they can easily accomplish what others work their whole lives for.
Even this was insulting. There are so many who work their whole lives for on chess, and there is only one Magnus Carlsen.
@@u.v.s.5583 yea, exactly. That’s what I was trying to get at.
I played a game of hand and brain with some friends a week ago and I've exclusively played Caro against e5 for probably a year. We got into a Spanish and I was lost immediately.
Wait , you have friends.
@@ayushanjay4983 This isn't reddit
“Amateur Challenges World Chess Champion.”
Ight, let’s see how bad Levy lost.
To this day Max’s algorithm is still computing..
Love the content Levy! Could maybe be filled with a little IRL stuff, but still really enjoying everything. Much love from Germany
@Newcious stop spamming
Grüße zurück :D
Den Irl stuff hätte ich gerne, wäre lustig. Aber nicht auf dem Hauptkanal für mich. Das wäre mir zu casual im Vergleich zu den Analysen
@@percyreiling yeah perhaps on the Gotham City channel
Isn't this a little like me saying "I will learn to write a program that rivals Photoshop from scratch in 30 days"? (or build an auto engine, or fly an airliner, etc.) Anyone with the remotest understanding of a particular field will know it's not possible. But my problem isn't with Max, it's the excruciating production values of that WSJ video.
It's a shame Max didn't share his algorithm with Nepo
Ouch.
First five moves 👌
Magnus shaking in his boots at main line Ruy Lopez
My favorite part of that video is that the guy believed he could not just beat Magnus, but also create an algorithm that beats stockfish with brutte force
I heard Magnus won't play on the next WCC, unless it's against Max.
Found this purely entertaining. Levy struggling not to angry whilst asking us not to get angry 😂 all jokes aside love this channel and all the content and learnt alot.
Won't blame anyone for hating him after listening to the amount of nonsense coming out his mouth. He brought it upon himself.
P.S. keeping Magnus what? 13:45
I actually read Max's entire blog where day by day he detailed his process of designing this algorithm. He didn't memorize moves at all. Instead, the model he sought to train is one which evaluates any individual move: e.g., given the positions of all the pieces on the board, if i move my pawn from h4 to h5, is that a positive move? Such a model would have thousands of parameters (real numbers) that he would have to memorize and then add and multiply in his head in order to use. The mechanics of using such a model would then be to assess the board position, and for any move he wanted to consider playing, run it through this process, which by his own admission would takes hours or days depending on his speed of calculation. And again, that's just to give a rough answer to the question "is this move good?" and not even "what is the best move?". As it happened, he blamed the slow speed of model convergence as the limiting factor. But in reality even if he had the best chess engine in the world, and it only required 10,000 parameters (a very low estimate), he would still be hopelessly lost, both because in reality a) I doubt he could memorize that many digits or b) actually perform the mental math on that memorized matrix, and c) even if he could do those things, it would take him far longer than any over the board chess game ever lasts. It was a terrible plan from conception to execution.
you can read more here: medium.com/@maxdeutsch/m2m-day-342-is-this-even-humanly-possible-c6c5910d5ab7
or watch Max explain his algo here: ua-cam.com/video/ql3KtXtX-Mk/v-deo.html&ab_channel=MaxDeutsch
well, it is sounds the same thing to me
but manual
"Learn Chess like a computer." Ah yes, the Dewa Kipas school of chess pedagogy.