Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

What is Chess? What to think and for how long

alexcrompton.com

Post reply on HN
Print view View original

101–110 of 126 posts

Re: What is Chess? What to think and for how long

#101

Post by fasterik »

ly3xqhl8g9 wrote: post #89
ajkjk wrote: post #70

Earlier quoted context omitted.

The florid writing is an elitist fad; you can express the same idea in much simpler language without losing any meaning.

So what? Famously, Anna Karenina is about a woman who jumps in front of a train, why bother to read the other 863 pages. Anti-Oedipus [1] and especially A Thousand Plateaus [2] is about living in rarefied, futural spaces. Deleuze would be the first to be disgusted even at the faintest idea that his writing has any meaning whatsoever: you want meaning? go and live your life, don't read a book, especially not a philoso…

There's a word for that, it's called "obscurantism". Profound and true ideas can always be put in simple language. See Richard Feynman for one example. When someone uses a lot of complicated words and sentences to make a trivial point, that's a red flag for me.

Re: What is Chess? What to think and for how long

#102

Post by cryptonector »

_a_a_a_ wrote: post #71
sadhorse wrote: post #2

"The ability to play chess is the sign of a gentleman. The ability to play chess well is the sign of a wasted life." Paul Murphy.

The definition of a gentleman - "a man who can play the bagpipes. But doesn't" (unknown attribution but I got it from a scotsman)

A true Scotsman?

Re: What is Chess? What to think and for how long

#103

Post by NewEntryHN »

fferen wrote: post #79

Something I wonder: are computers better at chess because they can think faster? What if we give a team of top grandmasters 1 day per move, versus like 1 hour for Stockfish. Will they at least consistently draw? To my knowledge, top human vs computer matches these days only use piece odds, not time odds.

"Classical" chess engine (those based on alpha-beta search) are indeed better because they calculate faster, although so insanely faster that there is probably not enough time in a _lifetime_ for a human to win a single game against it.

The more recent chess engines (LeelaChess, and the latest versions of Stockfish) use neural nets to allow for more shallow calculations without decreasing game quality. In this case the engine is better because it's just "more intelligent" (and nobody understands how).

Re: What is Chess? What to think and for how long

#105

Post by humanizersequel »

ajkjk wrote: post #98
humanizersequel wrote: post #94

Earlier quoted context omitted.

People love to say this, but this is extremely dense writing riddled with technical language, written for people trained in reading it. You could perhaps eliminate some of the jargon, but in doing so you'd have to substitute a book's worth of less "florid" writing to get the same point across.

"florid" as a word means "excessively complicated", not just "complicated". I'd describe this writing as florid because it its fanciness and complexity is obnoxious and unnecessary. Of course it's my opinion, but yeah, I think it's just bad writing, written to sound pretentious on purpose, doing a shit job of communicating its point because it wants to be inaccessible. You could easily write the same thing more clear…

>You could easily write the same thing more clearly in no more space.

I certainly couldn't, and I have spent a reasonable amount of time studying Nomadology (the text from which this passage is taken). Why don't you give it a shot? Put your money where your mouth is, let's say just the second paragraph, as that's the more serious offender in terms of jargon. I sincerely don't think you can rewrite it to be ~12 lines, forgoing all the technical jargon used, and retain all of the semantic content.

Frankly, barring a demonstration of this task you could supposedly "easily" accomplish, it reads like you're just denigrating the writing because you can't understand it.

Re: What is Chess? What to think and for how long

#106

Post by humanizersequel »

fasterik wrote: post #101
ly3xqhl8g9 wrote: post #89

Earlier quoted context omitted.

So what? Famously, Anna Karenina is about a woman who jumps in front of a train, why bother to read the other 863 pages. Anti-Oedipus [1] and especially A Thousand Plateaus [2] is about living in rarefied, futural spaces. Deleuze would be the first to be disgusted even at the faintest idea that his writing has any meaning whatsoever: you want meaning? go and live your life, don't read a book, especially not a philoso…

There's a word for that, it's called "obscurantism". Profound and true ideas can always be put in simple language. See Richard Feynman for one example. When someone uses a lot of complicated words and sentences to make a trivial point, that's a red flag for me.

>Profound and true ideas can always be put in simple language.

Sure, you can reduce the passage above to "Chess is a game of specific, regimented roles, whereas Go is characterized an emergent, contingent understanding of purpose" — but you lose a lot of information compressing it like that, in the same way you lose a lot of information when you tell kids that black holes work like a bowling bowl on a trampoline. Sure, neither are incorrect at a phenomenological level, but it's such a reductive understanding that you can't accurately reason about it without reintroducing the "complicated words and sentences".

Re: What is Chess? What to think and for how long

#107

Post by sireat »

armitron wrote: post #7

Precise thought patterns, acronyms, X seconds per move... This seems like a terrible, robotic way to play chess, at least when it comes to having fun. I've played in excess of 100k games on lichess (mostly 3 minutes, some 5 minute and bullet too, no increments) with my ELO rating being ~2100 and I don't consciously do any of these (whether these or approximations of such still take place unconsciously is a different…

This is the actual answer and one that pretty much all strong players will agree on.

This "feel right" ability goes by different names unconscious mastery, intuition, GM-RAM, drunken chess, Blink, pattern recognition and so on.

How to develop it is another question.

Sadly I fear it is almost impossible to develop it past teenage years.

There is something magical that happens around age 13-15(Fischer said he just got good when he made this "jump", of course he was around 2300 before that). Suddenly everything just clicks.

I was a lowly A level player when it just clicked around the age of 15 and I was a master. Sadly/luckily I did not progress past FM. The "click" is a required but not sufficient condition for GM at an OTB play.

Sure for some it happens at a bit later stage for some at a bit younger stage. You have to do some work before that and to become a super GM you have to do work afterwards. (as Anand famously said you read a bunch of Informators afterwards)

As Tal said to young Kasparov - Garry you have to sac first and calculate afterwards.

Re: What is Chess? What to think and for how long

#108

Post by sireat »

thefourthchime wrote: post #27
armitron wrote: post #7

Precise thought patterns, acronyms, X seconds per move... This seems like a terrible, robotic way to play chess, at least when it comes to having fun. I've played in excess of 100k games on lichess (mostly 3 minutes, some 5 minute and bullet too, no increments) with my ELO rating being ~2100 and I don't consciously do any of these (whether these or approximations of such still take place unconsciously is a different…

When you look at the board you are calculating moves though yes? You don't just think about puppies and ice cream cones I assume.

90% of time you do not calculate, you "feel" the moves.

There is a saying that a difference between SuperGM, GM, IM, and a master is how many times a game they have to really spend time on a move that they are not sure of.

SuperGM might be unsure about 1-3 times a game, GM 3-5, IMs 5-7, and as a master I have about 7-10 moves.

Notice that depending on your time control, you will still want to take time to verify-calculate your "feel" moves.

Re: What is Chess? What to think and for how long

#109

Post by buzzy_hacker »

I'm a decent chess player. 1600 elo over the board, 1800 online. I think the conclusions reached give too much time to non-critical moves and not enough time to critical moves.

Let's take the 90 minute game, where it's suggested to use 88 seconds for opening moves, 165 seconds for non-critical moves, and 330 seconds for critical moves.

330 seconds (5 1/2 minutes) is too short for the critical position in a 90-minute game. I would be comfortable using about 15-20 minutes for the most critical move.

I think my disagreement stems from the author's estimate of "7 or 8 critical moves per game." That's too many in my experience. In a 40 move game (the average), that's 18.75%.

I'd say there are typically 3 or fewer critical moves per game. And there's no need to spend anywhere close to 88 seconds in the opening. When you're in theory, you can bust those out in a few seconds.

Re: What is Chess? What to think and for how long

#110

Post by tepitoperrito »

gjm11 wrote: post #96
amalcon wrote: post #84

Earlier quoted context omitted.

An easier way to do this would be to limit the hardware resources available to Stockfish. You can get the 24x time ratio just by giving Stockfish 1/24th of the compute cycles. That said, 24x is not even close to enough advantage there -- that probably removes less than one ply of depth. Stockfish can read far deeper than any human; you'd need a dramatically larger reduction to compensate for this. You could probably…

24x is more than 1 ply. Typical branching factor isn't so far from 24, and with alpha-beta the effective branching factor is more like sqrt(nominal branching factor), and I think modern engines are pretty good at pruning and extending the tree to make the scaling even better than that.

Is a "ply" a unit I'm not familiar with? Or does it just refer to a single layer. (Honestly not sure and curious)
Post reply on HN