|
Chess discussion continues here |
On November 17 2013 21:31 Grumbels wrote:Show nested quote +On November 17 2013 21:05 GolemMadness wrote:On November 17 2013 20:39 Grumbels wrote:Solvability relates to Starcraft about as much as it relates to baseball. Theoretically there is an optimal way to play baseball based on simulating the universe. Except that this is completely nonsensical and irrelevant because many implicit assumptions you have made fail to be met: for instance that optimal play is the best way to play if you want to win. Use this approach for Starcraft and it might discover that one race is "stronger" than another race because it is theoretically always winning, except that this has nothing to do with real strength because humans (and computers too) have limitations and won't play Starcraft as if it's a turn-based game where they can use their database that has infinity more values than all the atoms in the universe. It's also completely irrelevant to building an AI (unlike chess where it is useful) These concepts just can't be used, it's honestly a scientific crime because in science you have to be very rigorous in defining not only concepts but also all the assumptions that need to be met before the concept can be applied to a problem. If you use them haphazardly it leads to ridiculous results like bringing up solvability of Starcraft. Also, try beating this by playing about a 100 rounds without any outside help: http://www.nytimes.com/interactive/science/rock-paper-scissors.html?_r=0 Played 31 rounds, and am up 9-15-7. Wow. You should try to outsmart it, like with "I played x1..xn last n rounds so now the computer will expect me to play a but I will play b". I don't think you can win. :o
I was too bored to go for 100 rounds but i was up 15 - 8 - 12 by just randomly clicking something, after all it is possible to make the game 100% random.
|
On November 18 2013 08:04 gaymon wrote:Show nested quote +On November 17 2013 21:31 Grumbels wrote:On November 17 2013 21:05 GolemMadness wrote:On November 17 2013 20:39 Grumbels wrote:Solvability relates to Starcraft about as much as it relates to baseball. Theoretically there is an optimal way to play baseball based on simulating the universe. Except that this is completely nonsensical and irrelevant because many implicit assumptions you have made fail to be met: for instance that optimal play is the best way to play if you want to win. Use this approach for Starcraft and it might discover that one race is "stronger" than another race because it is theoretically always winning, except that this has nothing to do with real strength because humans (and computers too) have limitations and won't play Starcraft as if it's a turn-based game where they can use their database that has infinity more values than all the atoms in the universe. It's also completely irrelevant to building an AI (unlike chess where it is useful) These concepts just can't be used, it's honestly a scientific crime because in science you have to be very rigorous in defining not only concepts but also all the assumptions that need to be met before the concept can be applied to a problem. If you use them haphazardly it leads to ridiculous results like bringing up solvability of Starcraft. Also, try beating this by playing about a 100 rounds without any outside help: http://www.nytimes.com/interactive/science/rock-paper-scissors.html?_r=0 Played 31 rounds, and am up 9-15-7. Wow. You should try to outsmart it, like with "I played x1..xn last n rounds so now the computer will expect me to play a but I will play b". I don't think you can win. :o I was too bored to go for 100 rounds but i was up 15 - 8 - 12 by just randomly clicking something, after all it is possible to make the game 100% random.
i don' t think, in reality any human being can go random. random is not possible for us. it may seem so, but only a computer can truly pick random
|
On November 18 2013 10:26 pebble444 wrote:Show nested quote +On November 18 2013 08:04 gaymon wrote:On November 17 2013 21:31 Grumbels wrote:On November 17 2013 21:05 GolemMadness wrote:On November 17 2013 20:39 Grumbels wrote:Solvability relates to Starcraft about as much as it relates to baseball. Theoretically there is an optimal way to play baseball based on simulating the universe. Except that this is completely nonsensical and irrelevant because many implicit assumptions you have made fail to be met: for instance that optimal play is the best way to play if you want to win. Use this approach for Starcraft and it might discover that one race is "stronger" than another race because it is theoretically always winning, except that this has nothing to do with real strength because humans (and computers too) have limitations and won't play Starcraft as if it's a turn-based game where they can use their database that has infinity more values than all the atoms in the universe. It's also completely irrelevant to building an AI (unlike chess where it is useful) These concepts just can't be used, it's honestly a scientific crime because in science you have to be very rigorous in defining not only concepts but also all the assumptions that need to be met before the concept can be applied to a problem. If you use them haphazardly it leads to ridiculous results like bringing up solvability of Starcraft. Also, try beating this by playing about a 100 rounds without any outside help: http://www.nytimes.com/interactive/science/rock-paper-scissors.html?_r=0 Played 31 rounds, and am up 9-15-7. Wow. You should try to outsmart it, like with "I played x1..xn last n rounds so now the computer will expect me to play a but I will play b". I don't think you can win. :o I was too bored to go for 100 rounds but i was up 15 - 8 - 12 by just randomly clicking something, after all it is possible to make the game 100% random. i don' t think, in reality any human being can go random. random is not possible for us. it may seem so, but only a computer can truly pick random
Well computers can't completely pick random either. The most random generators are based on very hard to reproduce variables external to the system, however the system itself isn't randoming.
|
On November 18 2013 10:26 pebble444 wrote:Show nested quote +On November 18 2013 08:04 gaymon wrote:On November 17 2013 21:31 Grumbels wrote:On November 17 2013 21:05 GolemMadness wrote:On November 17 2013 20:39 Grumbels wrote:Solvability relates to Starcraft about as much as it relates to baseball. Theoretically there is an optimal way to play baseball based on simulating the universe. Except that this is completely nonsensical and irrelevant because many implicit assumptions you have made fail to be met: for instance that optimal play is the best way to play if you want to win. Use this approach for Starcraft and it might discover that one race is "stronger" than another race because it is theoretically always winning, except that this has nothing to do with real strength because humans (and computers too) have limitations and won't play Starcraft as if it's a turn-based game where they can use their database that has infinity more values than all the atoms in the universe. It's also completely irrelevant to building an AI (unlike chess where it is useful) These concepts just can't be used, it's honestly a scientific crime because in science you have to be very rigorous in defining not only concepts but also all the assumptions that need to be met before the concept can be applied to a problem. If you use them haphazardly it leads to ridiculous results like bringing up solvability of Starcraft. Also, try beating this by playing about a 100 rounds without any outside help: http://www.nytimes.com/interactive/science/rock-paper-scissors.html?_r=0 Played 31 rounds, and am up 9-15-7. Wow. You should try to outsmart it, like with "I played x1..xn last n rounds so now the computer will expect me to play a but I will play b". I don't think you can win. :o I was too bored to go for 100 rounds but i was up 15 - 8 - 12 by just randomly clicking something, after all it is possible to make the game 100% random. i don' t think, in reality any human being can go random. random is not possible for us. it may seem so, but only a computer can truly pick random
They are only random if you include some physical phenomenon to make your numbers (like radioactive decay), otherwise the computer will only give you numbers based on some deterministic algorithm, that means the numbers are not really random (for example if you use the current time, you will get the same "random" numbers every 24 hours)
|
I hope Anand will win tomorrow's match, being back to 1 point behind and 5 games left to play, it would be awesome.
And it would boost his confidence quite a bit for another win.
|
On November 18 2013 08:04 gaymon wrote:Show nested quote +On November 17 2013 21:31 Grumbels wrote:On November 17 2013 21:05 GolemMadness wrote:On November 17 2013 20:39 Grumbels wrote:Solvability relates to Starcraft about as much as it relates to baseball. Theoretically there is an optimal way to play baseball based on simulating the universe. Except that this is completely nonsensical and irrelevant because many implicit assumptions you have made fail to be met: for instance that optimal play is the best way to play if you want to win. Use this approach for Starcraft and it might discover that one race is "stronger" than another race because it is theoretically always winning, except that this has nothing to do with real strength because humans (and computers too) have limitations and won't play Starcraft as if it's a turn-based game where they can use their database that has infinity more values than all the atoms in the universe. It's also completely irrelevant to building an AI (unlike chess where it is useful) These concepts just can't be used, it's honestly a scientific crime because in science you have to be very rigorous in defining not only concepts but also all the assumptions that need to be met before the concept can be applied to a problem. If you use them haphazardly it leads to ridiculous results like bringing up solvability of Starcraft. Also, try beating this by playing about a 100 rounds without any outside help: http://www.nytimes.com/interactive/science/rock-paper-scissors.html?_r=0 Played 31 rounds, and am up 9-15-7. Wow. You should try to outsmart it, like with "I played x1..xn last n rounds so now the computer will expect me to play a but I will play b". I don't think you can win. :o I was too bored to go for 100 rounds but i was up 15 - 8 - 12 by just randomly clicking something, after all it is possible to make the game 100% random.
Went 40-27-32 on Veteran. Not exactily impossible.
|
Carlsen so far is playing like Fischer in his 71-72 run. Solid chess with no mistakes, outlasting his opponent, sneaking wins out of games most give as draws, all of a sudden he's up 2 wins. Let's see what'll happen this week.
Great match so far.
|
I think Anand is dead in the water at this point, but for the sake of hype I'd love to see him win some.
|
Anand's tactic doesnt seem to be working. He needs to change things up and try something different because I dont see him beating Carlson the way he is going atm.
|
On November 18 2013 11:05 lightman wrote:
Carlsen so far is playing like Fischer in his 71-72 run. Solid chess with no mistakes, outlasting his opponent, sneaking wins out of games most give as draws, all of a sudden he's up 2 wins. Let's see what'll happen this week.
Great match so far.
let's just hope he literally doesn't pull a bobby fischer
|
On November 18 2013 07:32 ch33psh33p wrote:Show nested quote +On November 18 2013 05:29 Yurie wrote:On November 18 2013 04:15 mihajovics wrote:
so the only question remains...
how many decades of Carlsen domination will we see? He already broke Kasparov's record of highest rating, now it's time to break his record of being #1 on the rating list. Every single generation breaks the rating record. There is inflation in the system. ... Are you serious.
Reasons for this inflation are unclear, but the inflation itself is real. If you look at the average ELO of the top 100, it shows a steady increase.
![[image loading]](http://en.chessbase.com/portals/4/files/news/2009/sonas02.gif)
In that sense, it is not surprising to have a new top ELO every generation.
What makes the Fisher, Kasparov or Carlsen numbers "survive" for some time is the gap between those players and the rest of the field.
- Fisher was #1 with 2760 ELO in 1972, 70points over #2 at the time (Spassky), he was the first over 2700.
- Kasparov reached 2851 in 2000, 80points over #2 at the time (Kramnik), he was the first over 2800
- Carlsen had a 75 points lead to #2 (Kramnik again) at 2870 in october. He might get 2900 (would probably need the rest of the top 10 to get to 2800+ to feed those points)
There are 2 ways to get the ELO record: Be the best by a small margin when the curve leads you to the mark, or crush all opposition by a large margin and get there ahead of time. Those 3 have done it by getting far ahead, but even if no new genius comes after Carlsen to crush the field, his current rating could still be average for top 10 in 20 years.
(Fisher's 2760 would barely get him in today's top 10)
And of course, an absolute ELO value is by no means a "level" estimate (since it's only a comparison to current players), so the record itself is meaningless.
That said, go Anand, win one before it's over !
|
So... when I look at their past history, Anand would seem to have an advantage but everyone thinks that Carlsen will crush him... and in the light of the past two games it also seems like he's far superior as a player... whats changed?
|
Most of the losses were from when Anand was *arguably* at or around his peak, whereas Carlsen was still young? What is their most recent match history, ie their last 5-10 games?
|
Now we come to personal opinion about the rating.
I think there is inflation because chess players stay active after their peak (I would as well). So they peak at 2700, lose 70% of their games, then leave at 2500 or lower rating. They then fed 200 points into inflation.
Second part is new players. They feed around themselves, then often quit before stabilising, thus increasing total point pool while not increasing player pool as much. This is probably a factor that is decreasing since many people play online instead of in the normal rating system now a days.
edit, it could also be player skill is increasing this quickly and the system doesn't have inflation.
|
On November 18 2013 12:37 meeple wrote:
So... when I look at their past history, Anand would seem to have an advantage but everyone thinks that Carlsen will crush him... and in the light of the past two games it also seems like he's far superior as a player... whats changed?
Not sure but I think their first 4 games were wins for Anand when Carlsen was younger and somewhat weaker than he is right now.
The record since then actually favors Carlsen if I am not mistaken.
|
I can't remember their match history, but I think it was 3 years since Anand won over Carlsen. Adding the two losses in a row for Anand he must have had a tough time after game 6. Let's not count him out yet though! I think we might have a really interesting game today 
|
On November 18 2013 15:56 Arevall wrote:I can't remember their match history, but I think it was 3 years since Anand won over Carlsen. Adding the two losses in a row for Anand he must have had a tough time after game 6. Let's not count him out yet though! I think we might have a really interesting game today
It's do or die for Anand today. If he draws he has 2 whites left needing 2 wins. If Carlsen offers a sharp fight he might as well take it even if there's a chance it's Carlsen's preparation. Carlsen OTOH, seems to be almost stubborn in playing for a win even if it makes little sense in the tournament or match situation. It almost cost him his match against Anand, when he lost in the final round of the Candidate Tournament.
|
On November 18 2013 15:45 Yurie wrote:
Now we come to personal opinion about the rating.
I think there is inflation because chess players stay active after their peak (I would as well). So they peak at 2700, lose 70% of their games, then leave at 2500 or lower rating. They then fed 200 points into inflation.
Second part is new players. They feed around themselves, then often quit before stabilising, thus increasing total point pool while not increasing player pool as much. This is probably a factor that is decreasing since many people play online instead of in the normal rating system now a days.
edit, it could also be player skill is increasing this quickly and the system doesn't have inflation.
It's not really opinion. All Elo systems suffer from inflation from the influx of new players. FIDE attempts to avoid this by using a provisional-ish system where they will not give you a rating until you have played a certain number of games, but even so, a new player is much more likely to receive a higher than average rating, because he is more likely to be playing against active players who also hold above average ratings. Because of this, the average rating of all players increases over time, and this inevitably feeds all the way up into the top levels.
Your other point is also accurate; the skill gap for most games widens over time, and chess is no exception. As a game like chess or SC2 gets more and more figured out, the top tiers of players separate themselves more distinctly from the pack, and their lead grows accordingly. This is not true inflation per se, but it is definitely a factor in the ratings of top players.
In regards to Carlsen, Oshuy's great post already put everything into context. He is benefiting from the Elo inflation, sure, but so is everyone else, and none of his contemporaries have even come close to reaching the same heights. It's unfair to compare Carlsen to Kasparov, but it's entirely fair to compare him to his peers and say "Okay, this guy is exhibiting Kasparov-like dominance."
|
On November 18 2013 16:11 hypercube wrote:Show nested quote +On November 18 2013 15:56 Arevall wrote:I can't remember their match history, but I think it was 3 years since Anand won over Carlsen. Adding the two losses in a row for Anand he must have had a tough time after game 6. Let's not count him out yet though! I think we might have a really interesting game today It's do or die for Anand today. If he draws he has 2 whites left needing 2 wins. If Carlsen offers a sharp fight he might as well take it even if there's a chance it's Carlsen's preparation. Carlsen OTOH, seems to be almost stubborn in playing for a win even if it makes little sense in the tournament or match situation. It almost cost him his match against Anand, when he lost in the final round of the Candidate Tournament.
Well, if had a drawn the final round of the Candidates and Kramnik had won his match, it's Kramnik who would have had more points... So it's understandable that he tried to get a win there to make sure he'd qualify!
|
On November 18 2013 12:37 meeple wrote:
So... when I look at their past history, Anand would seem to have an advantage but everyone thinks that Carlsen will crush him... and in the light of the past two games it also seems like he's far superior as a player... whats changed?
Anand hasnt won a single set against Carlsen in the last three years. They have all been draws or wins for Carlsen. Anands wins stem from periods when Magnus was young and had lower rating
|
|
|
|
|
|
EPT
