Open AI Championship

Post #24

Finals: before start

Project

Sandbox

Round 1

Round 2

Finals

Recent comments

12 december 13:31: alex3d wrote comment on post Some more good news ;)
12 december 12:07: newpuk wrote comment on post Finals
12 december 09:50: leloy wrote comment on post Finals
11 december 23:48: VexfxjC wrote comment on post Some more good news ;)
11 december 23:38: alkozel wrote comment on post Some more good news ;)
11 december 18:26: Jeners wrote comment on post Some more good news ;)
11 december 16:59: VexfxjC wrote comment on post Some more good news ;)
11 december 16:57: VexfxjC wrote comment on post Some more good news ;)
11 december 16:08: OREZ wrote comment on post Some more good news ;)
11 december 11:58: Eugene713 wrote comment on post Some more good news ;)

VK Group

Telegram chat

Some good news

Some good news for the last week.

The performance of Ruby and Python 2 and 3 clients was improved. While Ruby and Python 2 received a very small speed increase, the boost of Python 3 is more significant. In turn, fans of Python 2 had the opportunity to run their strategies in an alternate way, namely PyPy 2 5.8.0+. This can improve the execution speed of some strategies. However the NumPy and SciPy library are not available. Choose wisely. Unfortunately, PyPy 3 does not have official support in Windows, except for the old slow version.

We noticed that after achieving significant advantage in score points, many strategies try not to force events and simply wait for the end of game time. In this regard, the rules were updated. Now, since the middle of the game, a strategy with fewer units regularly gains a small amount of score points. So to say, the bonus for survival. We hope that now the games will become more dynamic and interesting.

Updated:

  • replaced weak strategy award with vehicle maintenance payment.
  • December 3: reduced slightly vehicle maintenance cost, also fixed bug with vehicle production.


Adler

Adler

Dzyarzhynsk

OK, then we start the battle to end of the game

Dec 2, 2017 5:57:15 PM
selya_

selya_

Гимназия №1

The game at the end of which I have and opponent same number of points, and adds points to the rating in the sandbox, or not?

Dec 2, 2017 6:11:59 PM
Adler

Adler

Dzyarzhynsk

>Every $1000$ tick starting at $10000$ and not including the last tick, the strategy with less amount of technology will get $1$ point for every $10$ units of difference with the opponent. >strategy with less equipment gets… if you merge to 10000 frame all equipment except for one airplane, ideally you can get no less than 499 points for 1000 shots. not bad, so you can win.

Dec 2, 2017 6:12:11 PM
Adler

Adler

Dzyarzhynsk

Adler

Adler

Dzyarzhynsk

Every $1000$ tick starting at $10000$ and not including the last tick, the strategy with less amount of technology will get $1$ point for every $10$ units of difference with the opponent. >strategy with less equipment gets… if you merge to 10000 frame all equipment except for one airplane, ideally you can get no less than 499 points for 1000 shots. not bad, so you can win.

> Every $1000$ tick starting at $10000$ and not including the last tick, the strategy with less amount of technology will get $1$ point for every $10$ units of difference with the opponent.

.

> strategy with less equipment gets…

if you merge to 10000 frame all equipment except for one airplane, ideally you can get no less than 499 points for 1000 shots. not bad, so you can win.

upd: formatted

Dec 2, 2017 6:14:04 PM
selya_

selya_

Гимназия №1

Adler

Adler

Dzyarzhynsk

Every $1000$ tick starting at $10000$ and not including the last tick, the strategy with less amount of technology will get $1$ point for every $10$ units of difference with the opponent. >strategy with less equipment gets… if you merge to 10000 frame all equipment except for one airplane, ideally you can get no less than 499 points for 1000 shots. not bad, so you can win.

But this is absurd. Early got the advantage and survived, now merged units and survive. One airplane to escape much easier than a herd of equipment.

Dec 2, 2017 6:15:29 PM
selya_

selya_

Гимназия №1

Let points accrue, if your and the enemy’s technique without contact, for example, 1000 ticks. And up to a certain threshold (below the one with the least units could not score more points than the one who has more)

Dec 2, 2017 6:18:51 PM
spacistor

spacistor

AnswerPRO

The organizers got tired of the game themes and RTS, we grow up and instead of war games move on to the functional analysis: let’s maximize function with unknown behaviour?

Dec 2, 2017 6:21:18 PM
Kristopher

Kristopher

АГ СПбГУ

Good news? Pfft…

Dec 2, 2017 6:29:01 PM
oreshnik

oreshnik

Альфастрахование

Sorry, but the change is delusional. First, it does not help to cope with the stated problem. Change is not something that stimulates you more attack in the end how much is not stimulated to attack in the first part of the game. Second, it is unclear why you think it is possible to support the weaker strategy, while conditions should be equal for all. And finally, just to say “bonus for surviving” - if one remains 400 units as the opponent - 100, it seems to me that the first “survived” better, maybe better then it is still a bonus to give? :D

Dec 2, 2017 6:34:29 PM
dbf

dbf

A

It is not clear what problem trying to solve the organizers. It’s not just “sitting”, and the working way of dealing with the sandwich, which was ranked top in the first round. They are very difficult to take away on the part of groups of units, but in the second part of the opportunity to circumvent on points with control plants. Throw a homogeneous group of forces within the sandwich still will not make sense - where losses can be 10 times more points. Well, we’ll just have less to build to not fined, as it will add entertainment?

Dec 2, 2017 6:41:12 PM
AlexKol

AlexKol

BB

1) the Tactics to be allocated in addition to all of the aircraft. Persecuted to the enemy, poured in Maxim. Get the freebie. 2) Moisten their yaderku, and hide the plane 1… the opponent has 0 points we are in chocolate

If you want to get rid of the sandwiches, remove yaderku.

Dec 2, 2017 6:58:28 PM
selya_

selya_

Гимназия №1

dbf

dbf

A

It is not clear what problem trying to solve the organizers. It’s not just “sitting”, and the working way of dealing with the sandwich, which was ranked top in the first round. They are very difficult to take away on the part of groups of units, but in the second part of the opportunity to circumvent on points with control plants. Throw a homogeneous group of forces within the sandwich still will not make sense - where losses can be 10 times more points. Well, we’ll just have less to build to not fined, as it will add entertainment?

But by the way. Structure made sandwiches very inefficient on the map where these buildings are located far from each other and in large quantity. That allows you to not only gain a lot of points strategies with small groups, but to create a bunch of equipment. So maybe this change and balance both types of strategies on maps with buildings. But again, we need atleast limit - not the difference in the number of units, and the difference remaining from the original 500. Ie if I have 100 and I’ve built another 200, and another chela 400 and it was built 600, the 700 is the difference, and 300. Which is still nonsense, but at least nonsense)

Dec 2, 2017 6:59:16 PM
selya_

selya_

Гимназия №1

AlexKol

AlexKol

BB

1) the Tactics to be allocated in addition to all of the aircraft. Persecuted to the enemy, poured in Maxim. Get the freebie. 2) Moisten their yaderku, and hide the plane 1… the opponent has 0 points we are in chocolate If you want to get rid of the sandwiches, remove yaderku.

Is this a joke? Sandwich strong concentrated firepower. If you remove yaderku, there is little that will deprive ourselves of how to fight the sandwiches, and all your attacks will end in tears.

Dec 2, 2017 7:00:58 PM
AlexKol

AlexKol

BB

selya_

selya_

Гимназия №1

Is this a joke? Sandwich strong concentrated firepower. If you remove yaderku, there is little that will deprive ourselves of how to fight the sandwiches, and all your attacks will end in tears.

Actually, no. Sandwich this build, and he’s strong while he holds it. If to circle around him and shoot the weak spots on quietly you can win. Yaderka on it nothing gives, he filitsa quickly. But when he gets yaderku is the problem.

Dec 2, 2017 7:03:30 PM
dbf

dbf

A

> If to circle around him and shoot the weak spots on quietly you can win And there are examples where dismantled as a normal sandwich? I did a little test with her, and failed to achieve a stable analysis. Or need all sorts of complicated moves, like the attack immediately after a nuclear attack.

Dec 2, 2017 7:06:05 PM
oreshnik

oreshnik

Альфастрахование

AlexKol

AlexKol

BB

Actually, no. Sandwich this build, and he’s strong while he holds it. If to circle around him and shoot the weak spots on quietly you can win. Yaderka on it nothing gives, he filitsa quickly. But when he gets yaderku is the problem.

Really good, exactly folded sandwich no known weaknesses.

Dec 2, 2017 7:06:55 PM
AlexKol

AlexKol

BB

oreshnik

oreshnik

Альфастрахование

Really good, exactly folded sandwich no known weaknesses.

Corners are a weak point

Dec 2, 2017 7:09:10 PM
JIo6cTep

JIo6cTep

Samara

“Every 1000 ticks, starting from 10,000 and not including the last tick, the strategy with less equipment receives 1 point for every 10 units of difference with the opponent.” Something not clear - the difference of units modulo is taken? Or is it if I have less units, but more points, the opponent gets nothing?

Dec 2, 2017 7:11:01 PM
selya_

selya_

Гимназия №1

AlexKol

AlexKol

BB

Actually, no. Sandwich this build, and he’s strong while he holds it. If to circle around him and shoot the weak spots on quietly you can win. Yaderka on it nothing gives, he filitsa quickly. But when he gets yaderku is the problem.

It is quite difficult to write a nibble weaknesses slowly. I’m in the code, for example, take a group of allied units, enemy group, in which my team is moving and count the number of shots to kill the enemy me and my enemy, on the basis of these calculations to make decisions. To do better I’m just too lazy, because it is difficult and is already a kind of mini-project in which you put the search path, analyzer weaknesses and so on. A very small number of people here will do it. This means that to win a sandwich much more difficult than winning sandwich. And the same yaderka very strong against sandwich. Compare games where the sandwich goes on a sandwich. If one of them will not depart from yaderku, you lose with a score almost two times less) I mean that it is easier not to look for weaknesses, and to create their yaderku. And attack until he sahilinde.

Dec 2, 2017 7:11:34 PM
AlexKol

AlexKol

BB

selya_

selya_

Гимназия №1

It is quite difficult to write a nibble weaknesses slowly. I’m in the code, for example, take a group of allied units, enemy group, in which my team is moving and count the number of shots to kill the enemy me and my enemy, on the basis of these calculations to make decisions. To do better I’m just too lazy, because it is difficult and is already a kind of mini-project in which you put the search path, analyzer weaknesses and so on. A very small number of people here will do it. This means that to win a sandwich much more difficult than winning sandwich. And the same yaderka very strong against sandwich. Compare games where the sandwich goes on a sandwich. If one of them will not depart from yaderku, you lose with a score almost two times less) I mean that it is easier not to look for weaknesses, and to create their yaderku. And attack until he sahilinde.

Well, I have a very neat strategy, I’m struggling with this. But it just will not climb there where you should not

Dec 2, 2017 7:13:26 PM
AlexKol

Well, I have a very neat strategy, I’m struggling with this. But it just will not climb there where you should not

And now, even if I totally win I can not win, like chasing a crowd of disparate units is just nonsense

Dec 2, 2017 7:15:00 PM
selya_

selya_

Гимназия №1

AlexKol

AlexKol

BB

Well, I have a very neat strategy, I’m struggling with this. But it just will not climb there where you should not

I noticed. And yet, so to disassemble the sandwich to be quite difficult. At least not without yaderku or for a given number of ticks (only 20,000). By the way, if not a secret, you use something like a simplified steering behaviours, or something more static like a search path?

Dec 2, 2017 7:16:02 PM
selya_

selya_

Гимназия №1

In General, for this update: making a sandwich, separated the two airplane. One sent to bomb in advance, the other sent home (to escape). The enemy will not remain a lot of units. It will build new factories - make yourself is worse. No one using the group, not chasing them at top speed, not to destroy the structure, and hiding the airplane, respectively (in theory) no one will catch (at least 20000 ticks). Easy wines. Even if the airplane caught, you will have enough points (499+ for every 1000 ticks, i.e., until 4491+ points). Easy win. From one extreme to another, as they say

Dec 2, 2017 7:25:33 PM
selya_

selya_

Гимназия №1

selya_

selya_

Гимназия №1

In General, for this update: making a sandwich, separated the two airplane. One sent to bomb in advance, the other sent home (to escape). The enemy will not remain a lot of units. It will build new factories - make yourself is worse. No one using the group, not chasing them at top speed, not to destroy the structure, and hiding the airplane, respectively (in theory) no one will catch (at least 20000 ticks). Easy wines. Even if the airplane caught, you will have enough points (499+ for every 1000 ticks, i.e., until 4491+ points). Easy win. From one extreme to another, as they say

Sorry. 49, not 499. Was wrong

Dec 2, 2017 7:26:47 PM
AlexKol

AlexKol

BB

selya_

selya_

Гимназия №1

I noticed. And yet, so to disassemble the sandwich to be quite difficult. At least not without yaderku or for a given number of ticks (only 20,000). By the way, if not a secret, you use something like a simplified steering behaviours, or something more static like a search path?

Evaluation at each point, and finding the best way to best point.

Dec 2, 2017 7:35:22 PM
dbf

dbf

A

Here’s an example: http://russianaicup.ru/game/view/161419 under the new rules. The enemy in the area of the 16000-th step is a bit aircraft. He deftly leads them on a map, every 1000 ticks me fined on 80 points. An approximate equal number of points, each such fine displays of the opponent forward. I, of course, the sucker that couldn’t kill him, but where is this game an example of “a strategy of trying not to force things and just wait for the end of playing time”? :) Such penalties at a large advantage in units where required smooth allow all the advantage (and that they take the strategy “help” forward is quite strange).

Dec 2, 2017 7:39:19 PM
selya_

selya_

Гимназия №1

dbf

dbf

A

Here’s an example: http://russianaicup.ru/game/view/161419 under the new rules. The enemy in the area of the 16000-th step is a bit aircraft. He deftly leads them on a map, every 1000 ticks me fined on 80 points. An approximate equal number of points, each such fine displays of the opponent forward. I, of course, the sucker that couldn’t kill him, but where is this game an example of “a strategy of trying not to force things and just wait for the end of playing time”? :) Such penalties at a large advantage in units where required smooth allow all the advantage (and that they take the strategy “help” forward is quite strange).

The lack of units scores. Damn, this is really absurd. It’s like in a war in Afghanistan to give the glasses a little boy, deftly running from the troops, a bunch of points because he was able to escape. This because its STRATEGIC position it’s not getting better, right?

Dec 2, 2017 7:48:06 PM
SladeThe

SladeThe

Russian AI Cup Team

dbf

dbf

A

Here’s an example: http://russianaicup.ru/game/view/161419 under the new rules. The enemy in the area of the 16000-th step is a bit aircraft. He deftly leads them on a map, every 1000 ticks me fined on 80 points. An approximate equal number of points, each such fine displays of the opponent forward. I, of course, the sucker that couldn’t kill him, but where is this game an example of “a strategy of trying not to force things and just wait for the end of playing time”? :) Such penalties at a large advantage in units where required smooth allow all the advantage (and that they take the strategy “help” forward is quite strange).

“deftly leads on the map”

Cleverly is in place.

“many of the strategies try not to force things and just wait for the end game time”

This game is a striking example of this. Instead of finish off the enemy group of units randomly move around the map. And not even all the buildings were occupied, which has already led to the victory.

I haven’t seen a single example of a game that would not be lost solely because of the weakness of the strategy.

Dec 2, 2017 7:53:37 PM
Adler

Adler

Dzyarzhynsk

Now profitable to get rid of their BREM s to 10000 frame, and try not to kill enemy units up to 19000 frame.

Dec 2, 2017 7:54:06 PM
selya_

selya_

Гимназия №1

SladeThe

SladeThe

Russian AI Cup Team

“deftly leads on the map” Cleverly is in place. “many of the strategies try not to force things and just wait for the end game time” This game is a striking example of this. Instead of finish off the enemy group of units randomly move around the map. And not even all the buildings were occupied, which has already led to the victory. …

You want to say that the strategy is lost because of its weaknesses, however, weaker strategy proved to be more intelligent because they just “merged and deftly stood still”? Something any contradiction

Dec 2, 2017 7:57:50 PM
SladeThe

SladeThe

Russian AI Cup Team

selya_

selya_

Гимназия №1

The lack of units scores. Damn, this is really absurd. It’s like in a war in Afghanistan to give the glasses a little boy, deftly running from the troops, a bunch of points because he was able to escape. This because its STRATEGIC position it’s not getting better, right?

Not true. If the troops in Afghanistan won’t be able to catch him, they will sooner or later have to leave, but he will remain the winner. In history there are examples.

Dec 2, 2017 7:59:22 PM
SladeThe

SladeThe

Russian AI Cup Team

Adler

Adler

Dzyarzhynsk

Now profitable to get rid of their BREM s to 10000 frame, and try not to kill enemy units up to 19000 frame.

Or quickly finish off the enemy, get 1000 points and win. Why of all the possible ways you suggest is the most stupid? It seems that it’s just the usual whining because of the reluctance to Refine the strategy.

Dec 2, 2017 8:01:41 PM
AlexKol

AlexKol

BB

SladeThe

SladeThe

Russian AI Cup Team

“deftly leads on the map” Cleverly is in place. “many of the strategies try not to force things and just wait for the end game time” This game is a striking example of this. Instead of finish off the enemy group of units randomly move around the map. And not even all the buildings were occupied, which has already led to the victory. …

http://russianaicup.ru/game/view/160992 here is an example. for 10K I totally won. But lost on points in the end, because chase small groups of units larger direct sense.

Dec 2, 2017 8:02:18 PM
selya_

selya_

Гимназия №1

SladeThe

SladeThe

Russian AI Cup Team

Not true. If the troops in Afghanistan won’t be able to catch him, they will sooner or later have to leave, but he will remain the winner. In history there are examples.

That was an example, distant from the model on which we test. Immediately two troops went on neutral territory. One of the troops almost merged, but cleverly hid his remaining units. Time went, both troops go home. What will return one strategy? With two or three units and shouts of “HURRAH, YOU’re HEROES!”. And what about the second? With cries of “UGH, FUCKERS” (in your logic). Although initially the conditions were the same, and the effectiveness of the second strategy was higher

Dec 2, 2017 8:02:28 PM
dbf

dbf

A

SladeThe

SladeThe

Russian AI Cup Team

“deftly leads on the map” Cleverly is in place. “many of the strategies try not to force things and just wait for the end game time” This game is a striking example of this. Instead of finish off the enemy group of units randomly move around the map. And not even all the buildings were occupied, which has already led to the victory. …

This game is a striking example of this. Instead of finish off the enemy group of units randomly move around the map. And not even all the buildings were occupied, which has already led to the victory.

I.e. it is normal that “help” displays the backward bonus forward dozens of points?

Dec 2, 2017 8:02:51 PM
SladeThe

SladeThe

Russian AI Cup Team

selya_

selya_

Гимназия №1

That was an example, distant from the model on which we test. Immediately two troops went on neutral territory. One of the troops almost merged, but cleverly hid his remaining units. Time went, both troops go home. What will return one strategy? With two or three units and shouts of “HURRAH, YOU’re HEROES!”. And what about the second? With cries of “UGH, FUCKERS” (in your logic). Although initially the conditions were the same, and the effectiveness of the second strategy was higher

The “effectiveness” of the strategy is evaluated according to the rules of the game, not according to some ideas about global justice.

If that example had to do with the game, you shouldn’t have to bring him in.

Dec 2, 2017 8:05:38 PM
SladeThe

SladeThe

Russian AI Cup Team

dbf

dbf

A

This game is a striking example of this. Instead of finish off the enemy group of units randomly move around the map. And not even all the buildings were occupied, which has already led to the victory. I.e. it is normal that “help” displays the backward bonus forward dozens of points?

It is normal that a strategy that manages thousands of units, can’t finish a dozen?

Dec 2, 2017 8:06:35 PM
AlexKol

AlexKol

BB

selya_

selya_

Гимназия №1

That was an example, distant from the model on which we test. Immediately two troops went on neutral territory. One of the troops almost merged, but cleverly hid his remaining units. Time went, both troops go home. What will return one strategy? With two or three units and shouts of “HURRAH, YOU’re HEROES!”. And what about the second? With cries of “UGH, FUCKERS” (in your logic). Although initially the conditions were the same, and the effectiveness of the second strategy was higher

It would make sense if there were different game modes. The type of base defense, some attack other attack, and those who are protected would heroically run away.

Dec 2, 2017 8:06:58 PM
dbf

dbf

A

SladeThe

SladeThe

Russian AI Cup Team

It is normal that a strategy that manages thousands of units, can’t finish a dozen?

We are in the starting conditions the same. I scored more points and built more units. Why this is the reason that my opponent appears artificially forward in the account?

Dec 2, 2017 8:07:50 PM
selya_

selya_

Гимназия №1

SladeThe

SladeThe

Russian AI Cup Team

The “effectiveness” of the strategy is evaluated according to the rules of the game, not according to some ideas about global justice. If that example had to do with the game, you shouldn’t have to bring him in.

I understand that the rules of the game. But this rule, which was “slightly bustanut left behind” turned the game, and now those strategies that won will play weaker in all parameters. If you and it means nothing, then argue I will no longer

Dec 2, 2017 8:08:09 PM
AlexKol

AlexKol

BB

And when will the fog of war, a small group of aircraft, how to look?

Dec 2, 2017 8:09:20 PM
AlexKol

AlexKol

BB

Maybe the 2018 world Cup the same rules will introduce, after the first half, the side that loses record of 1 goal every 10 minutes. The stronger team on their toes and scored as many goals as possible….

Dec 2, 2017 8:19:18 PM
Kunik

Kunik

Heatherglade

SladeThe

SladeThe

Russian AI Cup Team

It is normal that a strategy that manages thousands of units, can’t finish a dozen?

According to this logic, 45 it was necessary to win a complete genocide, is this normal? :)

Dec 2, 2017 8:21:27 PM
dbf

dbf

A

AlexKol

AlexKol

BB

Maybe the 2018 world Cup the same rules will introduce, after the first half, the side that loses record of 1 goal every 10 minutes. The stronger team on their toes and scored as many goals as possible….

Better not even one goal, and by 2 to trailing by 1 it was possible to come out ahead.

Dec 2, 2017 8:22:20 PM
azt-yur

azt-yur

ОГУ

SladeThe

SladeThe

Russian AI Cup Team

It is normal that a strategy that manages thousands of units, can’t finish a dozen?

And if you do not these rules originally came up with? And is week the beta test is not over to rules to fundamentally change? Then it is officially announced before each round a week of beta, or conduct a survey on the changes (like last year)

Dec 2, 2017 8:28:36 PM
Kunik

Kunik

Heatherglade

Congratulations to the Hungry games! And let luck will always be with you!

Dec 2, 2017 8:37:38 PM
mixei4

mixei4

КГУ

azt-yur

azt-yur

ОГУ

And if you do not these rules originally came up with? And is week the beta test is not over to rules to fundamentally change? Then it is officially announced before each round a week of beta, or conduct a survey on the changes (like last year)

So officially it was the second week of beta test before the second round.

Dec 2, 2017 8:38:18 PM
Kreol

Kreol

RetailNext

You have there in the name of the news error “A bit of good news over the past week” -> “Not a lot of good news over the last week”

Dec 2, 2017 8:41:35 PM
jurifoll

jurifoll

Moscow

Perhaps, the organizers of this innovation wanted to weaken the sandwiches. They are still widely distributed and often win, even if they fail to catch up with all the enemy groups. But the rule is very strange and provokes perversion. Would be better if the limit of action of the increased if necessary to shift the balance in favor of distributed strategies.

Dec 2, 2017 8:42:12 PM
azt-yur

azt-yur

ОГУ

mixei4

mixei4

КГУ

So officially it was the second week of beta test before the second round.

Yes indeed, that is not paying attention. Then it turns out everything is legal ))), the government has the right to be indignant it is useless, postreverse people.

Dec 2, 2017 8:46:18 PM
selya_

selya_

Гимназия №1

jurifoll

jurifoll

Moscow

Perhaps, the organizers of this innovation wanted to weaken the sandwiches. They are still widely distributed and often win, even if they fail to catch up with all the enemy groups. But the rule is very strange and provokes perversion. Would be better if the limit of action of the increased if necessary to shift the balance in favor of distributed strategies.

The decision is good for balancing. But initially without buildings sandwich was very strong. Right, let’s do the second round of the building to small mobile groups make it easier to capture. And then there’s the fog of war let’s go to strategy is again divided into two types. Those who puts fighters to vision, and those that are looking for the enemy at random. I understand to add conditions that complicate the gameplay, and you have to add strategy and make it smarter. But I don’t understand the changes that revolutionize the essence of the game.

Dec 2, 2017 8:50:12 PM
JIo6cTep

JIo6cTep

Samara

1) Why to draw ridiculous Parallels with the real military conflicts? This competition has no relation to real life. You did not ask why in chess, the Bishop can take the rook. Better to just use common sense. 2) My common sense says that any points must be earned to some useful actions. So innovation is very strange. And in General artificial encouragement lagging in any competition it’s weird.

Dec 2, 2017 8:52:43 PM
ilt

ilt

Krasnogorsk

azt-yur

azt-yur

ОГУ

Yes indeed, that is not paying attention. Then it turns out everything is legal ))), the government has the right to be indignant it is useless, postreverse people.

Il just go away.

Dec 2, 2017 9:02:42 PM
lnuynxa
SladeThe

SladeThe

Russian AI Cup Team

It is normal that a strategy that manages thousands of units, can’t finish a dozen?

This competition gives you the opportunity to test your programming skills by creating artificial intelligence (strategy), managing a large number of units (equipment) in a special game world.

According to the description of your tournament strategy with 10 units failed in the management of a large number of units, and therefore the skills of its Creator failed validation?

Dec 2, 2017 9:03:15 PM
bypeso

bypeso

БНТУ

This competition gives you the opportunity to test your programming skills by creating artificial intelligence (strategy), managing a large number of units (equipment) in a special game world. According to the description of your tournament strategy with 10 units failed in the management of a large number of units, and therefore the skills of its Creator failed validation?

http://russianaicup.ru/game/view/161439

But the strategy collapsed on 6K tick all right. her another and give a head start. ODDS THE COLLAPSED STRATEGY

Dec 2, 2017 9:06:05 PM
GreenTea

GreenTea

Organization::ISDDesign

http://russianaicup.ru/game/view/161825 normal?) And I specifically didn’t kill him and it happened..

Dec 2, 2017 9:12:23 PM
azt-yur

azt-yur

ОГУ

Then it is exclude from the calculation units destroyed himself. And in some cases it may be advantageous to destroy and get points for it, especially in games without buildings.

Dec 2, 2017 9:15:43 PM
azt-yur

azt-yur

ОГУ

GreenTea

GreenTea

Organization::ISDDesign

http://russianaicup.ru/game/view/161825 normal?) And I specifically didn’t kill him and it happened..

It turns out if you win, you make the technique no longer profitable. By the way and how you can cancel the plant? but to reset it every 59 ticks?

Dec 2, 2017 9:22:19 PM
mixei4

mixei4

КГУ

GreenTea

GreenTea

Organization::ISDDesign

http://russianaicup.ru/game/view/161825 normal?) And I specifically didn’t kill him and it happened..

So in this case. So you have to specifically kill a weak enemy :)

Dec 2, 2017 9:25:45 PM
Pumba

Pumba

Omsk

So now write the algorithm as the end of the fight to escape the remaining airplane. What a bullshit…

Dec 2, 2017 9:33:34 PM
GreenTea

GreenTea

Organization::ISDDesign

In short, innovation is very controversial, because to do the dumbest strategy of operating it’s much simpler than a strategy which will look for the partisans, especially in the presence of fog of war. And it is necessary not only to find but to catch. What may not be so easy if they assume several groups of fighters, and you have at the moment, no fighters at all. It turns out that the strategy which has achieved an overwhelming numerical advantage, losing on points. If the organizers had a desire to unload the server, I would have advised them that: finish the game early if within say 2000 ticks were not active actions from the players (what is considered an active action - we have to think) and one of the strategies clearly achieved the benefits. Instead of typing the questionable changes that fundamentally affect the balance.

Dec 2, 2017 9:36:47 PM
xeltym

xeltym

Moscow

Pumba

Pumba

Omsk

So now write the algorithm as the end of the fight to escape the remaining airplane. What a bullshit…

Not quite, it is necessary that ten points are lost(kill yourself, well, or that the opponent scriptel something at the factory). Now it turns out for each killed unit, the difference in 0 points(opponent kills 10 units, over 10 times will have 10 points) in the end a draw, but if you kill yourself or let the opponent to produce equipment that will come in plus.

Dec 2, 2017 9:39:25 PM
SladeThe

SladeThe

Russian AI Cup Team

GreenTea

GreenTea

Organization::ISDDesign

In short, innovation is very controversial, because to do the dumbest strategy of operating it’s much simpler than a strategy which will look for the partisans, especially in the presence of fog of war. And it is necessary not only to find but to catch. What may not be so easy if they assume several groups of fighters, and you have at the moment, no fighters at all. It turns out that the strategy which has achieved an overwhelming numerical advantage, losing on points. If the organizers had a desire to unload the server, I would have advised them that: finish the game early if within say 2000 ticks were not active actions from the players (what is considered an active action - we have to think) and one of the strategies clearly achieved the benefits. Instead of typing the questionable changes that fundamentally affect the balance.

The change is not to do with speed test. Rather, the aim was to enhance the entertainment of games. If you reduce the game, it would be… shorter, but the rest will not differ from playing with a longer duration. In the Final, it may be disabling fog after 10,000 ticks.

Dec 2, 2017 9:43:23 PM
GreenTea

GreenTea

Organization::ISDDesign

Dec 2, 2017 9:59:36 PM
GreenTea

GreenTea

Organization::ISDDesign

now all you have to write a strategy that picks on all the corners of the map?

Dec 2, 2017 10:00:33 PM
UZERE
SladeThe

SladeThe

Russian AI Cup Team

The change is not to do with speed test. Rather, the aim was to enhance the entertainment of games. If you reduce the game, it would be… shorter, but the rest will not differ from playing with a longer duration. In the Final, it may be disabling fog after 10,000 ticks.

Drawers and running away does not increase the entertainment games

Dec 2, 2017 10:01:59 PM
GreenTea

GreenTea

Organization::ISDDesign

playing hide and seek and catch-up will increase the entertainment of the game…

Dec 2, 2017 10:06:25 PM
GreenTea

GreenTea

Organization::ISDDesign

And, I’m sure you can improve the tactics of the carpet to such an extent that the fighters are like flies running around the field and run. and you can’t catch stupid. but if all is not to kill - lost..

Dec 2, 2017 10:13:07 PM
SladeThe

SladeThe

Russian AI Cup Team

GreenTea

GreenTea

Organization::ISDDesign

http://russianaicup.ru/game/view/162117 this is true, tell me?

I think not. This game shows that the factor rewards are high. It will be reduced, and the movement slightly redesigned to be more logical.

Dec 2, 2017 10:16:55 PM
imrinat

imrinat

Ketovo

>bonus for survival

As I understand from the rules is the main spirit of competition that the victory goes to those who more effort was applied to the destruction of the enemy (as far as I understood it from the rules in the first paragraph of Chapter 2). And if it loses the one who put more efforts to the destruction of the enemy is not very logical.

Dec 2, 2017 10:17:29 PM
jurifoll

jurifoll

Moscow

>Rather, the goal was to increase the entertainment games. If you reduce the game, it would be… shorter, but the rest will not differ from the game with >a longer duration. In the Final, it may be disabling fog after 10,000 ticks.

Yeah, let’s 100000 after every 100 ticks will occur random nuclear explosion. And its radius will increase by 2 times each time. And will begin to appear neutral mob, attacking everyone. And cells areas will gradually become a new type of landscape - “trampoline”, which yields units of acceleration in 2 times. And still give + 1500 points a player if his units are lined up in the logo mail.ru

Do you think the participants specifically not finish off the enemy? Rather not have time to implement something and take everything into account. The potential to improve tactics is far from exhausted even at the tops, and then an additional headache. In the end, the entertainment games in the future, only go down, because need to write another crutches, to more advanced tactics that will not come.

Dec 2, 2017 10:19:05 PM
Kreol

Kreol

RetailNext

GreenTea

GreenTea

Organization::ISDDesign

http://russianaicup.ru/game/view/162117 this is true, tell me?

What norms someone has a weak strategy simple :)

Dec 2, 2017 10:19:27 PM
tyamgin

tyamgin

Simferopol

SladeThe

SladeThe

Russian AI Cup Team

I think not. This game shows that the factor rewards are high. It will be reduced, and the movement slightly redesigned to be more logical.

A crutch on a crutch :D

Dec 2, 2017 10:20:27 PM
lnuynxa
tyamgin

tyamgin

Simferopol

A crutch on a crutch :D

The order “not one step back”! Better one but at least somewhere

Dec 2, 2017 10:22:32 PM
tyamgin

tyamgin

Simferopol

SladeThe

SladeThe

Russian AI Cup Team

The change is not to do with speed test. Rather, the aim was to enhance the entertainment of games. If you reduce the game, it would be… shorter, but the rest will not differ from playing with a longer duration. In the Final, it may be disabling fog after 10,000 ticks.

Let all the fog off, why waste time on trifles. And for the finale may be what else will figure it out.

Dec 2, 2017 10:23:35 PM
Alexardos

Alexardos

Odintsovo

It is time the terrible abusow. Rollback until it is too late.

Dec 2, 2017 10:30:15 PM
jurifoll

jurifoll

Moscow

Sorry to repeat, but it would be better to enhance the entertainment would mitigate the limitation of actions. First efforts to overcome this limitation remain “behind the scenes” and entertainment do not contribute. Second, without this constraint, the participants would probably like something more interesting could demonstrate that it would be possible to check visually.

Dec 2, 2017 10:34:47 PM
mixei4

mixei4

КГУ

GreenTea

GreenTea

Organization::ISDDesign

http://russianaicup.ru/game/view/162117 this is true, tell me?

It turns out that starting with 10,000 tick the enemy has destroyed more units, while having less units. As for me, too achievement, although due to the yaderku it is not very difficult to do.

Don’t know if this is right, is it fair, but it certainly gives you the opportunity to come back and win the drain of the beginning of the match.

Dec 2, 2017 10:36:01 PM
Mystifier

Mystifier

Simferopol

And it’s called the good news? The current strategy seems simply destroyed such changes 6 days before the start of the round. Bravo to the organizers. After these changes, lost all desire to do something. It looks like it is some crutch to solve the performance problem.

Dec 2, 2017 11:24:48 PM
F10PPY

Not such a major change to make in the middle of the season, you will not regret the time human? Write 10 days one, then the new all. As for the Cape hailed also changed 5 days before the end of the amount of data and had to shovel at all. Of course you all try to make more fair and more interesting, but for this beta test was supposed to be, or do a survey. In General, to invent not need anything вк3/SK everything has already been invented. Depending on the limit of units you can increase the time of production of the equipment for example, or something, but that’s all I had to decide.

Dec 2, 2017 11:34:14 PM
stulentsev

Not such a major change to make in the middle of the season, you will not regret the time human? Write 10 days one, then the new all. As for the Cape hailed also changed 5 days before the end of the amount of data and had to shovel at all. Of course you all try to make more fair and more interesting, but for this beta test was supposed to be, or do a survey. In General, to invent not need anything вк3/SK everything has already been invented. Depending on the limit of units you can increase the time of production of the equipment for example, or something, but that’s all I had to decide.

but for this beta test was supposed to be

Now just week beta test.

Dec 2, 2017 11:36:19 PM
F10PPY

but for this beta test was supposed to be Now just week beta test.

Who said that?

Dec 2, 2017 11:39:51 PM
stulentsev

Who said that?

The rules are not read? :)

Dec 2, 2017 11:41:32 PM
GreenTea

GreenTea

Organization::ISDDesign

“the award strategy with a smaller number of units replaced by the costs of the technology.” are you sure that it will make the games more interesting and dinamichnee? I now think that many games will end in a draw 0 - 0. Where one bot was minimal advantage for the dead units it will leave at 0 will Not work to realize the victory at the expense of small advantages.

Dec 2, 2017 11:44:54 PM
F10PPY

The rules are not read? :)

Perhaps overlooked, I mean it means that the organizers have the right to change anything roughly an hour before round 2, and since this is a “beta test” any claims to him, well, so-so rules.

Dec 2, 2017 11:47:03 PM
Adler

Adler

Dzyarzhynsk

GreenTea

GreenTea

Organization::ISDDesign

“the award strategy with a smaller number of units replaced by the costs of the technology.” are you sure that it will make the games more interesting and dinamichnee? I now think that many games will end in a draw 0 - 0. Where one bot was minimal advantage for the dead units it will leave at 0 will Not work to realize the victory at the expense of small advantages.

you just have everything at the start to throw for 250 and the rules will be

Dec 2, 2017 11:48:11 PM
GreenTea

GreenTea

Organization::ISDDesign

And the game with factories it is virtually no impact due to the large bonus for the capture of the factory. But if one bot wins largest (let’s say he has 1000 units, and your opponent has 10 runaway), it is likely that the gap in points he has much more than 250 fine (500 for their units - 250 units for the opponent).

Dec 2, 2017 11:51:44 PM
Mystifier

Mystifier

Simferopol

According to the rules, the organizer can change everything he wants. But concentrated negativity from such changes still remains :( And the desire to spend time all of it abruptly disappears :(

Dec 2, 2017 11:52:11 PM
F10PPY

Generally limit mercy to enter and all you have reached the 500 points difference, the end of the game, quick, fast, no sad endings, servers free.

Dec 2, 2017 11:57:06 PM
Adler

Adler

Dzyarzhynsk

so the ending was understandable not tightened and not rude you can just add this fix:

if a team has less than 75 units, then we should start to burn they have HP.

Dec 3, 2017 12:07:04 AM
SladeThe

SladeThe

Russian AI Cup Team

Adler

Adler

Dzyarzhynsk

so the ending was understandable not tightened and not rude you can just add this fix: if a team has less than 75 units, then we should start to burn they have HP.

Yes, but I want to finish off the carried out strategy, not a punishment from heaven.

Dec 3, 2017 12:19:13 AM
Adler

Adler

Dzyarzhynsk

SladeThe

SladeThe

Russian AI Cup Team

Yes, but I want to finish off the carried out strategy, not a punishment from heaven.

finish off the last 75 units - boring and dull activity which usually does not add to the entertainment.

PS: BRAM, too, should die by fire :)

Dec 3, 2017 12:43:06 AM
GreenTea

GreenTea

Organization::ISDDesign

http://russianaicup.ru/game/view/162631 here is an example of the game close to a draw after the change. In fact, the winner will be the one who gets in last 1000 ticks more points.

Dec 3, 2017 12:51:04 AM
mixei4

mixei4

КГУ

Perhaps overlooked, I mean it means that the organizers have the right to change anything roughly an hour before round 2, and since this is a “beta test” any claims to him, well, so-so rules.

The beta test is almost over.

Dec 3, 2017 1:25:13 AM
kuzmalex

kuzmalex

pekach dev

Intuitively it seems to me that innovation is very artificial and not fair. To spur the game due to the fact that to take advantage of the player received up to the middle of the game? This creates a victory condition weaker strategy and gives a demotivator to the exchanges.

Dec 3, 2017 2:12:58 AM
alkozel

alkozel

Калужский филиал ГУ СДА при Спецстрое России

GreenTea

GreenTea

Organization::ISDDesign

“the award strategy with a smaller number of units replaced by the costs of the technology.” are you sure that it will make the games more interesting and dinamichnee? I now think that many games will end in a draw 0 - 0. Where one bot was minimal advantage for the dead units it will leave at 0 will Not work to realize the victory at the expense of small advantages.

Totally agree. Bite from the sandwich slices, create at least a little, but hurt, then it is in the zero output and the last ticks stray bullet makes the score 1-0 in your favor. Mmm… perfect….

Dec 3, 2017 6:29:37 AM
AlexKol

AlexKol

BB

Nafig to write something clever. All drain and cook partisans

Dec 3, 2017 7:21:34 AM
AlexKol

AlexKol

BB

Is that excuse what happened? http://russianaicup.ru/game/view/164065 A fair fight, or the enemy is some sort of a more clever strategy, my stupid glasses all the time was shot and I would be in region 0

Dec 3, 2017 7:29:52 AM
azt-yur

azt-yur

ОГУ

alkozel

alkozel

Калужский филиал ГУ СДА при Спецстрое России

Totally agree. Bite from the sandwich slices, create at least a little, but hurt, then it is in the zero output and the last ticks stray bullet makes the score 1-0 in your favor. Mmm… perfect….

http://russianaicup.ru/game/view/164532 - that’s exactly the scenario you described

Dec 3, 2017 7:54:17 AM
azt-yur

azt-yur

ОГУ

alkozel

alkozel

Калужский филиал ГУ СДА при Спецстрое России

Totally agree. Bite from the sandwich slices, create at least a little, but hurt, then it is in the zero output and the last ticks stray bullet makes the score 1-0 in your favor. Mmm… perfect….

http://russianaicup.ru/game/view/164532 - that’s exactly the scenario you described

Dec 3, 2017 7:54:18 AM
alkozel

alkozel

Калужский филиал ГУ СДА при Спецстрое России

azt-yur

azt-yur

ОГУ

http://russianaicup.ru/game/view/164532 - that’s exactly the scenario you described

Of course with buildings, such situations are almost excluded, but what about for example those who are breaks in the sandbox in 60 additional seats. I am sure that many of the strategies having advantages in points in the passive opponent does not climb on the rampage. And it is quite normal. But people want bread and circuses. Let’s take another fatality to enter, who is more beautiful smear the opponent 100,500 points. I believe that in games without buildings, this innovation should be removed.

Dec 3, 2017 8:20:34 AM
alkozel

alkozel

Калужский филиал ГУ СДА при Спецстрое России

It would be logical for example with a margin of at least 100 points to award a draw. But to write “acquired by overwork” scores too much.

Dec 3, 2017 8:34:59 AM
selya_

selya_

Гимназия №1

> Add buildings > Kill the profit of creating new technology Now buildings are really checkpoints. Here it is spectacular, Yes! And proposed will do nothing, because sitting there, really smart people, they know how to do better. Definitely. And I? I do Skalnik, where there I?

Dec 3, 2017 9:33:33 AM
AlexKol

AlexKol

BB

selya_

selya_

Гимназия №1

Add buildings > Kill the profit of creating new technology Now buildings are really checkpoints. Here it is spectacular, Yes! And proposed will do nothing, because sitting there, really smart people, they know how to do better. Definitely. And I? I do Skalnik, where there I?

Student and already involved?) Respect.

Dec 3, 2017 9:38:15 AM
alkozel

alkozel

Калужский филиал ГУ СДА при Спецстрое России

selya_

selya_

Гимназия №1

Add buildings > Kill the profit of creating new technology Now buildings are really checkpoints. Here it is spectacular, Yes! And proposed will do nothing, because sitting there, really smart people, they know how to do better. Definitely. And I? I do Skalnik, where there I?

Hang in there, students are not particularly favor. Do not know where the limit is 18 in the competition, and probably innovation, a few days before round has a deleterious effect on the immature mind.

Dec 3, 2017 9:50:07 AM
selya_

selya_

Гимназия №1

alkozel

alkozel

Калужский филиал ГУ СДА при Спецстрое России

Hang in there, students are not particularly favor. Do not know where the limit is 18 in the competition, and probably innovation, a few days before round has a deleterious effect on the immature mind.

I asked a few days ago, the final can be like, Yes, and stories too. Here with the final prizes of some kind of backed up. But it’s not that I don’t seek, just not so steep to get into the top 6

Dec 3, 2017 9:52:08 AM
Kreol

Kreol

RetailNext

AlexKol

AlexKol

BB

Is that excuse what happened? http://russianaicup.ru/game/view/164065 A fair fight, or the enemy is some sort of a more clever strategy, my stupid glasses all the time was shot and I would be in region 0

and I’m cool :) I also have glasses filmed that not so?

Dec 3, 2017 11:07:34 AM
Kreol

Kreol

RetailNext

Dec 3, 2017 11:11:44 AM
AlexKol

AlexKol

BB

Kreol

Kreol

RetailNext

and I’m cool :) I also have glasses filmed that not so?

I have acted faster since I essentially won. But due to the new “chips”, I stupidly went to 0

Dec 3, 2017 11:23:32 AM
Kristopher

Kristopher

АГ СПбГУ

@STARTED THE GAME WITH A MAN WHO DOES NOTHING @OH, NOW IZI WINES @BEAT THE CROWD REGULARLY NUCLEAR STRIKES @GAIN 250 POINTS @REACHED ARRV @DON’T KILL @WELL, I’M MORE KILLED, YES @LOSE ALL YOUR POINTS UP TO 20000 TEAK @BECAUSE ITS RATING IS LESS THAN YOURS, FOR A DRAW LOSE RATE

Dec 3, 2017 11:34:57 AM
ilt

ilt

Krasnogorsk

Kristopher

Kristopher

АГ СПбГУ

@STARTED THE GAME WITH A MAN WHO DOES NOTHING @OH, NOW IZI WINES @BEAT THE CROWD REGULARLY NUCLEAR STRIKES @GAIN 250 POINTS @REACHED ARRV @DON’T KILL @WELL, I’M MORE KILLED, YES @LOSE ALL YOUR POINTS UP TO 20000 TEAK @BECAUSE ITS RATING IS LESS THAN YOURS, FOR A DRAW LOSE RATE

The organizers believe that it is not easy wines, as fire draws.

Dec 3, 2017 11:37:26 AM
juso

juso

АО Финам

Interesting, by what principle selects the next game in the sandbox? From my last 10 games 8, not buildings. Getting ready for format especially makes no sense, and there stupid win and gain rating strategy, which generally can’t do anything with the buildings :(

Dec 3, 2017 11:52:58 AM
kuzmalex

kuzmalex

pekach dev

ilt

ilt

Krasnogorsk

The organizers believe that it is not easy wines, as fire draws.

the organizers believe that this strategy newbies, but because no matter what there will result

Dec 3, 2017 11:59:47 AM
Kristopher

Kristopher

АГ СПбГУ

kuzmalex

kuzmalex

pekach dev

the organizers believe that this strategy newbies, but because no matter what there will result

“Until you write strategy GreenTea, the sandbox will not rise!”

Dec 3, 2017 12:10:06 PM
MikeWazowski

MikeWazowski

Moscow

>Every 1000 ticks, starting from 10,000 and not including the last tick, the strategy spends 1 point on the contents of every 20 units, under the control of this strategy, but a total of at least 25 points for one payment. The number of points cannot fall below zero. but if units remain 19, also 25, or 0 points ? is this on all maps or only where there is structure ?

Dec 3, 2017 12:34:44 PM
dbf

dbf

A

> The points cannot fall below zero. but if units remain 19, also 25, or 0 points Even if you only have 1 unit, you pay 25 points

> this is on all maps or only where there is structure ? At all

Dec 3, 2017 12:35:59 PM
MikeWazowski

MikeWazowski

Moscow

if after 10,000 teak on the map without buildings the first player in 20 of the planes, and the second player 100 ARRV, provided that the first manages to fly away from a nuclear explosion, and the second time to heal, the winner is the first player ?

Dec 3, 2017 1:26:06 PM
azt-yur

azt-yur

ОГУ

MikeWazowski

MikeWazowski

Moscow

if after 10,000 teak on the map without buildings the first player in 20 of the planes, and the second player 100 ARRV, provided that the first manages to fly away from a nuclear explosion, and the second time to heal, the winner is the first player ?

No. To win the second. In a game without buildings both stable for every 1000 after 10K off for 25 points total 250 points

Dec 3, 2017 1:42:08 PM
Firen95

Firen95

Саратовский ГУ

The most popular score in games without buildings will now be 0-0. Too many points removed.

Dec 3, 2017 2:01:55 PM
novich-OK

novich-OK

Nizhny Novgorod

I sincerely wish that the organizers of the contest have to be so negative, but I’m afraid this time he fully deserved. Changes frankly ugly. To take the strategy that performed better in battle just because she is not spectacularly finishes off the remains of it is beyond good and evil.

Do not really understand this emphasis on entertainment. Assumed that the competition is primarily aimed at programmers who consistent the world is many times more important than pretty pictures on the website. Obviously, the administration has other priorities and they are extremely important to maintain the attractiveness of the event for the uninitiated. Well, maybe so, but the method of achieving this goal is, at best, puzzling. It would be funny if entertainment only suffer (the hunt for the remnants of istrebitelej this tedious Senate), but even without this the warped scoring system, leveling reward for the advantage in the battle of the main forces, severely discourages.

Dec 3, 2017 3:04:34 PM
imsohate

imsohate

Saint Petersburg

Not a bad idea. Logic is clumsy. Not practical to have considerably exceeding the size of the army by the end of the game, and it is clear in advance, if you have practically no air units, and your opponent has mostly air survived. Again, the optimal strategy (i.e., we consider both players as effective as possible in current position) in the case described above is the collection of the player with the larger army units in the pile and the destruction of significant parts. Entertainment obviously there will be more =)

Offer for every 600 ticks (after 10,000) if your strategy was not applied and NEITHER is 1 POINTS DAMAGE to any of the units of the opponent, to take away from the player with the bigger army 4% checkuseraccess (IMHO, to take a constant number, so that players with different number of points both in the end went to 0 is nonsense). The second player similarly, but taking, say, 2%, or in proportion to the size of the armies.

Dec 3, 2017 3:19:00 PM
mi5

mi5

Simferopol

http://russianaicup.ru/game/view/166962 Still 100 ticks, and would have lost. Need less to build.

Dec 3, 2017 5:30:35 PM
kuzmalex

kuzmalex

pekach dev

mi5

mi5

Simferopol

http://russianaicup.ru/game/view/166962 Still 100 ticks, and would have lost. Need less to build.

thresh

Dec 4, 2017 5:18:02 AM
bsn-iam

bsn-iam

Saint Petersburg

Good question. What is the point in the battle format is different from the next round? I would like to get an answer.

Dec 4, 2017 1:28:06 PM
bsn-iam

bsn-iam

Saint Petersburg

juso

juso

АО Финам

Interesting, by what principle selects the next game in the sandbox? From my last 10 games 8, not buildings. Getting ready for format especially makes no sense, and there stupid win and gain rating strategy, which generally can’t do anything with the buildings :(

Good question. What is the point in the battle format is different from the next round? I would like to get an answer.

Dec 4, 2017 1:29:18 PM
Predelnik

Predelnik

Rock Flow Dynamics

I agree. I would like in the sandbox fighting with buildings only.

Dec 4, 2017 3:29:40 PM
mixei4

mixei4

КГУ

bsn-iam

bsn-iam

Saint Petersburg

Good question. What is the point in the battle format is different from the next round? I would like to get an answer.

Sense such - on the results of the sandbox is extend it to the finals. If there are games only finals, you can connect to the competition after 2 rounds and only implement part of the strategy for the final. If implemented well, to reach the final.

Still in the sandbox, give prizes, and the strategy needs to fight equally well in all types of games to win.

Dec 4, 2017 6:07:35 PM
bypeso

bypeso

БНТУ

mixei4

mixei4

КГУ

Sense such - on the results of the sandbox is extend it to the finals. If there are games only finals, you can connect to the competition after 2 rounds and only implement part of the strategy for the final. If implemented well, to reach the final. Still in the sandbox, give prizes, and the strategy needs to fight equally well in all types of games to win.

and what’s the problem? if a person with no previous rounds of strata implements the finals, and will be on top of the sandbox, probably the last would be implemented? based on the fact that the final a priori - the hardest part. the complexity is growing, not declining.

Dec 5, 2017 5:06:39 PM
stulentsev

stulentsev

Moscow

bypeso

bypeso

БНТУ

and what’s the problem? if a person with no previous rounds of strata implements the finals, and will be on top of the sandbox, probably the last would be implemented? based on the fact that the final a priori - the hardest part. the complexity is growing, not declining.

Well, even realizes, if he can. :)

Dec 5, 2017 5:45:00 PM
Jeners

Jeners

Moscow

bypeso

bypeso

БНТУ

and what’s the problem? if a person with no previous rounds of strata implements the finals, and will be on top of the sandbox, probably the last would be implemented? based on the fact that the final a priori - the hardest part. the complexity is growing, not declining.

the trick is that when the rules change, have to change the approach. For example stupid in the first round slipped a ton of sandwiches, Yes, meat balls. Now this tactic is secondary, and probably those who moved in the other direction has a better chance than before, simply for the reason that rules change. And who knows how much will affect the behavior strategy, the fog of war in the final?

Dec 5, 2017 8:20:04 PM
PlayerDark

PlayerDark

Miass

Very little time is devoted to the development of strategies, if you are not able to sit the whole day debugging. I just finished up a decent strategy that could fight with the tops, but have no time and I most likely will not gain the necessary ranking to get the second round.

Dec 5, 2017 11:33:02 PM
mixei4

mixei4

КГУ

PlayerDark

PlayerDark

Miass

Very little time is devoted to the development of strategies, if you are not able to sit the whole day debugging. I just finished up a decent strategy that could fight with the tops, but have no time and I most likely will not gain the necessary ranking to get the second round.

A month and a half is the tournament. How much it needs to go in your opinion? In my opinion, even it’s a little long all.

And by the way? even if you do not pass the second round, you can go directly to the final :)

Dec 5, 2017 11:39:34 PM
Sanda

Sanda

JetBrains

PlayerDark

PlayerDark

Miass

Very little time is devoted to the development of strategies, if you are not able to sit the whole day debugging. I just finished up a decent strategy that could fight with the tops, but have no time and I most likely will not gain the necessary ranking to get the second round.

If the strategy is really good, and can play 50/50 against players from the second hundred, you can reset rating, and it will easily come out in the second round transoms. Now it needs about a dozen victories that should not be difficult for a good strategy in games with the bottom of the table.

Dec 5, 2017 11:40:47 PM
PlayerDark

PlayerDark

Miass

Sanda

Sanda

JetBrains

If the strategy is really good, and can play 50/50 against players from the second hundred, you can reset rating, and it will easily come out in the second round transoms. Now it needs about a dozen victories that should not be difficult for a good strategy in games with the bottom of the table.

And what does “reset rating”? I only see obscure button to “increase the fluidity rating.”

Dec 5, 2017 11:44:29 PM
PlayerDark

PlayerDark

Miass

mixei4

mixei4

КГУ

A month and a half is the tournament. How much it needs to go in your opinion? In my opinion, even it’s a little long all. And by the way? even if you do not pass the second round, you can go directly to the final :)

Well, I basically every day for a few hours on it was busy, and still did not, because a huge amount of time spent on code to gather the squad together, which in the end is not really even needed was because of the nukes.

Dec 5, 2017 11:46:32 PM
Sanda

Sanda

JetBrains

PlayerDark

PlayerDark

Miass

And what does “reset rating”? I only see obscure button to “increase the fluidity rating.”

This is it. In fact, this checkbox sets the rating of 1200, but greatly increases the growth for the win. After the first game the rating drops (even after the victory, this is the norm), but then grows by leaps and bounds.

I recommend first to download a new version, ponasati a bunch of games with different participants, to make sure that the new version is really much better than the current one, and if so, to drop the rating. If it is only slightly better, dumping is not necessary.

Dec 6, 2017 12:02:41 AM
bypeso

bypeso

БНТУ

PlayerDark

PlayerDark

Miass

Well, I basically every day for a few hours on it was busy, and still did not, because a huge amount of time spent on code to gather the squad together, which in the end is not really even needed was because of the nukes.

So the nukes were introduced before the start of the sandbox. Why write unnecessary code?

Dec 6, 2017 10:28:44 AM
PlayerDark

PlayerDark

Miass

bypeso

bypeso

БНТУ

So the nukes were introduced before the start of the sandbox. Why write unnecessary code?

I was somewhat baffled that I have not found a simple way to solve this problem, so I decided to bring code to make it working out of principle. I’d like to see after the competition, as the other made because I feel there is much more simple and quick method that I made.

Dec 6, 2017 11:20:58 AM
dimir

dimir

СПбГЭТУ (ЛЭТИ)

PlayerDark

PlayerDark

Miass

I was somewhat baffled that I have not found a simple way to solve this problem, so I decided to bring code to make it working out of principle. I’d like to see after the competition, as the other made because I feel there is much more simple and quick method that I made.

Did 4 games with You. Now let’s see what You got.

Dec 6, 2017 11:29:42 AM
PlayerDark

PlayerDark

Miass

dimir

dimir

СПбГЭТУ (ЛЭТИ)

Did 4 games with You. Now let’s see what You got.

Well, I have noticed that my AI is weak against nukes and it should fix, I’m going to do this, but it probably will die again because of this.

Dec 6, 2017 11:33:43 AM
ilt

ilt

Krasnogorsk

mixei4

mixei4

КГУ

A month and a half is the tournament. How much it needs to go in your opinion? In my opinion, even it’s a little long all. And by the way? even if you do not pass the second round, you can go directly to the final :)

Measured in full weekend for the formulation without the beta form is obtained: 1 round - 1st weekend Round 2 - 0 weekend 3 round - 0 output Given that three rounds is in fact different games, it turns out the paradox - physically-time of 1.5 months, and time to develop a little.

Dec 6, 2017 11:18:42 PM
mixei4

mixei4

КГУ

ilt

ilt

Krasnogorsk

Measured in full weekend for the formulation without the beta form is obtained: 1 round - 1st weekend Round 2 - 0 weekend 3 round - 0 output Given that three rounds is in fact different games, it turns out the paradox - physically-time of 1.5 months, and time to develop a little.

Why no beta test? Why during the round, not to prepare for the next one?

Dec 7, 2017 12:04:06 AM
peter.alexeev

Sandbox usually finally dies somewhere in February - including you all the holidays to write code. No prizes, of course. But if it is about prizes, then physically if time is more, those participants that measure time not the weekend, and just days just fly into space by the time the sandwich is ready)

Dec 7, 2017 12:34:55 AM
yukas66

I think it’s time in the sandbox to do 100% of the games with buildings. Strategies that play on maps with buildings, still did not make it to the finals. Why let them in the second round? Only place someone will take

Dec 7, 2017 2:20:01 AM
mixei4

mixei4

КГУ

I think it’s time in the sandbox to do 100% of the games with buildings. Strategies that play on maps with buildings, still did not make it to the finals. Why let them in the second round? Only place someone will take

300 first passed the second round without playing with buildings. Some of them may not have optimized the strategy for the second round. Why should they start the second round?

Dec 7, 2017 3:08:27 AM
ilt

ilt

Krasnogorsk

mixei4

mixei4

КГУ

Why no beta test? Why during the round, not to prepare for the next one?

It is theoretically possible to prepare for and during a round. The problem of preparing for round 2 and 3 is that no package of tools to adequately test your bot without the presence of the sandbox.

But practically, many people do not want to do this from time to time flying in the “good” news.

Dec 7, 2017 8:43:36 AM
mixei4

mixei4

КГУ

ilt

ilt

Krasnogorsk

It is theoretically possible to prepare for and during a round. The problem of preparing for round 2 and 3 is that no package of tools to adequately test your bot without the presence of the sandbox. But practically, many people do not want to do this from time to time flying in the “good” news.

The tools are all there. All modes in advance is localendar. 2 their strategies against each other to run.. of course it is necessary..

Dec 7, 2017 8:53:09 AM
ilt

ilt

Krasnogorsk

mixei4

mixei4

КГУ

The tools are all there. All modes in advance is localendar. 2 their strategies against each other to run.. of course it is necessary..

Tools that when out of the box )

In the box, even the quick guy, who minimally meets the requirements of round 2 and 3 no.

Dec 7, 2017 9:25:00 AM
ilt

ilt

Krasnogorsk

Sandbox usually finally dies somewhere in February - including you all the holidays to write code. No prizes, of course. But if it is about prizes, then physically if time is more, those participants that measure time not the weekend, and just days just fly into space by the time the sandwich is ready)

Mess around with the sandbox RAIC in the absence of competition is a dubious pleasure. Sandwich is spelled for 6-8 hours if it is a mast HEV. There is no need to delude ourselves with the advantage those who have more time. This time, you can merge code that will be very disgusting to watch in a week.

Dec 7, 2017 9:31:31 AM
Acuion

Acuion

НИУ ВШЭ - НН

ilt

ilt

Krasnogorsk

Tools that when out of the box ) In the box, even the quick guy, who minimally meets the requirements of round 2 and 3 no.

The final is not checked, but minimally takes into account the second round of the Smart strategy (to be replaced Quick Smart config Android lokal)

Dec 7, 2017 4:24:59 PM
ilt

ilt

Krasnogorsk

Acuion

Acuion

НИУ ВШЭ - НН

The final is not checked, but minimally takes into account the second round of the Smart strategy (to be replaced Quick Smart config Android lokal)

To call the production facility accidentally captured in the plant is minimal subject to the rules of round 2 as the language is not rotated.

Dec 7, 2017 5:13:08 PM