Fortress tie breaking

What do you want to see in Armagetron soon? Any new feature ideas? Let's ponder these ground breaking ideas...
User avatar
Z-Man
God & Project Admin
Posts: 11715
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Fortress tie breaking

Post by Z-Man »

Ok, so there are some zealots out there who don't want anyone to even think about taking the winzone. What else can we do to make sure that fortress rounds actually end? As much as I like 1 vs 1 fights, if I want to see those, I play Sumo.

I have this simple idea: in addition to the two fortress zones, have two slow sumo zones (meaning that they don't get conquered by the enemy, but need to be kept alive by the owning team every once in a while). Each team's sumo zone would be centered at the enemy fortress zone and initially be large enough to cover more than half of the grid so it can be reached easily at round start. It then shrinks over the course of maybe 5 minutes like a regular sumo zone does. The sumo zone would decay slowly it would only collapse after 30s without an owner inside. In addition to keeping enemies outside of your fortress, you'll also have to have at least one player in the enemy half, and as time progresses, he has to get closer and closer to the enemy fortress.

Effect: a team that is not attacking at all automatically loses. The classic 1 defender vs N attackers situation gets resolved in favor of the attackers, acknowledging the obviously at least slightly superior play of their team leading to the advantage. The less common case where both fortresses are fought over will be resolved when the sumo zones shrink; eventually, they'll be small enough for the defender to keep the attackers out. If the defenders on both sides are so good that they can hold the fortress against the attackers, the defenders that manage to keep a larger piece of space around the fortress free from attackers will win.

Of course, this requires zones 2.0. Branch 0.2.8 can't have different conquest settings for the two zone types and can't make you own a zone that is closer to the enemy than to you.
User avatar
Rain
Round Winner
Posts: 300
Joined: Sat Apr 15, 2006 2:59 pm
Location: a random empty server playing with bots

Post by Rain »

Idea is interesting and problem is real, but i think this solution could change players behaviour in game. first problem with it is that objective of game change from "defend own zone and conquer enemy zone" to "stay inside the extra zone". in fact teams will be more worried about to stay inside that zone than to the real objective. also this means no possibility to play 1 on 1 or 2 on 1 at all. by the way the unbalace between teams will affect more the possibility to win.
i think your solution perfectly solve the problem of "waiting", meantime it creates some others.
i find more interesting to set a max time for the round. like 5 minutes as you said. during those 5 minutes round can end in 1 of these 4 possible cases:
1) team gold conquered blue zone - team gold won
2) all blue players died - team gold won
3) team blue conquered gold zone - team blue won
4) all gold players died - team blue won
after 5 minutes round will end anyway. if there is at least 1 player for each team, round ends in a tie. 5 points to each team. this add two important elements present in most sports: time and tie.
also it doesnt deviate to main objective ("defend own zone and conquer enemy zone") that could be improved giving 1 point for each kill instead of 2.
(it needs a timer in clients HUD, i think...)
User avatar
Joe
Core Dumper
Posts: 161
Joined: Fri Aug 27, 2004 6:11 am
Location: C eh N eh D eh
Contact:

Post by Joe »

What if you took z-mans idea and but had it so the other zones only showed up after a certain time. Or make it like the old nuke zone, if someone doesn't die after this alloted(sorry for spelling) then the game ends in a tie. This would make teams more eager to go to the zone and take away from the long boring 1v1's like the one with durka vs obli in the ladle final.
User avatar
Lucifer
Project Developer
Posts: 8747
Joined: Sun Aug 15, 2004 3:32 pm
Location: Republic of Texas

Post by Lucifer »

We already have philippe's suggestion, it's a death zone that covers the whole arena and appears at some time t after the round starts.

I'm partial to z-man's original suggestion, and if phillipe's going to stick around long enough to do it, it'll be in 0.3.1. :) That is, the two zones move towards each other throughout the round, so that we can aim for a 5 minute deadline (so to speak) where the centers of the two zones will be near each other (I'm thinking of a function that has an asymptote at 0, shifted to the middle of the grid).

However, I'd like to consider ideas that instead of forcing the team to attack, either strengthen attack or weaken defense. The way it was originally, attack was stronger in general. Then sumo was created to train and became wildly popular on its own. Now the balance is strongly tipped toward defense (as it should be, most games have this as a trait), and I don't think it's going to shift back to attack. I've got two ideas here that may not actually do anything to make attacking worthwhile, but I think are interesting on their own. :)

* Defender can collapse his zone. Now the risk to deciding to attack is gone, other than attacking a fortified position. A variation of this would be allowing the defender to collapse his zone and *move* it (i.e. he must set it up within some period of time or he is killed and the zone reconstructed at its origin or some random place). This might actually make defense stronger, though, but it creates a time where the defender is weaker than attackers because he has no zone and can be killed.
* When it becomes 1 v 1, or some arbitrary time period has elapsed, both zones just disappear, and in say 2 more minutes, a death zone will kill both players. This is achievable for 0.2.8.3, I think, it's just another setting to determine when to collapse the zones, and the death zone (as previously mentioned) is already there. This weakens defenders in favor of attack by the simple mechanism of removing the thing they're defending.
Check out my YouTube channel: https://youtube.com/@davefancella?si=H--oCK3k_dQ1laDN

Be the devil's own, Lucifer's my name.
- Iron Maiden
User avatar
Rain
Round Winner
Posts: 300
Joined: Sat Apr 15, 2006 2:59 pm
Location: a random empty server playing with bots

Post by Rain »

it sounds to me a "sumoization" of fortress. hmm....
User avatar
Z-Man
God & Project Admin
Posts: 11715
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Post by Z-Man »

Rain wrote:in fact teams will be more worried about to stay inside that zone than to the real objective,
Well, maybe, but hat zone will be on the enemy side and will shrink around the enemy fortress, so staying inside it is synonymous to attacking. Yes, the plan is completely unsuitable for 1 vs 1 or 1 vs 2, but Fortress is a broken game mode in those cases anyway, so no harm done. Well, actually, 1 vs 1 could be fun with these modifications, due to the different timescales, each player will have to prevent the other from rushing to his fortress while at the same time being forced to touch the enemy half every once in a while.

The problem with any mechanism of the "wait x minutes (y minutes without death), then do stuff" is that if the stuff done is of benefit to a team, they may try to stall the game to get it to happen. Example: a weak attacker against a strong defender, knowing he has no chance to take the enemy fortress alive, would be tempted to sit and wait for the nuke.

Moving zones are very hard to defend. The attacker can just install a roadblock in front of the zone's path, essentially he just has to defend an area the zone will be at some time in the future. Knowing that he only has to survive for a finite time, a spiral of death would probably be enough sophistication if the zone's speed is fast enough.

Lucifer has a good point. The basic problem that causes games to stall simply is that the defender has a natural advantage in the current settings. Let's analyze the possibilities one attacker has against one defender (many attackers is less of a problem, they can squeeze from several sides if a false sense of honor or the free points for the defender or other considerations rule out a hole attack):
1. he can try to push the defender away or squeeze him, sumo style. The defender has the clear advantage here of already being in the position he needs to hold, so this takes time, effort and risk.
2. he can try to squeeze through the gap, possibly killing the defender and at least reversing the situation.
3. he can move about threateningly, pretending to plan 2., hoping the defender will make a mistake and die.

Basically, we want to avoid 3. to be the best strategy the attacker can choose for any skill of attacker and defender. It has been suggested to remove or reduce the tail shrink setting; that makes defense harder. Unfortunately, that means the defender is more likely to screw up on his own, making 3 a better strategy. Strategy 2 gets easier, too, since the defender will have to leave a larger gap again. We'll have to experiment; I'll put up my tournament server with a .1 tail shrinkage.

I'll also do a reverse experiment: I'll put up a server with massive tail shrinkage. There is no way a defender will suicide, so 3 is no option at all. Of course, conquest settings are modified; the zone will get conquered eventually if one attacker and one defender are inside it. The massive tail shrinkage currently causes severe client glitches, I'll be on it, of course.

The server names are "Fortress Test: * tail shrink".

Hmm, another very simple meta-rule solution would be to immediately declare the round a draw once it gets to 1 vs 1 (and both teams had more than 1 player to begin with).
User avatar
Rain
Round Winner
Posts: 300
Joined: Sat Apr 15, 2006 2:59 pm
Location: a random empty server playing with bots

Post by Rain »

z-man wrote:Well, actually, 1 vs 1 could be fun with these modifications, due to the different timescales, each player will have to prevent the other from rushing to his fortress while at the same time being forced to touch the enemy half every once in a while.
usually 1 on 1 is played one player attacking and the other defending alternating round by round.
z-man wrote:they may try to stall the game to get it to happen
true.
z-man wrote:Moving zones are very hard to defend.
hmm... wait... hm... moving zones...
and what about both fortress zone moves to center and fix in a final sumo. actually zones are x=250 y=50 and x=250 y=450. how about they meet in x=250 y=250? should be very cool.
each team will be forced to attack, in the way they are forced to defend their own zone.
so the zones dont chahge their radius, but they move with a slow rigid movement till they meet in center of the grid.
also time and tie elements can be added simply fixing a time (like when zones meet the round ends). in this way a supposed attacker must attack or the enemy defender will conquer his zone.
problem is when it happens that a lot of players die at start or too soon, leaving only one player per team. this could case a stasis of two zones defended till they meet in center (so boring) or a nice fight against those 2 players. we cannot know yet.
z-man wrote:Hmm, another very simple meta-rule solution would be to immediately declare the round a draw once it gets to 1 vs 1 (and both teams had more than 1 player to begin with).
going back to sumo idea, we could make zone move only when it becomes 1 on 1 and with a faster movement still ending in a sumo in center.
i dont know, must think about it again.
User avatar
Z-Man
God & Project Admin
Posts: 11715
Joined: Sun Jan 23, 2005 6:01 pm
Location: Cologne
Contact:

Post by Z-Man »

Yeah, I know what people usually do when it's 1 vs 1, but it requires a minimum of coordination with the enemy.
Rain wrote:hmm... wait... hm... moving zones...
and what about both fortress zone moves to center and fix in a final sumo. actually zones are x=250 y=50 and x=250 y=450. how about they meet in x=250 y=250? should be very cool.
That's exactly the hard to defend case. It may be worth a try anyway, though: it's only hard to defend a moving zone when you're alone. When you're not alone, the other players can defend the future zone locations. Different areas of interst will emerge: the current location of the zones, the future path of the zone, and the final zone position. Maybe we'll see the best sumo players of each team rush for the final zone position and start battling for it right away, while the other players try to guard/disrupt the zones' journey there. If it's 1 vs 1 from the start or if it gets to be 1 vs 1 early and both players are defenders, they won't sit around and defend the zone and wait for the zones to meet in the center. This strategy would be beaten by defending the enemy zone's future position, and that strategy would be beaten by rushing to the enemy zone, which in turn is beaten by defending the zone where it is now, waiting for the attack and doing a counter rush. A classic RPS situation, forcing the players to try to secure the arena center while keeping an eye open for rush opportunities and blocking off the opponent. Could be fun.
User avatar
2020
Outside Corner Grinder
Posts: 1322
Joined: Thu Dec 29, 2005 9:21 pm
Location: the present, finally

Post by 2020 »

interesting ideas posited
but most clutter the general minimal design...

may i suggest looking at the final scoring within a time limit?

teams have 100 seconds to finish the game once it reaches 1v1
(much like the 30s shot rule in basketball)

5-5 split if no team has entered the enemy zone
(ie they are in midfield or one circles the other without taking any space from the def)

7-3 split if a team has entered the enemy zone


so
the attacker may risk trying to enter the zone to turn a 5-5 split into a 7-3 advantage
and the def may keep hanging onto a 5-5 split
but once his def is breached in any way and a sumo ensues
then the def is facing a 3-7 loss

that is
the game-design favours the defence
but the scoring favours the attack

keeps the game clean and simple :)
hold the line
User avatar
wrtlprnft
Reverse Outside Corner Grinder
Posts: 1679
Joined: Wed Jan 04, 2006 4:42 am
Location: 0x08048000
Contact:

Post by wrtlprnft »

Don't for get my very simplicistic idea (requires some ugly gGame hacking or zonesv2): Have the winzone give only half as many points as conquering the enemy zone/being the last team alive. Maybe that would make less people complain if someone uses it.

Or maybe don't have the zones move together but just have the zones disappear after some time, wait maybe 15 secs and then create a new sumo zone in the middle of the grid?
There's no place like ::1
User avatar
philippeqc
Long Poster - Project Developer - Sage
Posts: 1526
Joined: Mon Jul 12, 2004 8:55 am
Location: Stockholm
Contact:

Post by philippeqc »

Humm... somehow it feels that you are trying to force complex actions onto the game to break what seems to be a defense advantage. Why not simply breaking the defense advantage by making the zone slightly bigger?

-ph
Canis meus id comedit.
User avatar
2020
Outside Corner Grinder
Posts: 1322
Joined: Thu Dec 29, 2005 9:21 pm
Location: the present, finally

Post by 2020 »

philippeqc wrote:Humm... somehow it feels that you are trying to force complex actions onto the game to break what seems to be a defense advantage. Why not simply breaking the defense advantage by making the zone slightly bigger?

-ph
from the outset?
in which case
crowding might be too easy
since players seem to be so good these days

nice solution thought...
we should try this before introducing moving zones et al
hold the line
User avatar
Rain
Round Winner
Posts: 300
Joined: Sat Apr 15, 2006 2:59 pm
Location: a random empty server playing with bots

Post by Rain »

2020 wrote:so
the attacker may risk trying to enter the zone to turn a 5-5 split into a 7-3 advantage
and the def may keep hanging onto a 5-5 split
but once his def is breached in any way and a sumo ensues
then the def is facing a 3-7 loss
i think three possible ends are enough and complete: win-lose-tie
moreover if i know i can get 7 meanwhile i can get 3, i would prefer 5.
this means the game stalls, again. imho.
wrtlprnft wrote:Don't for get my very simplicistic idea (requires some ugly gGame hacking or zonesv2): Have the winzone give only half as many points as conquering the enemy zone/being the last team alive. Maybe that would make less people complain if someone uses it.
most people just dont like to end the game with a win zone. usually only new players take it. the risk is it will be useless, because even if it appears, players will go on playing without touch it.
wrtlprnft wrote:Or maybe don't have the zones move together but just have the zones disappear after some time, wait maybe 15 secs and then create a new sumo zone in the middle of the grid?
yes, maybe. but i think the real problem is slipping out of our hands.

the problem is a round can stall in a boring 1 on 1 attacker-defender.
i want to underline too, the necessity of "tie" element, imo.
i think we should analize what is positive and negative in a round referring to the main objective of fortress type: conquer enemy fortress.
damn, we always considered to gain points, but we didnt consider to lose points.
down my thoughts:
1) kill a enemy gives no points to the killer and -1 to the victim team
2) kill a teammate 2 times in the same match auto-generate a poll
3) no player is allowed to start a poll
4) run time of a round is T:200 (200 seconds)
5) a round ended with no zone conquered is declared tie and gives -3 points to each team
6) a round won by conquering a zone gives 5 points to the winners and -5 points to the losers

hmm.. i forgot some while writing.. lol

oh, ok

7) a match ends after 10 rounds, wins the team who reach more points, if points are the same the match is declared tie.

explaination:

1) this should help focusing the objective and to make a team plays like a team (note "-1 to the victim team": players dont have personal score)
2 and 3) it could solve the problem that made z-man close his server, if i remember well. also it helps people re-knowing the "silence" option, that seems to be forgotten. it surely needs a feedbacks webpage of the server and a little of in-game admin stuff.
4) usually rounds take from 100 to 150 seconds.
5) introduces the tie and forces in 1 on 1 situation, the attacker to attack and the defender to sumo him and try to kill him, in fact both risk to lose 3 points.
6) teams will be more interested to conquer enemy zone.
7) interesting for tournaments. and it is funny that a team could win with negative score. lol

i like this way because it introduce the negative points, not used so much, but very useful to direct players behaviour.
it is interesting too how a match could change while playing, in fact scores are more dynamic.

what do you think?
User avatar
2020
Outside Corner Grinder
Posts: 1322
Joined: Thu Dec 29, 2005 9:21 pm
Location: the present, finally

Post by 2020 »

Rain wrote:
2020 wrote:so
the attacker may risk trying to enter the zone to turn a 5-5 split into a 7-3 advantage
and the def may keep hanging onto a 5-5 split
but once his def is breached in any way and a sumo ensues
then the def is facing a 3-7 loss
i think three possible ends are enough and complete: win-lose-tie
moreover if i know i can get 7 meanwhile i can get 3, i would prefer 5.
this means the game stalls, again. imho.
try hanging on for a 5-5 split --
not so easy if you are the def since the attacker may get a slice of the zone
and if you are the attacker
you may be tempted to risk it in order to get that 7-3 split...

the solution is essentially simple:
100 time limit with 1v1
since 1v1 is entertaining to watch on the whole
it's only the rare event when the def and attacker can't resolve it
the 7-3 split is just to tilt it gently to the attacker's advantage
hold the line
newbie
Core Dumper
Posts: 138
Joined: Wed Jun 01, 2005 5:04 pm

Post by newbie »

the easiest way would be to end after some time the 1vs1 fortress part and give an extra sumo round only for those two players
1, 4, 6, 9, 11, 21, 24, 33, 34, 35

Image
Post Reply