Non-instantaneous battles
Printed From: Illyriad
Category: Miscellaneous
Forum Name: Suggestions & Game Enhancements
Forum Description: Got a great idea? A feature you'd like to see? Share it here!
URL: http://forum.illyriad.co.uk/forum_posts.asp?TID=1066
Printed Date: 17 Apr 2022 at 13:49
Software Version: Web Wiz Forums 12.03 - http://www.webwizforums.com
Topic: Non-instantaneous battles
Posted By: HonoredMule
Subject: Non-instantaneous battles
Date Posted: 25 Sep 2010 at 20:15
|
Currently, whenever opposing armies meet, battle is automatic and instantaneous. This is a simple and functional implementation of battle, but I believe it can be much better, more realistic, and more exciting with little increase in complexity. Battles should take time. Results can still be calculated immediately, but instead of reporting them immediately, add a calculation for battle duration based on relative and total size of the armies as well as the stratagem used (i.e. raids should be shorter). Keep the attackers and defenders occupying the square "in battle" until the duration is concluded, then send the messages and return surviving troops, with the following exceptions: - An occupying (defending) army that is scheduled to leave (or recalled by messenger) at a time between battle start and end will still do so. Battle result becomes truncated to a percentage of losses/side equivalent to the percentage of battle duration completed. If forces from both sides are still present, a new result and duration is calculated and set for the remaining troops. Emails are generated showing progress so far and what has changed (an army retreated). - Another army arrives and adds its forces to either side. The same sort of result scaling + calculation of additional result again occurs. - A scout arrives. In this case, the email generated will report troop counts adjusted once again by the same "interrupted duration" technique, though without other intervention the final result is still not known nor altered. - An army arrives that is or NAP/ally of both sides. The army returns without doing anything. - An army arrives that is enemy/neutral of both sides and set to occupy. The army waits for existing battle to complete (at least until occupation duration expires) and fights the winner if the winner stays. During this time, another player sending to occupy that is enemy of all will battle the waiting 3rd party. This final exception could get complicated and therefore be excluded, but could add considerable excitement if included. --------------------- In plain English then, battles should take time relative to the balance and magnitude of participating armies, and by this means attackers should be able to work together by right of landing at the same time. This potentially upsets balance, but I believe it's fair and controllable, especially since failure to coordinate timing well could mean each army is destroyed by overwhelming force before the next arrives to reinforce it. The gameplay advantages are not really about balance, though, but excitement and getting more bang for your buck than a single battle report. When battles take time, the tide of battle can change by fluctuating circumstances (such as reinforcements and retreats), and various automatic messages and interim scout reports would give players a feel of greater control over and opportunity for involvement in the proceedings of battle. Also, large coordinated alliance operations would be more equitable in that everyone who shows up on time gets a piece of the action, rather than multiple battles being calculated in short succession where the first to arrive are the first served both death and experience,while others get nothing. |
Replies:
Posted By: bartimeus
Date Posted: 25 Sep 2010 at 21:38
|
I like it that every single one of your post is both accurate and elegantly written. I dont like it that every single one of my post seam quite dull in comparaison, as if they were written by a 15 years old. Anyway, I am 110% positively for your ideas. even Ikariam (another browser game, quite boring after a while) doesn't have instantaneous battle. (by the way, one of the few thing I liked about Ikariam was the concept of having multiple players having to work together because the rescource production upgrades were common to the whole island. maybe we could reuse this idea for something else in this game? specialy reguarding Faction? start another topic for that so as not to pollute HM's topic.)
------------- Bartimeus, your very best friend. |
Posted By: col0005
Date Posted: 26 Sep 2010 at 00:41
|
+1 very simple solution
I'm pretty sure many of us have been thinking something like this would be fantastic. later it would also allow for a new stratergem whereby smaller armies could delay larger armies on the move, gurrilla tactics or something. This option would obviously have to create higher casualties for the smaller army
|
Posted By: iluvpie3
Date Posted: 26 Sep 2010 at 00:53
| I like your idea HM,but I have one question about it:Would we actually control our army while its fighting and give it orders,or would it still be controlled by the computer and just take a longer time to finish? |
Posted By: HonoredMule
Date Posted: 26 Sep 2010 at 01:14
|
Without being interrupted, the outcome would be exactly the same--it just takes longer to finish. However, you could use scouts to track progress even before you know the final outcome, and recall if you don't like how its turning out...or if close enough, send reinforcements. The basic behavior of battle and calculations that determine outcome would be unaltered, but making it span a non-instantaneous time frame and holding the participants there means that additional forces can participate in the same battle--yet only to the extent that said forces are present during that battle. Balance would be altered somewhat given that (for example) two equal armies could attack an occupation matching their combined strength and the battle would be balanced instead of outnumbered 2/1 twice (provided the attackers arrive at the exact same time). Some additional territorial control is enjoyed as well; battles close to a participating player would be in that player's advantage as he can reinforce or withdraw quickly and the long-distance enemy cannot. But by far the greater difference would simply be that warfare becomes something that participants can observe over time. This is more about making the game interesting than fairness or even control. Battle conditions could actually change in response to additional (mostly pre-planned and executed/launched) actions. And instead of simply being told the final result, players would be told every time the course of battle is altered, having the opportunity to carry out further instructions. Players become much more involved/engaged spectators. ---- In my opinion, live player-controlled battle is a drastically huge change from how browser-based games normally work and infeasible socially if not technically as well. |
Posted By: col0005
Date Posted: 26 Sep 2010 at 02:02
|
actually I just realised that you mechanism is perhaps a little too simple as a re-calculated battle is very much in the favour of the larger side. Eg a battle interupted at 50% time by 1 unit and commander on the smaller players side.
Lets say p1 (100 troops) V.s. p2 (50 troops) with 25 %casualties for P1
so at 50% completion P1 has 87.5 troops P2 has 25 +2 (re-inforcments)
As you can see halfway through the battle P1 has gone from being twice the size of P2 to being more than 3 times the size.
Therefore even though P2 recieved the re-inforcments this system actually hurt P2.
Therefore the mathematics of a balanced system is actually a lot more complex than it seems on the surface.
I know this example is messy but I have to go to work. Let me know if I need to make things clearer.
Basically what I'm trying to say is that a larger force will recieve a smaller percentage loss of troops. Therefore as the battle progresses the smaller side will become more and more outnumbered meaning that a re-calculation will change from say 100 vs 50 (X2) to 90 V.s 30 (X3).
|
Posted By: TGE
Date Posted: 26 Sep 2010 at 03:30
|
I think this is a great idea because of all the new tactics it could allow. For example, if army 1 attacks army 2 from the west, then the owner of army 1 might be able to send another attack to the same point, but from the east (Using the not yet implemented rally points maybe?). This flanking could give a % attack bonus to the flankers, or maybe even scale the bonus with speed, which would make a lot of sense in my mind. This is just one of the many possibilities. ------------- 
Comic Sans started global warming |
Posted By: HonoredMule
Date Posted: 26 Sep 2010 at 04:21
|
(all numbers presented as theoretical units with equal attack and defense as all others) col, the outcome of a 50 vs 100 battle under the new system would be exactly the same as under the current one. If the 50 were reinforced by another 50 half-way through, then the losses on both sides for the original parties would be cut in half and a new calculation would occur based on the survivors of both sides plus the extra 50 units. So if the original outcome was 0 and 65 (losses of all and 35), then the halfway losses are 25 and 17 for a remainder of 25 vs 83. Add in the reinforcements and the new outcome is based on starting troop count of 25+50=75 vs 83, which is almost balanced (90.4% of the 83). However you look at it, the new outcome is nothing but favorable to the smaller party compared to how it was before. Instantaneous battle would resolve 50 vs 100 leaving 0 and 65 when the next 50 arrives for a new clash of 50 vs 65 (76.9% of the 65). And if both attacks had landed at the same time, the battle would effectively be 50+50=100 vs 100 for a perfectly even match. If you can't be bothered to well-coordinate and tightly time attacks, you shouldn't enjoy the full benefit of attack stacking anyway. Note also that integer rounding always favors the smaller party, and there would be several such occurrences when multiple events each trigger re-calculation of the outcome and apply percent-completion-based scaled losses. |
Posted By: some random guy
Date Posted: 26 Sep 2010 at 04:41
|
this idea needs to have a bit more refinement, as all good ideas do. for instance, if two armies begin an engagement and a third party wanted to reinforce one of the original combatants, how would the third army know what to do? ------------- Soon, very soon, my name will become synonymous with chicken alfredo.... mmm.... chicken alfredo.... |
Posted By: HonoredMule
Date Posted: 26 Sep 2010 at 04:56
|
I'm pretty sure that's already precisely covered. The 3rd party has to be in a NAP or confederation with the side it is supporting (as it would have to be anyway), and then automatically stacks with that side against the other upon arrival...regardless of whether the side it is supporting is the attacker or defender. And I also already noted that if the 3rd party is friends with both sides, it just turns around and goes home. The only time things get even a little vague or complicated is when the 3rd party has no friendly relationship with either side and in particular also wants to occupy the square. |
Posted By: HonoredMule
Date Posted: 26 Sep 2010 at 05:04
|
I suppose it's also noteworthy that a decision would have to be made as to whether a party was considered attacker or defender (important for determining their battle effectiveness relative to their unique stats). As I see it, there's two reasonable ways it could be handled: 1) Anyone present before a battle starts is a defender (like now), and anyone initiating battle or arriving after it begins is an attacker (regardless of who's side they take). 2) Anyone present before a battle starts is a defender (like now), and whoever sides with the defender after battle begins is also a defender. The other side's participants are all attackers regardless of when they arrive (thought it's obviously after the first defender). Which is better isn't nearly so clear-cut as what I've discussed so far. #2 is probably simpler/more compatible with existing battle resolution code (handling terrain modifiers etc.) and also more equitable, but I haven't fully thought it through yet. |
Posted By: xilla
Date Posted: 26 Sep 2010 at 05:50
|
I think you are making a meal out of it. If a party is on a square he is the defender, then strategems: Attack attacks the initial party, assists the attacker Reinforce assists the defender This makes it possible for neutral friends to assist their chosen sides as desired without the need to ally/NAP. If you are allied with the opposing army you bounce. Occupy becomes hostile to both parties if neutral, waits and attacks the winner of the battle (winner becomes the new defender). If ally/NAP then Reinforce. Multiple occupies stack and wait. If next occupier is allied to any reinforcing army but warring with another, bounce. Raid is a quick attack that instantaneously wears the opposition. Being instantaneous it would revitalise it and make it useable again. Raiding a battle: If you are not already involved and are neutral/enemy, you wear down both parties. If already involved, obviously raiding army raids opposition I like the idea. It makes battles more engaging and adds to the fun. Also, what col was saying Battle starts 50 v 100 (P1 has 50% of P2 troops) Battle allowed to resolve results in 50 v 35 killed. Battle interrupted halway. P2 reinforces 2 troops Battle starts 25 v 85 P2 initially had 100% more troops, now has 240% more troops. P2 won't lose 20 troops. P2 just abused the system, no? (Why wont P2 lose 20 troops- in the first battle P2 killed 43% more troops than s/he lost. If P2 lost 20 troops it would result in P2 killing only 25% more troops than lost despite having a 240% advantage on the opposition as opposed to 100% advantage.) |
Posted By: Mandarins31
Date Posted: 26 Sep 2010 at 06:14
|
HM, your ideas are interesting regarding the excitement of a battle. And we must consider that would be useful for fighting Factions as we could reinforce an attacking ally/NAP member. More, this is intersting because we could have big battles bettween 2 sides without doing any siege and risk to demolish/destroy a city. And this is important to precise with that idea you are creating a flee option with messengers. For me there are some things to think about this subject: 1) to be more realistic an army may not flee instantaneously, it may have a certain amount of loses by trying to flee... for exemple, we can imagine that when occuppying time is ended or when a messenger arrives, it put the army in raid mode. 2) we may consider if it would be judicious to add an option when you are occupying a square. I think about order to this occupying army either to flee if it is underattacked either to defend against any attacks... either to flee only if the attacking army is bigger. Finally what about assassins? For you would they be able to enter a battle what stays on a square? and if yes what would happen if a division lose its commender, or if the army lost all its commenders. |
Posted By: HonoredMule
Date Posted: 26 Sep 2010 at 06:55
This: "Multiple occupies stack and wait." is a heavily loaded sentence. Stack with whom and against whom and in what order? In this and other statements there are way too many variables and combinations to properly support all the possibilities in code--at least in a way that makes sense and can be understood by participants. Is it reasonable that other attackers sit around and pick off the winner--and then get picked off by the next in line themselves? How do waiting late Nth party arrivals team up with their allies, or do they at all? It's easy for us to throw around these loosely-defined ideas, but chewing all that fat down to actual code-level rules is quite a meal indeed, and one that will include details that will doubtlessly be hotly contested after release.
Maybe this should be mitigated somewhat, but did P2 abuse the system, or just outflank and surround a smaller foe? It would appear to me to still be (close to) a reasonable reward for actively directing either a brilliantly-planned strategy or a very lucky happenstance. I had a big spiel discussing this point, but removed it as a pointless rant. This is a matter that would need to be explored in code and simulation rather than speculation. |
Also, what col was sayingPosted By: HonoredMule
Date Posted: 26 Sep 2010 at 07:09
Perhaps it would be sufficient that messenger travel time still delays withdrawal. Otherwise, adding a relatively small delay with continued losses (at rates altered by yet another recalculation from the latest change) should suffice. Presumably the person recalling troops is losing and will lose troops even faster but for a shorter period, thus saving troops overall but at the cost of reduced killing effectiveness.
It sounds like you want more fine-grained stratagem options, in which case I think that could already be added as a separate feature with or without non-instantaneous battles, and it's an idea worth considering (at least in my opinion). In the mean time, just having the option to personally react by sending messengers is an increase in human interactivity thanks to having options where before there were only reports of foregone conclusions.
I don't think diplomatic options against battles should be implemented until diplomatic options against occupations are implemented. The issues and features required to resolve them are virtually identical for both. The latter is already on the dev list, I believe.
As I said, this could get very complicated. But ideally, I think (and that's an uncertain thought) that anyone trying to hold the square should get a chance to fight for it. It might be good enough to pigeonhole everyone into a round-robin affair, where newcomers get paired with the earliest nap/confed found, or if there are none, takes a new side. Every time there's 2 idle sides, due either to new arrival or battle completion, those two sides duke it out. The results would be very chaotic and costly, but that's probably how it should be in hotly-contested space. |
Posted By: Mandarins31
Date Posted: 26 Sep 2010 at 07:25
|
These are good points  |
Posted By: col0005
Date Posted: 26 Sep 2010 at 08:29
|
Actually I think what I was saying has been completely mis-interpreted. What I was saying is that if P2 is the smaller player and gets re-inforced by a VERY small force then P1 will recieve less casualties.
Ie initial calculation
P1 has 100 troops P2 has 50 troops end result is P1 75 troops P2 has 0 ie P1 has 25 casualties.
However P2 recieves a re-inforcment of 2 half way through so troop size is now P1 87.5 P2 25+2=27
P1 is now 3 times larger than P2 so final outcome may be 80 0 dispite the fact that the re-inforcments were for the smaller player.
My point is that the act of interupting a battle and re-calculating the result will in itself reduce casulaties for P1 (the larger player) as the larger player will outnumber the smaller player by a greater percentage halfway through a battle than at the begninning.
|
Posted By: bartimeus
Date Posted: 26 Sep 2010 at 08:58
|
I think thats because the casualty calculation doesn't take into account that most of P1 casualty are made at the beginning when P2 still has most of its troops. Thats a very good point you brought up col0005. ------------- Bartimeus, your very best friend. |
Posted By: col0005
Date Posted: 26 Sep 2010 at 10:46
|
sort of.... it's more that half way through a battle the smaller player will have lost 50% troops where as the larger player will have lost perhaps a simmilar number, but a much smaller percentage of their total force. Say 100 Vs 50 to 80Vs 30 (same change in numbers much greater difference in percentage) Given this it may be a lot simpler just to introduce rounds where a battle3 lasts a specific number of rounds depending on terrarin, troop type and numbers. This would mean that a re-calculation would occur regardless of re-inforcments being recieved or not. (Yes I do realise there has been a post on exactly this topic)
|
Posted By: bartimeus
Date Posted: 26 Sep 2010 at 11:30
|
Making same amount of casualty on both side like you show us in your exemple doesn't satisfy me. A very large army won't lose as much as the enemie armie if this enemie army is much smaller. thats just common sense. Not only will the percentage be smaller, but the numbers will be smaller too. on a different subject, maybe once pathfinding comes into play, we can make the [tiles the army crossed just before reaching the occupied tile] to intervein in the battle outcome with a small bonus if for exemple, archers come from the forest to the plain, or if cavalry gallop down a slope from a hill toward a plain. (i put the brakets so as to avoid ambiguity concerning the subjet of last sentence.)
------------- Bartimeus, your very best friend. |
Posted By: Zangi
Date Posted: 27 Sep 2010 at 07:51
|
+1 to this general idea. Push and pull of a battle with outside influence... Make it so its ok for people to split their armies up instead of lumping them into 1 huge one. The only problem is that military will be 'gone' longer then it is now. (Perhaps return time can be faster to make up for it?) 1. Basic military reinforcements. Simple and predetermined. 1a. Flanking or Hitting the rear of engaged enemy, with pathfinding. Only affects 1st turn. 1b. Positive or negative modifier from terrain reinforcement comes from. Only affects 1st turn. 2. Combat Spells 2a. Temporary enhancement and/or debuff spells. Players can cast onto on-going battles, if in range. (Perhaps send a 'Messenger Mage' to cast a spell on an on-going battle... or time their arrival with the battle.) 2b. Perhaps these spells can be prepared before hand, so that it doesn't favor the more active player too much... Commanders have limited spell slots/cast conditions. With limited conditions like... 'if enemy majority is sword, use Tenaril's Rust' Debilitates sword weaponry, -% for opponents swords for X turns. (Does not apply to reinforcements) Even a spell to allow the commander to reflect 1 debuff and/or steal a bonus off of the enemy in reaction. (Would take more spell slots or something.) Also, add a after X turn condition. >.> Especially when you expect reinforcements of your own or enemy reinforcements. 3. 'Diplomats' assigned to the army and/or incoming as support from other players. 3a. Scouts, if you have more can enhance the terrain effects in your favor. (Scouts tell the commanders what the best spots to fight on.) [Bound to be losses as they get caught.] 3b. Spies, increases command efficiency(Divisional Soldier Bonus) of the commander. (Spies give reports on the enemy movements, plans and troop make-up.) [Bound to be losses as they get caught.] 3c. Thieves, either assigned to 'steal' stuff from enemies, lowering their battle efficiency -or- defends your camp from getting stolen from. [Bound to be losses as they get caught or fight each other off.] 3d. Assassins, either assigned to try to specifically kill enemy diplomats, wounding/killing the commanders, or defending the camp from that happening. [Bound to be losses as they get caught or fight each other off.] A 'wounded' commander would not fight(no 'hero' bonus), but still be able to give command bonus. A dead commander is just dead, no more bonus. [Very low chance to happen? Or at least commander needs to be at really low health and 'wounded' first for this to happen.] 3e. Saboteurs, blows stuff up, chance of 'killing' some enemy troops, lower chance of backfiring on yourself, nothing happening, or just blowing themselves up. Yes, more is still better, but, now, players that specialize in stuff other then military can now contribute. Heavily. My biggest issue with this game is the battles, being as it is, 'win it or lose it all' in an instant and that defenders can stack in their favor for an all in-one-overwhelming force, while attackers from multiple players have to trickle in one by one. (Though the game at least balances it to favor attackers.) The only way to have mass battles of this sort is siege... Though.... siege could be a long on-going battle with this change... |
Posted By: col0005
Date Posted: 27 Sep 2010 at 08:44
|
The example with same lossess on both sides wasn't intended to illistrate (and certainly not promote) a scenario where large armies have no bonus. It was intended to illistrate how as the battle progressess the battle will shift further and further in favour of the the larger player, the smaller player recieving more casulties will only further prove my point.
If a re-calculation is done this change in balance will be taken into account. I'm not arguing that this shouldn't happen. What I'm arguing is that this change in balance should either never be taken into account, or should always be taken into account.
If a battle was calculated in turns the balance change would always be taken into account.
However if the battle is only ever interupted by re-inforcments this balance wont be taken into account, will always favour the larger player when it does happen.
|
Posted By: HonoredMule
Date Posted: 27 Sep 2010 at 14:09
|
I do (now) recognize the issue you're describing, and I could speculate at a few methods of addressing it. However, it's more of an implementation issue and only the devs will ever really know what works best. I'm gratified that others see my vision of potential in the basic concept. It is more for the devs' sake that I presented it in a manner attempting to prove by example that the idea is at a medium level both feasible and implementable without bringing chaos to the higher game balance. |
Posted By: Hora
Date Posted: 27 Sep 2010 at 20:35
|
Only problem is how to encourage players to attack squares instead of towns. Maybe a solution comes with pathfinding, when occupying armies can intervene with moving ones. There could be two options: 1. One army blockading the way, and the other one attacks to get through. 2. One army waits next to the way to attack everything that comes by (making the one on the road the defender). Sure needs some working over, but I think you get the picture. |
Posted By: Shrapnel
Date Posted: 27 Sep 2010 at 21:58
Let's see if I understand how this would work:
If 3rd party is already at location at beginning of battle:
NAP'd or Confed with defender: Sides with defender
any other time: attacks defender
If 3rd party arrives in middle of ongoing battle:
If NAP'd or Confed with one side and orders are to attack or raid: sides with that party
If NAP'd or Confed with both sides and orders are to attack or raid: Goes home
If NAP'd or Confed with one side and orders are to occupy: sides with that party
If NAP'd or Confed with both sides and orders are to occupy: occupies square
If NAP'd or Confed with one side and orders are to reinforce: sides with that party
If NAP'd or Confed with both sides and orders are to reinforce: goes home? (I think that the 3rd party should reinforce the defender unless the 3rd party has a Confederation with the attacker and reinforce without attacking either party in the case of a Confederation)
In the case where the 3rd party has no NAP or Confederation with either party, I propose the army gets to choose which party it sides with, then if any survivors are left that are on different sides, the survivors finish the battle. Either that or it is considered the completion of a "battle round" and survivors go home or the victor continues to occupy if that was its order.
Regardless of how it works out, excellent idea HM. |
Posted By: KarL Aegis
Date Posted: 27 Sep 2010 at 22:42
|
So instead of slaughtering both armies with a flank attack, they wait it out until the victor reforms its attack lines and then engage all fancy like? My, what great manners warriors have today.------------- I am not amused. |
Posted By: xilla
Date Posted: 28 Sep 2010 at 01:45
|
Only for occupy order: Instead of stacked occupies, an army could wait on the last adjacent square it was on before it encounters the battle. Once battle is resolved, the army could move in and occupy if winning party went home, and if winning army is occupying, then reinforce/attack based on same principles as reinforcing a siege. If encountering a waiting party then wait at your last adjacent square, etc.. Time the army starts waiting is the decider for order. |
Posted By: col0005
Date Posted: 28 Sep 2010 at 02:26
|
Waiting in that manner is far to ordered. If an army wants to get to a location only to find 2 forces already fighting then yes they may wait. However If another force arrvies I imagine that they'd attack the rear of the force watching the battle. If there are multiple sides then an impartial algorithm should allow a joined battle but with a random distribution of attacks and therefore a random outcome. Ie P1 is the defender.
for P3 213 troops attack P1 537 attack P2
P2 763 troops defend against P3 321 attacks P1
P1 375 defend against P2 472 defend against P3
Each attack and defence set is determied seperately each round ie there are 3 randomly generated, independant battles.
Note that a higher numbered player is the attacker against a lower numbered player (order that they arrived)
Oh commander divison bonus applies to all battles but individual stats would apply to only one battle.
|
Posted By: Zangi
Date Posted: 28 Sep 2010 at 13:31
I like this. 'Hey guys, just wanted to see what you are doing.' *Attacks the other players.* 'Now get out of my area.' |
Posted By: Shrapnel
Date Posted: 28 Sep 2010 at 13:58
|
So far we've only assumed 3 parties. Does this get exponentially more difficult with n parties? |
Posted By: Zangi
Date Posted: 28 Sep 2010 at 16:02
Well, seems pretty simple. Why not have a 4th or 5th party too? (That is if you somehow have +3 different non-NAP/Confederation parties fighting over a square or 'investigating' a battle with extreme prejudice. "Really, get off my road, you are stopping traffic." To promote that sentiment, maybe it would be in the interest of nearby players to 'encourage' a quicker, more decisive end to the battle happening on their 'toll road'. A reason to patrol and police 'toll roads'... If they are paying you for faster passage, they should expect travel to be free and clear. Not slower then taking the off-road. And if you can't take care of it... what right do you have collecting? Tie-in to banditry on 'toll roads' too... Unaligned caravans, diplomats and armies passing through may unwittingly be dragged into the fighting or at least be heavily delayed by the nearby fighting. Danged drive-by gawkers slowing traffic... That is... once pathfinding comes into place and armies can meet each other on the road... ~~~~~~~~~~~~~~~ Player intervention on the road can either end the battle early, after 2-3 battle rounds, they go their own way after exchanging insults. (This option should only be available to the alliance/player that runs the 'toll road' or within X squares of their city.) Or just simply try to destroy all trespassers for the experience... (The default option I guess...) |
Posted By: HonoredMule
Date Posted: 28 Sep 2010 at 16:07
It does seem that some of the others discussing this point are swallowing vast tracts of complexity or injecting vagueness and confusion without batting an eyelash. Just imagine how much more complex things get if you expect idle parties to wait on adjacent squares, which then get occupied/attacked by parties! And what if the required square (or all of them) is lake or a city? However, I think my initial proposal addressing this point was actually fairly complete and functional:
|
Posted By: Hora
Date Posted: 22 Jan 2011 at 00:40
|
Why stacking those armies on neighboured squares? Those armies already on the square would continue fighting, but both would have to engage some percentage of their army in defending against the 3rd player being the aggressor (I think a percentage based on troopnumbers would suffice). Same for every attacking army coming next... sounds like a big deal calculating victims, but I think it might turn out quite simple  Could favour the attacking 3rd force a bit, but that would be realistic crushing into a ongoing battle, with noone looking around  kindly Hora Edit: I know I'm actually reviving an old thread (3 month inbetween), but it was refered to in an other thread, so I needed to reread all that stuff...  (not meant negative) |
Posted By: some random guy
Date Posted: 22 Jan 2011 at 01:11
|
I knew this would be dug up! Good job Hora!  A new army should face the 2 others like they were united against it, and the 2 others should have the same predicament. ------------- Soon, very soon, my name will become synonymous with chicken alfredo.... mmm.... chicken alfredo.... |
Posted By: Zeus
Date Posted: 22 Jan 2011 at 02:08
|
Posted By: Hora
Date Posted: 22 Jan 2011 at 12:58
Thanks  Well, yes, but I thought of the new army only facing parts of those armies already there. Let's make a very easy example, with 3 persons a 100 of the same units (like the other ones with equal defense, attack and evrything..) Player1 occupies the square with 100 units Player2 attacks with 100 unit, that would give a hard battle player3 attacks after player2. With the current rules he would take over the square killing off the on remaining unit, and looses noone... not instantanious (or something..) battles would give some other scenarios... ok, then scenario 1: player3 attacks immediately after player2, also with 100 units he would find two equal armies struggling each other, so half of his army attacks player1, other half player2. Those would have to defend while continuing their fight with half the force. So it would look like 3 battles 50:50. If one battle is over, the one remaining unit of the winner would join in on one of the other fights, until only one (or perhaps even two ) units of one player are left...well, he would have won obviously.now scenario 2: player 1 and 2 are halfway through their battle when player three arrives makes 50 players each. player three would see two equal armies, so split his army 50:50 player1 has now to fight with two enemies, same player3 That inflicts some calculation actually...hmm...oh, ugly I tried to calculate for a dividing the forces they get actually attacked with, so player 1 divides 25:25, player 2 then must divide 17 to 33, but that inflicts back on player 3, giving an ugly three body problem (I think most of you know that from physics) after some rounds I got a result, where player1 and 2 defended with their whole army against player3, so exactly what I didn't want  So it seems, we should go for total size, means players 1 sees 50 and 100, so divides 17:33, same player 2, player 3 stays with 50:50 So that gives battles 17:17, and twice 33:50, giving all in all 0 Survivers for players 1 and 2, with player3 taking the square with aproximatly 40 remaining units (still haven't fully understood the victim calculation  ).A further possibility is to readjust armies during battle, means introducing some steps. That would inflict even more victims on player3 (don't want to calculate victims depending on number of steps  )So I'd prefer a new calculation each time an army arrives, parting the armies on total starting number of possible targets. Gives the most realistic result and further would inflict smallest calculations at least... |
Posted By: Createure
Date Posted: 22 Jan 2011 at 17:27
|
I like HM's idea much more having been in some large battles during this tournament. I think it would be much more fun if alliances mounting 'joint attacks' could actually really work together without having many seperately worked-out battles. Also I reckon it would mean that alliances that can coordinate well-timed assaults gain alot more benefit from their efforts, loosely spaced attacking army arrival times would mean defenders can deal with the incoming attackers 1-by-1 instead of all together. I think the idea about "a third party, unallied/NAPed to either sides in a currently battle" shouldn't just wait until the battle completed before they join in the fight though, I think it could result in alot of: '2 alliances set up a good evenly matched fight to enjoy' and then some 3rd party gets put into a much stronger position as soon as the other 2 alliances fight each other... this results in a stale-mate type game social mechanic where no alliance will ever attack another for fear of a third one stepping in and wiping out them both. IMO a 3rd party in battle should just join in fighthing against both the 1st and 2nd parties at the location simultaneously... with the amount of attack power split between each according to the ratio of troops (or unit strength) that the 1st and 2nd parties currently have. Similarly the 1st party would then have to direct it's attack against the 2nd and 3rd parties based on the ratio of troops (or total strength) between the 2nd/3rd etc. |
Posted By: HonoredMule
Date Posted: 22 Jan 2011 at 18:24
|
Hora, I avoided 3-way battles at all cost expressly to avoid the spider's nest of complexity that would result. I wanted to minimize not only the complexity but also the amount of code reconstruction required to implement progressive battles. Features that are too ambitious have a nasty habit of exploding in spectacular fashion, and as it is Stormcrow weighed in on the other thread alluding that the idea was too big an overhaul on the current system. The fallout of making others wait in line is that the odd party enjoys an easy battle against the leavings, but I believe that is fair, realistic, and interesting. It introduces risk on players who would conduct battles whenever there is a 3rd party likely to take interest. It represents an opportunistic advantage that the 3rd party could and would take if given the opportunity. And, it interacts with new forces in interesting ways. For example, the 4th party could be reinforcement for one of the first 2 parties, but it gets delayed by fighting the third. I could suggest other fallout and intregue, but I've got to run at the moment.  |
Posted By: Hora
Date Posted: 22 Jan 2011 at 22:31
That ratios are the problem, I tried to work it out for my example, but that's a problem called three body problem, with three things influencing each other. Those you can only solve numerically, means step by step, and that really goes down on the server. My solution was just to take start values as when there are 50 units of player 2 and 100 of player 3 attcking, player 1 splits units 1:2, and leaves it so until the battle is over, or a new army arrives. That would just inflict one ratio calculation for each army, at every point in the battle, where actually a new army arrives. In the example of three armies, the mechanic now would result in one battle calculation for player 1 and 2, and one for the surviver and player 3 (makes 2 battle calcs) With my idea, that would actually develope to one battle calc, one time ratio for the interruption, then each army doing a ratio, how many units there are (important: total number, not attacking numbers!) and consequently 3 battle calcs. Then one army goes down, leaves another battle calc for the remaining two. Sums up to 5 battle calcs and 4 ratios... Problem is for big fights, that goes up facultativelly with number of factions, but how many battles do we have, where there are more than 4 factions not being confederated to one of the other three? Well tournament would be such a case,... , hmm..., but I think for such special cases would turn up special solution |
Posted By: Createure
Date Posted: 23 Jan 2011 at 02:09
|
The '3 body problem' relates to planetary mechanics I think? does that even apply here? But yeh maybe HM (and SC) have a point here with regards to complexity. I think maybe we are overthinking this a bit. |
Posted By: Llyorn Of Jaensch
Date Posted: 23 Jan 2011 at 08:18
|
Being fashionably later to the party, and only reading the 1st couple o' entries to this thread (before Homer, monkeys clanging symbols kicked in...) I'd like to firstly express my support for the idea. Secondly and most importantly I'd love some feedback from the Dev's. Is the idea feasible? Is the above in any form planned for implementation? Etc etc. Whilst we can continue to discuss the particulars in ever more increasing detail the point is really moot unless the Dev's can confirm it is an option as the overall concept of more detailed battles is itself a universally good (supported) concept. |
Posted By: HonoredMule
Date Posted: 23 Jan 2011 at 15:51
| Stormcrow called it (as I roughly recall) an "interesting" idea that the devs considered but are not currently planning, and hinted that it would require too much work and reconstruction of current functionality. I suspect the functionality issue may be more related to the server's event queue than battle resolution, but that's just a guess. |
Posted By: Rorgash
Date Posted: 04 Jan 2013 at 01:13
Hey I like this +1 (NOT BUMPING AT ALL )------------- 
|
Posted By: Arctic55
Date Posted: 04 Jan 2013 at 16:36
NONONONONONONONONONO!!!!!!!! I left Ikariam because of that. Nobody would help me in that game. They were all selfish people. NONONONONONONONO!!!!!!!!! I completely hate this idea. I want my commanders back as soon as they die. I like the battle system the way it is. I BEG YOU DEVS, DON'T CHANGE THE BATTLE SYSTEM!!!!!! |