FreeOrion

Forums for the FreeOrion project
It is currently Mon Dec 11, 2017 5:06 pm

All times are UTC




Post new topic Reply to topic  [ 21 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Thu Aug 29, 2013 5:41 pm 
Offline
Design & Graphics Lead Emeritus
User avatar

Joined: Sat Sep 23, 2006 7:09 pm
Posts: 3858
Location: USA — midwest
It seems to be there's a lot of interest in adding new features to combat to make it less one-dimensional: parts that will be useful in some contexts, ships that will be powerful against some other types and weak against others, etc. Basically because combat gets boring.

Actually a linear progression of power was something purposefully aimed for. We used to have different kinds of weapons/shields and armor that were supposed to have a rock/paper/scissors type of relationship (I'm not sure if it was ever functional).
I took them out because of the limitations of non-tactical combat UI. When you have ships A > B > C >A, and no way to try to deploy ship C against ship A, while deploying other ships against something else-- that's just complications that make for a worse, more frustration experience. There's the secondary issue of how to figure out that C >A, when you don't have the tangible experience of seeing graphics of C ships destroying A. Gleaning that info from a highly detailed sitrep is not my idea of a good or enjoyable interface. (having the data there for those who want to really go in depth is fine, but you shouldn't have to pour over such lists to have a clue about whats going on.)

Its no good designing cool features that don't work in the actual game context, or that make the game less fun for years until something else is implemented.


To make combat less boring, at this time i think we would be better off focusing on giving the player more control over combat, rather than adding more wrinkles to the hulls and parts. But how does that fit in the with long-delayed 3D tactical combat?


eleazar wrote:
Personally, i'd be happy to abandon the idea of 3D RTS combat. It just adds so hugely to the art/code/balancing/ai required to complete this project.
If suitably skilled and motivated programmers for implementing the 3D interface don't become available, abandoning a 3D combat system will be unavoidable. So far it hasn't really prevented other development work, and there are still numerous other game systems that need implementing as well, so I've been content to wait. Despite the recent enthusiasm for complicating the combat system, I still feel like that is the case. At the least, efforts could be made to nominally use some of the above design ideas for any combat complicating that is done.


If we are still sitting on the fence about weather the game will ultimately include 3D tactical RTS combat, i think there are still some things we can do that will:
    * make combat more interactive and interesting
    * take much less time than 3D RTS combat
    * and be equally useful weather we ultimately have 3D RTS combat or not

:arrow: It is this: implement the Combat Queue, and the Auto-Resolve part of combat, in which you can give your fleets basic orders like "Retreat", "Defend", & "Charge!".

Here's basically how i see it playing out:

When Enemy ships end up at the same system, a "Battle" is added to the combat queue. When you click on that item a window opens up, and the map zooms to the combat location, and you can examine the involved fleets, and whatever else you want.
You have several options, "Retreat", "Defend", & "Charge!", and sometimes maybe special options enabled by certain ship parts. Each side would make his choice without knowing the other's choice.

Once the orders issued, that "Battle" in the queue gets a checkmark or something to show that it has been dealt with. The player can change his mind at any point before the end of the turn. If the player doesn't issue orders on a particular battle the same AI that the AI players use will choose an order for the fleet.

As it works now, combat will be resolved between turns, and the results displayed on the next turn's sit rep -- including the orders each side issued.

Later more bells and whistles could be added such as different orders for different fleets in the same battle, and more tactically specific orders.

_________________
—• Read this First before posting Game Design Ideas!
—• Design Philosophy

—•— My Ideas, Organized —•— Get an Avatar —•— Acronyms —•—


Top
 Profile  
 
PostPosted: Fri Aug 30, 2013 6:41 am 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12040
Location: Munich
It might help to discuss what the different orders would do / mean.

Overall, I'm not enthusiastic about asking for / requiring orders for every combat... It seems like it would slow down turn processing by requiring repeated player input.

There's also already an aggressive / passive (or whatever they've been renamed in the GUI) setting for fleets, which seems a bit redundant with issuing orders of this sort every turn. Giving a single order per combat (potentially affecting multiple fleets) also seems like it's less useful than setting the aggression of fleets separately... Could not a similar result be achieved by modifying ship fleet behaviours in combat depending on their fleet's aggression setting?


Top
 Profile  
 
PostPosted: Fri Aug 30, 2013 3:43 pm 
Offline
Design & Graphics Lead Emeritus
User avatar

Joined: Sat Sep 23, 2006 7:09 pm
Posts: 3858
Location: USA — midwest
Geoff the Medio wrote:
It might help to discuss what the different orders would do / mean.

This is by no means a highly refined chart, but it should give you a clearer idea:

Attachment:
battle-orders.png
battle-orders.png [ 15.77 KiB | Viewed 2226 times ]


Geoff the Medio wrote:
Overall, I'm not enthusiastic about asking for / requiring orders for every combat... It seems like it would slow down turn processing by requiring repeated player input.

I share that concern somewhat-- which is why the interface i describe does not require the player to do anything. He could totally ignore the battle queue if he wanted, and for battles where there is little question of the outcome, i expect the player to do so.

Additionally i'd like to see the queue sorted by the most important battle at the top, so they player doesn't need to scan through a bunch of minor skirmishes. "Important" could be complicated, but simply defined it would be the battles with the highest total ship weapons+structure+shields.

As another way to waste less time, the default order for a multi-turn battle should be whatever the player chose in the previous turn.

Certainly it will add time to turns if the player can issue some orders before battle-- but only a tiny fraction of the amount of time some sort of RTS combat would add. And generally speaking, i think players will consider it time well spent-- the total hands off system can be frustrating.


Geoff the Medio wrote:
There's also already an aggressive / passive (or whatever they've been renamed in the GUI) setting for fleets, which seems a bit redundant with issuing orders of this sort every turn. Giving a single order per combat (potentially affecting multiple fleets) also seems like it's less useful than setting the aggression of fleets separately... Could not a similar result be achieved by modifying ship fleet behaviours in combat depending on their fleet's aggression setting?

The core rational of the aggressive/passive setting is for stealth, to allow the player to choose when the stealthy ships, hide and when they strike. That rational remains, i believe, though i'd love to see a way around it. In short, the stealthy player needs to decide to be stealthy before battles are added to the battle queue, because then the other player will know that something is up at that location when battles don't actually happen, and enemies aren't seen.

Besides stealth i've never like how the aggressive/passive setting functioned. Most of the time it just doesn't matter because the other side will be set to aggressive and can see you. If we give those settings more significance in how combat plays out and/or add more options, it simply requires the player to make decisions about combat without usually knowing what that combat is going to be like. Players usually don't send a war fleet against an enemy knowing that they will be greatly outnumbered. It's when you make that discovery that a "Retreat" order really becomes useful. Yeah if "passive"="retreat" it would work great for unarmed ships, but that's about it.

Considering efficiency, IMHO for a non-wreckless player using those fleet settings (if they mattered more) would take more time, since you have to hunt around to make sure all fleets are set appropriately for any potential battles, many of which won't actually occur. Making a decision only when battle occurs, provides with the info on the actual battle should allow quicker, better decisions, and the more fun, more in-control, less frustrating experience that is the goal for this topic.

_________________
—• Read this First before posting Game Design Ideas!
—• Design Philosophy

—•— My Ideas, Organized —•— Get an Avatar —•— Acronyms —•—


Top
 Profile  
 
PostPosted: Sat Aug 31, 2013 9:51 pm 
Offline
AI Lead, Programmer
User avatar

Joined: Sat Sep 22, 2012 6:25 pm
Posts: 4390
eleazar wrote:
Geoff the Medio wrote:
Overall, I'm not enthusiastic about asking for / requiring orders for every combat... It seems like it would slow down turn processing by requiring repeated player input.
I share that concern somewhat-- which is why the interface i describe does not require the player to do anything. He could totally ignore the battle queue if he wanted, and for battles where there is little question of the outcome, i expect the player to do so. ... Certainly it will add time to turns if the player can issue some orders before battle-- but only a tiny fraction of the amount of time some sort of RTS combat would add. And generally speaking, i think players will consider it time well spent-- the total hands off system can be frustrating.
Your combat options are interesting ideas that seem they could work well as advance settings like aggressive/passive is handled now. I suppose that for single player games it could also work out ok as a responsive system, though for AI v AI battles we might want to bypass it to save time.

On the other hand, for a mid-turn response system in a multiplayer game I share Geoff's concern about how much time this would take, which seems like even if just a fraction of the time that would be required for RTS, would still be a potentially long additional time, during which everyone's game-state is uncertain and possibly frozen. It doesn't seem to me that it helps at all that a player might choose to ignore the battle until their response ability timed out.

Suppose you have three players, all having visibility and ships near systems X and Y. On turn 50, players A and B each move aggressive armed ships to systems X and Y; player C is biding their time. Presumably in the new scheme, as now, at the start of turn 51 everyone gets an initial update showing visible movement. Currently there is then fairly fast combat resolution during which the UI is somewhat responsive, accepting orders but not yet updating the UI in response to those orders. In a small to medium size galaxy that lax of response is hardly noticeable but it can be a quite noticeable period of time in a large galaxy. Then there is a totally-frozen UI period during which the main post-combat update is processed. Then the UI updates and becomes normally responsive to orders. If we allow player input after the turn movement phase but before combat resolution, then all players will be delayed, whether they personally have any combat or not. They might be able to click on their screen, but the actual gamestate for the start of that turn is still uncertain; even if we tried changing the local client to let it respond to orders before the post combat update, it not even have enough info to reliably determine if an order was going to be valid to give. And how much time would the combatants be given to evaluate a battle situation and make a decision? Geoff has previously cited concerns about being able to add any limit to turns and those same concerns would apply here. Even if a limit could be applied, would 5 seconds or 10 seconds be enough? Even with only two battles in this case, that means that player C will be waiting for an extra 20 seconds during which their UI is not fully responsive, and they will notice that for sure.

_________________
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0


Top
 Profile  
 
PostPosted: Sat Aug 31, 2013 10:25 pm 
Offline
Design & Graphics Lead Emeritus
User avatar

Joined: Sat Sep 23, 2006 7:09 pm
Posts: 3858
Location: USA — midwest
Dilvish:

You misunderstand my proposal. There would be no mid-turn, battle phase before the regular turn. Battle orders would be issued at any point during the player's turn, and be executed after "end turn" is hit. And then battle orders, like all the other orders would be executed together, and results displayed on the next turn.

This does mean that the resolution of a battle would happen 1 turn latter than it currently does now, but that's fine.

_________________
—• Read this First before posting Game Design Ideas!
—• Design Philosophy

—•— My Ideas, Organized —•— Get an Avatar —•— Acronyms —•—


Top
 Profile  
 
PostPosted: Sun Sep 01, 2013 12:47 am 
Offline
Designer and Programmer
User avatar

Joined: Tue Aug 14, 2007 6:33 pm
Posts: 2058
Location: Orion
eleazar wrote:
There would be no mid-turn, battle phase before the regular turn. Battle orders would be issued at any point during the player's turn, and be executed after "end turn" is hit. And then battle orders, like all the other orders would be executed together, and results displayed on the next turn.

How do we know which battles are going to happen before the movement phase occurs?

_________________
Warning: Antarans in dimensional portal are closer than they appear.


Top
 Profile  
 
PostPosted: Sun Sep 01, 2013 12:57 am 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12040
Location: Munich
eleazar wrote:
There would be no mid-turn, battle phase before the regular turn. Battle orders would be issued at any point during the player's turn, and be executed after "end turn" is hit. And then battle orders, like all the other orders would be executed together, and results displayed on the next turn.

This does mean that the resolution of a battle would happen 1 turn latter than it currently does now, but that's fine.
Since battle orders would presumably be per-system, presumably a player could just have a standing battle order for all systems, include one that haven't been explored yet and which don't contain any ships owned by that player (but which could after the next movement phase).

I was under the impression you were suggesting a separate stage during turn resolution where all battles that are about to happen that turn are determined and presented to players, and they would then pick what battle order to use for that battle based on the visible contents on the system (and with a default to the previous order in that system).


Top
 Profile  
 
PostPosted: Sun Sep 01, 2013 1:47 am 
Offline
AI Lead, Programmer
User avatar

Joined: Sat Sep 22, 2012 6:25 pm
Posts: 4390
eleazar wrote:
Dilvish: You misunderstand my proposal...This does mean that the resolution of a battle would happen 1 turn latter than it currently does now, but that's fine.
So, in my example, on turn 50 players A and B enter their orders that move armed aggressive ships into systems X and Y. In the current setup, that movement would be shown at the next movement phase (which I think of as the start of turn 51, though I don't have a large enough saved game right now to check and distinguish the phases after hitting the turn button); then this combat at systems X and Y is auto-resolved, system and planetary status is updated, and then the players can enter their turn 51 orders. As I understand it, in the proposed setup, these turn 50 orders would be shown, but there would be no corresponding combat auto-resolution in turn 51. Instead the players would have the chance to enter combat orders for the system, and combat would then be resolved after turn 51 orders are entered, probably just before the corresponding movement phase, so that players would see the combat results at the start of turn 52. Is that right?

In many ways, it seems kind of like if two fleets got into combat now but for some temporary reason didn't do any damage to each other on the first turn of combat, and were then going to be put through another combat on the next turn even if they were going to retreat then (but subject to the extra settings). So I guess that could be the viewpoint used to decide system status (blockaded or not, planetary production, etc) during the interim turn until combat was resolved. Although, some of the current rules regarding colonization attempts or blockades depend on whether an armed aggressive enemy is present at the end of a turn, which could be easily maintained indefinitely with weak throwaway ships under such a setup. But maybe I'm still not understanding right.

_________________
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0


Top
 Profile  
 
PostPosted: Sun Sep 01, 2013 10:51 am 
Offline
Vacuum Dragon
User avatar

Joined: Sun Sep 25, 2011 2:51 pm
Posts: 500
Dilvish wrote:
Although, some of the current rules regarding colonization attempts or blockades depend on whether an armed aggressive enemy is present at the end of a turn, which could be easily maintained indefinitely with weak throwaway ships under such a setup. But maybe I'm still not understanding right.
That's entirely up to the ordering of turn processing. If combat is resolved before colonization and colonization before movement, then the colonization will succeed before the enemy forces move in. It will also succeed on the next turn if the enemy forces are eradicated.

_________________
[...] for Man has earned his right to hold this planet against all comers, by virtue of occasionally producing someone totally batshit insane. - Randall Munroe, title text to xkcd #556


Top
 Profile  
 
PostPosted: Sun Sep 01, 2013 3:04 pm 
Offline
Design & Graphics Lead Emeritus
User avatar

Joined: Sat Sep 23, 2006 7:09 pm
Posts: 3858
Location: USA — midwest
Dilvish wrote:
As I understand it, in the proposed setup, these turn 50 orders would be shown, but there would be no corresponding combat auto-resolution in turn 51. Instead the players would have the chance to enter combat orders for the system, and combat would then be resolved after turn 51 orders are entered, probably just before the corresponding movement phase, so that players would see the combat results at the start of turn 52. Is that right?

In many ways, it seems kind of like if two fleets got into combat now but for some temporary reason didn't do any damage to each other on the first turn of combat, and were then going to be put through another combat on the next turn even if they were going to retreat then (but subject to the extra settings).

Yeah that.

To be abundantly sure everyone understands here's an example:

Turn 1
Blue sends warship to unknown system Foo. Trip will take 1 turn

(hits "end turn")

Movements, and all other stuff is processed. Ship arrives at Foo, which contains a monster.

Turn 2
Since there's a potential battle, there's an entry re: Foo added to the battle queue.
On the galaxy map there's an explosion/starburst icon around the Foo system circle indicating the unresolved battle.
The player can give orders for the battle and set research, change the production queue, order ships to move, etc. in any order he wants. Blue chooses to "defend", to prolong the battle so backup can arrive. He directs another warship to Foo.

(hits "end turn")

Stuff is processed. Monster and ship take damage. Lastly 2nd warship arrives.

Turn 3
Explosion/starburst icon remains around Foo. The battle at Foo is again in the queue, and the default action is now "Defend". But blue now has 2 ships, one undamaged in the fight. Blue orders his fleet to "Charge!".

(hits "end turn")

Stuff is processed. Monster is destroyed, but Blue looses one ship.

Turn 4
Dark explosion/starburst icon is around Foo now indicating resolved battle. Sitrep reports the conclusion of the battle.

_________________
—• Read this First before posting Game Design Ideas!
—• Design Philosophy

—•— My Ideas, Organized —•— Get an Avatar —•— Acronyms —•—


Top
 Profile  
 
PostPosted: Tue Sep 03, 2013 4:28 pm 
Offline
Creative Contributor
User avatar

Joined: Sat Sep 13, 2003 6:52 pm
Posts: 1396
So essentially what is happening is

Aggressive v. Passive (2 options) moves to 4 options (retreat, defend, attack, charge)

And battles take one extra turn to update. (since they all have the possibility of receiving player orders.. ie change in default status..attack/defend being the only 'default' statuses that a new battle would have..ie whether fleet in motion is aggressive or passive)


Top
 Profile  
 
PostPosted: Thu Sep 12, 2013 9:24 pm 
Offline
Space Krill

Joined: Thu Sep 12, 2013 1:08 pm
Posts: 5
Choosing rock, paper, or scissors before every fight just isn't interesting, even if you can buy some kind of super scissors tech (RPS-style is so simple you could reduce any such tech to a simple combat bonus). Real strategy comes from figuring out how to reach high priority targets as a result of having interesting mechanics and diverse combat types (e.g. how you checkmating a King). But the chess board, chess pieces, and chess strategy come together, not alone, which is why I think you are having this difficulty trying to implement one at a time.

I understand that every game has in it choices that are strong against other choices and in that way they're similar to RPS. But the best games have that as an emergent thing where there are many best actions over the course of the battle and they are all hard to do, either physically or mentally. RPS gameplay is pretty one dimensional by itself, I'd rather be thinking about what I want my board, my pieces, and my game mechanics to be like.

IDK, just an outsider's opinion, but there's no shortcut to interesting combat. When you gave FreeOrion interesting civilization mechanics you didn't do it by asking the player growth/production/research and then directly penalizing them based on whether their opponent chose the "counter".


Top
 Profile  
 
PostPosted: Fri Sep 13, 2013 4:51 pm 
Offline
Creative Contributor
User avatar

Joined: Sat Sep 13, 2003 6:52 pm
Posts: 1396
phyvo wrote:
Choosing rock, paper, or scissors before every fight just isn't interesting, even if you can buy some kind of super scissors tech (RPS-style is so simple you could reduce any such tech to a simple combat bonus). Real strategy comes from figuring out how to reach high priority targets as a result of having interesting mechanics and diverse combat types (e.g. how you checkmating a King). But the chess board, chess pieces, and chess strategy come together, not alone, which is why I think you are having this difficulty trying to implement one at a time.

I understand that every game has in it choices that are strong against other choices and in that way they're similar to RPS. But the best games have that as an emergent thing where there are many best actions over the course of the battle and they are all hard to do, either physically or mentally. RPS gameplay is pretty one dimensional by itself, I'd rather be thinking about what I want my board, my pieces, and my game mechanics to be like.

IDK, just an outsider's opinion, but there's no shortcut to interesting combat. When you gave FreeOrion interesting civilization mechanics you didn't do it by asking the player growth/production/research and then directly penalizing them based on whether their opponent chose the "counter".


Except RPS success only depends on what the opponent chose.

The success of a "retreat/defend/attack/charge" depends on
1. how powerful+fast the enemy fleet is (compared to yours)
2. What your goals are (retreat will probably save more of your ships than charge, but it won't attack/defend any planets)

Those parts can currently work (mechanics of space combat aren't interesting much, but fleet movement is).. and planets are a high value target through the economic part of the game.

I do think that expanding agressive/passive by adding charge+ retreat is a good idea.


Top
 Profile  
 
PostPosted: Sat Sep 14, 2013 6:37 pm 
Offline
Design & Graphics Lead Emeritus
User avatar

Joined: Sat Sep 23, 2006 7:09 pm
Posts: 3858
Location: USA — midwest
My proposal above isn't really supposed to be a RPS system, though i suppose it has some element of that.

It is more of an attempt to be interesting by allowing the player some involvement, and control -- i.e. making significant decisions.

_________________
—• Read this First before posting Game Design Ideas!
—• Design Philosophy

—•— My Ideas, Organized —•— Get an Avatar —•— Acronyms —•—


Top
 Profile  
 
PostPosted: Sat Sep 14, 2013 8:46 pm 
Offline
Space Squid
User avatar

Joined: Wed Aug 14, 2013 1:42 pm
Posts: 68
Location: Johannesburg, South Africa
eleazar wrote:
My proposal above isn't really supposed to be a RPS system, though i suppose it has some element of that.

It is more of an attempt to be interesting by allowing the player some involvement, and control -- i.e. making significant decisions.


I think it's a good call in the absence of some sort of tactical combat solution. It echoes the idea of planetary focuses, while giving some degree of control over the outcome. Far from being a Rock-Paper-Scissors scenario, it's really all about giving a fleet a mission and telling them to adopt a stance that will best help them achieve that mission.

If nothing else is adopted, an option to retreat if resistance is met would be a very useful addition to the current 2 stances, especially as it's often necessary to send ships to unexplored systems with no idea of the dangers there, and because sometimes an enemy is beyond one's scanning level, and one wouldn't want them to put up a fight that they have no hope of winning.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 21 posts ]  Go to page 1, 2  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group