Page 1 of 2

A Team-based RTS

Posted: 05 Apr 2008, 09:57
by Saktoth
Just some musings on mod design. This is going to be an Argh-worthy post so ill include a tl;dr:
An RTS built for team play should take a page out of RPG's or team-based FPS's such as TF and implement a system of complimenting classes.

A response i often get to question of why players play team games over 1v1 is simply because of the teamwork, interaction and group accomplishment that comes with it, rather than that the game is suited to or plays better that way- BA for example is made for 1v1. Most mods do nothing to encourage teamwork, even ones supposedly balanced towards team games. The majority of the time players work in a vacuum ignorant of the actions of their allies, let alone actively attempting to take account of or coordinate them. The assistance of allies is rarely needed or sought after unless to defend against an opponent who is beating you. Although i personally often cut sideways and lend my support to an ally if the enemy opposite me porcs, this is more a strategic decision to strike the weakest point in the enemy line than a truly active effort at teamwork.

So how to make an RTS that actually encourages the style of play most people claim they play spring for? An example can be taken from team-based FPS's, namely a class system. With each player having a unique and essential ability or part to play in the overall effort, teamwork becomes natural. In addition to this, players can find a class that suits their playstyle.

A series of factions, each distinct in abilities, is one way to implement this, ala War Evo (though not a strictly traditional RTS). However, if one faction is specifically strong against another then you get R-P-S play before the game even starts. War Evo solves this by not allowing you to see what class the enemy has chosen. However, then you cannot see what your allies have chosen and get people weighing the team towards one class (everyone picks pyro, or for a horrifically long game, everyone picks engineer) with no ability to switch to a more appropriate one in game. As a side note War Evo is an excellent example of a class-based system fostering strong teamplay in my experience.

Another implementation might be a game with expensive starting factories, each with a distinct set of units (possibly a distinct constructor making distinct structures). The starting resources could be considerable in comparison to the rest of the later economy, meaning that while dual-facs would be possible, it would be more difficult than in some other games. However, you would be able to reclaim and change factories (at the cost of spending the time and a portion of the resouces- E and BT in the traditional sprig setup). This could also be done with hardcodedly limited 'one type per player at a time' factories, which would alleviate some issues with the above.

In both cases unit sharing would probably have to be forbidden.

The specific classes can find rough analogs in team-based FPS. A defensive 'Engineer' faction, a faction with heavy expensive units, another with aggressive spam and another focused on intel and precision strikes (air and stealth units, say). There are a wealth of possibilities here so i wont go into them.

Critique
It is debatable if players truly play team games 'for the teamwork', despite what they might say. As noted above, they make very little effort to actually work together, despite the fact that this can actually benefit them an awful amount, if only to coordinate troop movements. An extreme example of teamwork might be starting all commanders next to eachother, pooling the starting metal into a single factory, then blowing up the commanders one by one and taking their metal as the starting resources run out. This would probably be an incredibly powerful teamwork strategy but its not something you'd ever see outside of a very, very rare Greenfields game.

I would speculate that mostly people play team games due to diminished responsibility. This is especially appealing to casual players (I wont say 'those with fragile egos or of lesser skill' because that would be mean...). Team games are seen as 'less stressful' because of the perception that less relies on that player. There may be other factors in the way it alters the gameplay, as team games usually allow for more porcing and simbasing, which a lot of players seem to enjoy. There may also be a sense of disparity as 1v1 players are usually of a higher skill level, so most players are unable to find an even matchup.

So, it may not be the case that an RTS built specifically for team play would be popular with the proportion of the playerbase that claims they enjoy team games due to the teamwork- but i think there is little doubt that it would find a niche if well executed.

No, im not planning on making this game any time soon although there are lessons to be learned here for CA (see if you can connect the dots to some of my other design philosophies for CA).

Re: A Team-based RTS

Posted: 05 Apr 2008, 10:24
by KDR_11k
Ever played World in Conflict? It's a good game but you can forget most matches as few if any players ever work together, it's mostly just everyone doing what he feels like. In WiC cooperation is enforced HARD, if your air, armor and support players don't cooperate you get nothing done. Especially the supports are trouble, they are needed as antiair but often just go arty and happily shoot random points while their allies get torn apart by airstrikes.

BTW, WiC allows joining midgame (not sure you'll need that) and changing your class freely (so if the situation calls for more tanks you can get more tanks).

You think too conventional it seems, expensive factories are a TA-ism that means attacking the enemy fac has very high priority and that you cannot really move your production forward. Spring has Lua now, use it! If you want classes let players declare them in a class selection window and get the matching units, if they want another let them change (but kill all their class-specific units).

I don't make team-based mods because I dont think you'll find enough people to play them with. The closest I've made to one might be THIS with its perk system that would allow team members to pick somewhat different roles (e.g. scout player with advanced gravidar, cap ship player using dark matter tech and mag shields, ...)

I think the diminished responsbility is true though it's silly, in a 1v1 you have no responsibilities, in a team game you have the responsibility to be a good part of your team and not fail your teammates. Noone cares if you fuck up in a 1v1, everyone does if you do in a team game and you failing is enough to lose the game, team or no team. Then again, many people don't even seem to notice they fucked up, see e.g. Orakios thread about how his ALLIES failed him when HE didn't claim his mex spots and built a three layered DT wall.

Re: A Team-based RTS

Posted: 05 Apr 2008, 15:18
by Saktoth
KDR_11k wrote:Ever played World in Conflict?
I havent, but it seems more of an RTT game than a traditional RTS. Also seems to follow some other FPS designs (such as the victory scheme).
You think too conventional it seems
True enough, i probably delved too much into implementation there when this is mostly a discussion of broader theory, though i think the engine should generally be used as much as possible, especially given how crashy lua can be.
I don't make team-based mods because I dont think you'll find enough people to play them with.
Well yeah, see my critique. I dont think most players have that much interest in genuine cooperative play.
I think the diminished responsbility is true though it's silly, in a 1v1 you have no responsibilities, in a team game you have the responsibility to be a good part of your team and not fail your teammates.
Its like you are reading my mind. This is what im saying to people all the time, though i didnt go into it because that was kinda tangential to begin with.

Re: A Team-based RTS

Posted: 05 Apr 2008, 15:49
by KDR_11k
Victory conditions that go beyond "kill everything" are on the rise as they provide more interesting gameplay. Basically in a kill-to-win game the combat dynamics are between the two bases, pretty much one line plus a few small skirmishes over resource spots. Those bases must be fortified because otherwise you take damage and the functioning of your base is inhibited, making you lose even faster. In a VC game an attack on your base is separate from a bid to win and if you're behind in the victory condition your ability to fight isn't diminished, preventing a slippery slope. In a KTW game you get crippled more and more until it's clear that you cannot recover but then the game isn't over, you still have to play it to the end or be counted as a ragequitter. With a VC a game is only really lost when you have no chance to cancel the enemy victory which is usually only for a few seconds or a minute before the victory is declared, keeping it more interesting until the end of the match.

Sure, WiC was mostly tactics but IMO few RT"S"es actually go higher.

Lua is as crashy as you write it. THIS is very stable AFAIK despite having handed its entire resource system to Lua.

The hard to find players thing was about non-BA mods in general, I rarely see games get larger than 1v1 or 2v2 for those simply because noone wants to play. If you think you can do better feel free to try it but don't think you're the only one who did.

IIRC Draco made a team-based mod once.

Re: A Team-based RTS

Posted: 05 Apr 2008, 16:59
by Saktoth
Thats quite interesting what you have to say on victory conditions, but i think this has had a very bad track record in RTS. I think most people find such victories unsatisfying- that it is a cheap way to win or that the game isnt really 'over' until you kill the enemy. AoE 2 for example allowed Wonder (build and countdown) and Relic (collect and countdown) victories in the default game mode, but people almost always play conquest to disable them. I've heard numerous times that OTA and Supcom are both usually played com ends though- which is a sort of Assassination game mode. Spring team player, though, is almost never played com ends.

I think the aversion to other victory conditions has something to do with the economic buildup and base-building aspects. There is a lot of time and energy devoted to defenses and economy, for a buildup towards the eventual goal of destroying the enemy and his base (stomping through lovingly placed solar panels is a part of the fun). To have that truncated can make it seem like half the game wasnt finished, and half the effort never bore any fruit. WiC for example seems to eschew economic buildup and base building entirely. The 'Buildup to victory' is through the acquisition of the victory conditions. Hrrm, this has given me a lot to think about.
The hard to find players thing was about non-BA mods in general.
My comments on this were also in response to your comments on WiC, i should have quoted that part too:
It's a good game but you can forget most matches as few if any players ever work together, it's mostly just everyone doing what he feels like.
If you think you can do better feel free to try it but don't think you're the only one who did.
As i said ive actually no intention to. This is just spitballing. Im not claiming anything new here either: these are ideas that have been used for years in RPG's (even P&P) and FPS's.

Re: A Team-based RTS

Posted: 05 Apr 2008, 17:26
by KDR_11k
Well, wonders and relics do seem fairly silly as a VC to me anyway. A wonder is pretty much an economic victory (porc win) and the mobility of relics didn't exactly make them strategic locations IIRC. Dawn of War and Company of Heroes have strategic locations that AFAIK get used, forcing you to get out of your base and take those damn spots if you want to survive (instead of just having an economic disadvantage and getting crushed by a bigger army after a while)

Com ends is flawed in that it means one mistake is instant death, with a point-based VC a mistake would just mean losing a point or two that you can recapture if you don't keep messing up. Also com ends is flawed for team games since the victory condition is per-player instead of per-team, if a player loses his com then his contributions to the team explode which means the team's base is no longer the team's effort but several individual efforts with different weakpoints each.

Something to think about: SimBase and Fibre have an economic and capture-and-hold victory condition respectively that doesn't win the game instantly but basically consists of an unstoppable superweapon.

Re: A Team-based RTS

Posted: 05 Apr 2008, 20:41
by SpikedHelmet
We're trying something vaguely similar with S:44, except more reductionist and almost accidentally.

There are several very distinct unit types with vastly different combat capabilities. It is similar to the "Vehicle-Kbot-Air-Hover" dynamic, except more extreme. Infantry, Light Vehicles, Towed Guns, Tanks, and Self-Propelled Guns, each with their own (relative) usefullness and task.

Nemo came up with an idea a couple weeks ago for something nearly exactly that you're describing, focusing S:44 on teamplay where the access for any one player to these unit types is limited to two or three, making for games where one player may choose artillery, another infantry, another combat tanks, etc. I thought it a severe limitation but I'm warming up to the idea. But the basic premise, where each of these unit types fulfills certain roles, all of which are useful in certain conditions, lends to better teamwork; I had a game where I focused on scouting enemy positions and directing artillery fire, while my ally (we were com-sharing) concentrated on a simeltaneous attack with tank forces. It worked amazingly.

Of course, the fact that S:44 is rather complex helps (as opposed to hindering individual control). Tanks and self-propelled guns need to be positioned so their frontal armour is facing the enemy; artillery and anti-tank guns need to be set up and deployed, and then directed to bombard enemy positions out of LOS; scout infantry needs to be micromanaged to avoid detection; etc. Too much for many people to handle alone.

Maybe Comm Ends needs to be played more? I really like Comm Ends.

Re: A Team-based RTS

Posted: 05 Apr 2008, 23:40
by zwzsg
An RTS built for team play should take a page out of RPG's or team-based FPS's such as TF and implement a system of complimenting classes.
There was one made: Allianz Arena

Re: A Team-based RTS

Posted: 06 Apr 2008, 17:15
by SpikedHelmet
But Zwz, that guy was French... you know that doesn't count...

Re: A Team-based RTS

Posted: 06 Apr 2008, 19:42
by rattle
Am I the only one who read the topic as A-Team based RTS? :P

Re: A Team-based RTS

Posted: 06 Apr 2008, 22:37
by Forboding Angel
I PITY THE FOOL! 8)

Re: A Team-based RTS

Posted: 08 Apr 2008, 05:16
by Neddie
A number of players pool resources via that team example in BA these days, unfortunately, the groups who do so are habitual stackers, and they would never play a group as cohesive as their own.

Re: A Team-based RTS

Posted: 08 Apr 2008, 07:30
by Sleksa
neddiedrow wrote:A number of players pool resources via that team example in BA these days, unfortunately, the groups who do so are habitual stackers, and they would never play a group as cohesive as their own.


There are no other groups ;]

Re: A Team-based RTS

Posted: 08 Apr 2008, 23:38
by Neddie
Yes, and [K] would be no threat to [WarC] at all. Of course, we will never know, since they banned WarC.

Re: A Team-based RTS

Posted: 10 Apr 2008, 11:20
by Sleksa
[DRF] Were the coolest, but they disappeared :[

Re: A Team-based RTS

Posted: 11 Apr 2008, 11:10
by TheMightyOne
Saktoth wrote: It is debatable if players truly play team games 'for the teamwork', despite what they might say. As noted above, they make very little effort to actually work together, despite the fact that this can actually benefit them an awful amount, if only to coordinate troop movements.
If you want to make ppl play as a team, give them rewards for the stuff they should do as their class. I'm thinking on the "Enemy Territory" system where every class got XP not only for killing enemies but for doing stuff they are ment to do. Medic gets XP for giving medpacks and "reviving" bad wounded. Engineer got XP for repairing tanks and bridges. This way you were forced to help ppl, heal them, give them ammunition since it brought you better weapons, more HP and other stuff so you WANTED to play as a team. As a medic you're running through the map searching for wounded allies, once you find one both become happy coz they get what they want.
To make ppl play as a team, make them want to play as a team.

Re: A Team-based RTS

Posted: 11 Apr 2008, 14:51
by KDR_11k
I've seen the result in world in Conflict, players running HUGE hordes of repair tanks around that would just go near enemies in order to get hit and then repair each other for points.

Re: A Team-based RTS

Posted: 11 Apr 2008, 15:55
by TheMightyOne
KDR_11k wrote:I've seen the result in world in Conflict, players running HUGE hordes of repair tanks around that would just go near enemies in order to get hit and then repair each other for points.
hahaha

well, this is no proof that the idea is bad ^^

its just a proof for the designers being not able to think about the consequences of their own ideas.

but this might be an idea for one of the classes :)
a support player, with units without or maybe with few light armed units whose main task is to keep the army of his allies healed at the battlefield.

Re: A Team-based RTS

Posted: 13 Apr 2008, 00:10
by Ixoran
KDR_11k wrote:I've seen the result in world in Conflict, players running HUGE hordes of repair tanks around that would just go near enemies in order to get hit and then repair each other for points.
Because it's so hard to only give them EXP for healing non-healing units, and unit's that they don't control?

Also, just because people farm, doesn't mean that is the way the game was intended to be played...

But hey...

Re: A Team-based RTS

Posted: 13 Apr 2008, 03:57
by Zpock
If you make the game to be about collecting points then people will do just that.