Game Enders

While posting in another thread, I stumbled upon an elegant way of explaining something. Game Enders, by there very name, end games. Despite the fact that I do want games to end, I think they shouldn't exist, and I'm here to explain why.

First let's define what I mean by 'Game Ender'. A 'Game Ender' is a unit (or combination of units) which, if successful built, nearly always mean the builder wins the game. This is a bad thing, because it's a very simple strategy, one that does not promote an interesting or dynamic game. It also, ironically, encourages turtling (please note that I have a defensive playstyle, so this alone isn't bad). But it encourages the bad kind of turtling, because guess what you never have to attack. You just build your super unit, and then you win. You never need to learn offensive strategy, and your opponent never gets to do anything but madly rush to kill you before you build 'doomsday X'.

All of this boils down to one simple concept. There should be units in situations which cause me to lose, but 'don't let them build it' should never be the only counter-strategy.

Let's talk TA. TA had a doomsday unit, in fact it had three. Nukes could be spammed until you won, but doing so was not cost effective, as your enemy could spam counter nukes at about a tenth of the price and build-time. The Buzzsaw, however, was a little more of a 'game ender'. It was somewhat limited by terrain, which would create safe zones, and by the fact that shields didn't exist and thus it could be bombed(and was easy to find, too). The Krogoth is a lovely example of a super unit. Krogoths could be incredibly deadly, and they didn't have a real 'counter' (except nuking the hell out of it). At the same time, due to how TA worked, they could be swarmed under by conventional units. Their balance was such that they literally were only available to one side. Yet players still played as ARM, and won.

I believe this is the quote that you posted in another thread that got you thinking about this whole "Game Enders" thing,
correct?

While I agree with you that One Unit should never be able to end the game on its own I disagree with your "combination of units" postulation.

I also disagree that if you allow your enemy to build a very powerful combination of units, that you should be able to dig your way out of your own grave... that you dug for yourself by allowing him to gain critical mass.

If you let him gain critical mass you should lose.

Ah, but there is the key, I didn't say 'he horribly outnumbers me' I said, 'he builds a combination of units'. If I have roughly the same amount of mass(in terms of units) in play as he does, the game shouldn't be a forgone conclusion. In other words I'm not talking about him building a lot of units, I'm talking about him building specific units.

In short, if my economy is in the roughly the same shape, then I should be able(assuming skill) to fend off anything he can throw at me. In other words, there should not be a bomber snipe swarm that cannot be stopped by enough AA. I shouldn't lose because he built 500 bombers, I should only lose because I didn't have enough AA, or I left my Commander vulnerable.

Objectively defining Game Enders

Right: it's all well and good to talk theory and hypothesis and even personal conjecture and experience, but data rules all when it comes to balancing decisions.

RTS games, for a while, have been very good at logging gameplay data for a) the stats screen at the end of a match b) unit veterancy stuff and c) persistent ranking for competitive multiplayer purposes. I'm going to assume PA definitely logs all this stuff due to the replay function.

So: game enders ought to be quite visible in after-action data reports, seen as:
[*]spikes of unit deaths
[*]specific units killing the commander
[*]a specific single unit killing hundreds of units
[*]that kind of thing

and so Uber can tell if a unit is overpowered by looking at the reported data, performing a complex enough analysis of it, and seeing how many games are ended by eg a nuke killing the commander. In other games obviously this is used to see if a faction were OP compared to another faction. And of course other things beyond the actual unit stats play into this, like the metagame, but in general it's very powerful and useful.

In your opinion, what percentage of victories-by-superweapon is acceptable? What was that number in other games?

A very good question, thank you.

I can't give you the answer in terms of math, because it all depends on what you mean by 'victory-by-superweapon'. Is an asteroid a super-weapon? Well, it can countered by missiles, and it can be simply avoided. In short, one can assume that we will see more than one asteroid used in a single game, often without securing a kill. By my logic then, it's not a superweapon.

So I guess . . . none? I feel as if I'm avoiding your question, and I'm sorry about that. Obviously powerful units will turn the tide of battle, that's just what powerful units do. My point is there should never be a unit that, when I see it, I know the game is over, no matter what condition I'm in.

That is quite the leap in logic. I'm not sure I agree with your definition of a "Superweapon"

It heavily depends on how easy scouting is.

Lets say scouting (especially in late game where game enders usually come in play in form of super weapons) is as easy as in old TA or SupCom.

Then, anything that kills me but which I could have defended against if scouted in time is ok in my book.

For example, if my enemy builds up a huge stash of nukes in some faraway corner an hour into the game and I never scout it nor build nuke defenses. She then nukes me and kills me. Totally fine, my fault for not scouting.

On the other hand, I scout my enemy building up a huge mass of bombers. If I can't stop them because neither AA nor interceptors are powerful enough to stop them and they're guaranteed a snipe on my commander. Bad.


In other words, I'm fine with exceedingly strong game enders, IF they can be stopped by appropriate preparation and scouting it in time.


So yes, if I let my enemy build a buzzsaw in range of some of my bases, totally my fault. Building a buzzsaw is a slow and expensive process, with a bit of preparation its unlikely to be suprised by it.

I could have built a big bertha (which is both cheaper and faster to build) and taken it out for about 1/8th of the cost of the buzzsaw without even leaving my base. I could have attacked it head on. I could have bombed it or just exploited the fact that my enemy is pouring tons of ressources into a single building and attacked elsewhere. Etc.

Oh look, a Backers' Only forum. Maybe it will be less terrible than the main forum (haha, no).

If your opponent has 500 bombers and your economy is in the same shape as his, you deserve to lose. Why? Because he has devoted a huge portion of his economy (and unit cap) into building a purely offensive air-to-ground unit and you did not take this opportunity to wreck his base. If he goes for a purely offensive strategy and his economy is still as good as yours, it's because you got outplayed. Amassing a zillion bombers leaves you very weak in most areas, and a good player would use that opportunity to take map control away from the bomber-spammer.

And if the game's gone on so long that 500 bombers does not represent huge portion of a player's economy, you probably still deserve to lose because at that point it's better that someone wins, and "being the first guy to attack with 500 bombers" is as good a sudden death condition as "being the first guy to build a Vulcan/Mavor".

So by your logic you'd be fine with a timelimit on games, after which the player with least mass(in terms of units) automatically loses. Doesn't matter if it's by 2%, it just happens.

Sudden death doesn't belong in a strategic game. Two equally matched players should be able to deadlock for quite a while if neither screws up. We have multiple maps, huge battlefields. I've played Total Annihilation games with a 5000 unit cap, and I've bloody well reached the cap. You could easily have 10 players in a game, so yes, one guy could spend all his time building bombers, fueled by his 4 allies. His 4 allies might be losing badly, but they could hold long enough till the death fleet sails. What's that, 'you can counter that' you say. You're right, I can, by getting 500 bombers before he does. Let's call the game 'bomber spam', then, shall we?



You should simplify your post to "I like game enders, so I disagree with your premise." That's fine, you can disagree, that's fair. But don't even start to use the word 'should'. Strategy is, as you said, about out-playing your opponent. But out-playing your opponent does not consist of racing toward an arbitrary win-condition, be it bombers, missiles, or asteroids.

I will invoke chess, if you don't mind. Notice the lack of superweapons? Notice that the only way to win is to actually win, not just fail to lose long enough?

Fair enough, I'm fine using a different term for my definition. "Game Ender," perhaps? Call it what you wish, it's the concept that matters.

Game Ender is more appropriate.

Superweapons can end games (what happens when your Commander is hit in the head with an asteroid), but they are not necessarily "Game Ender" units.

Game enders are fine as long as they are significantly less cost effective than unit spam, take a long time to build and are very difficult to produce within the normal length of a game (30-45min?). If you don't have them you just end up with stalemates which are not fun, because if neither opponent has any chance at victory without the other conceding then pro matches are going to just be a war of attrition, you need to have something to be able to change up the gameplay and shift the balance of power otherwise games will end up monotonous and get boring VERY quickly.

Notice that chess has stalemates, and isn't particularly entertaining to watch. Since Uber want PA to be a potential esport, stalemates are going to be something they want to avoid, and game enders solve that as well as usually being crowd pleasers too. Game enders build up a good sense of suspense and anticipation that drives the excitement of watching a match with situations like: "will the buzzsaw be spotted before it's built?" or "is he going to be able to kill the galactic colossus before it starts tearing apart his base?".

IMO game enders make the games more interesting to both watch and play, sure it can suck getting beaten by one but as long as they are balanced such that they are only really useful for breaking stalemates after a very long game, have easier to build counters, and are not something that you can just to rush to in order to win then they really shouldn't be a problem as by the time anyone could build them you should be far enough into the game that you can do something about it if you spot it. And even if you don't spot it soon enough you should have your base spread out enough that you can deal with it well before you are completely out of the game.

At the very least game enders are a lot better than players just pointlessly throwing units at each other until someone gives up.

Nice strawman there. A game ender should never represent 2% of a player's economy. Try attacking points I actually made.

Wow, you must have only played with bad players who don't even try to win by any method but sniping. If one player is doing nothing but building bombers, that team is down 20% on eco from the start. If the teams are 'equally matched' as you say, that's a loss right there. The counter to some idiot amassing 500 bombers in his base isn't to get 500 bombers before him, it's to blow up all his stuff with your vastly superior land army.

EDIT: Also the board game community would tear you apart for saying sudden death doesn't belong in strategy games. Pretty much every grand strategy board game has a hard turn limit. Also most 4X games end the moment one player satisfies one of their victory conditions.

No, because I use logic to support my arguments. I don't actually like game enders, because they're kind-of anticlimactic, but their existence is necessary to ensure large games actually end for reasons other than one player quitting because they have work in the morning. You're also misrepresenting the point of the game ender. If you race towards one, you lose because they're so expensive that doing so puts you at a crushing economic disadvantage. Nobody starts the game thinking "I'm gonna rush a Mavor" and wins. They exist as an option so players know they can't turtle in their base forever because sooner or later the shells are going to start raining down and none of their layered defences are worth anything any more. Any player that wins with a game ender probably would have won without it, but it would have taken longer and been a waste of time for both players.

Funny that you mention chess. Chess actually does have a timelimit. The threefold repetition rule and the "no pawn advanced, no piece taken for fifty moves" rule effectively limit how long two players can be deadlocked for. And, more importantly, the number of pieces on a chess board can only decrease so in fact, chess already includes a sudden death condition, because eventually only one player will have enough materiel to force a win. SupCom and TA have economies that, in general, only increase, so Chess's lack of game enders isn't relevant.

Sudden death is arguably in a lot of strategic games this century, and earlier. Any game that has "loss of specific unit X" (e.g. commander) as a victory condition is a sudden death scenario.

I've always thought of myself as an itinerant turtler. I build bases and defenses up the wazoo in an effort to protect my investment. But srsly, maxing a 5000 unit cap in TA? It sounds like you and your fellows actively avoid offense where at all possible. Compared to that, I don't turtle, I'm an eco spammer :shock:

How many krogoths could I have built during the time it took you to max out your unit cap, and how many krogoths could you have defended against during that time? What were you actually building?

Actually, outplaying your opponent is always about getting to the arbitrary win condition first (blow up your opponent's commander). If your opponent is ahead of you in some strategy to achieve that win condition, you will need to scupper their plans somehow (i.e. blow up their nascent asteroid base and take it over yourself), otherwise you have indeed been outplayed.

Chess: sudden death condition (arbitrary win condition) = capture the opposing King, superweapon = Queen

PA: sudden death condition (arbitrary win condition) = destroy enemy commander, superweapon = asteroid

Do you agree? If not, what do you disagree with about my interpretation of chess? I'll stand by my assertion that checkmate (since games are never played to capture) is a sudden death condition; if that happens you have definitively lost, and it does not finalise until the last move (until your opponent makes that last move to turn check into mate, you can still respond). You do not have to capture any other pieces in the game, just the King.

I'll also stand by my assertion that the Queen in chess is a "superweapon"; she has almost all the movement options in the game, is conventionally considered the most powerful piece, and you can make more of them using an in-built (if convoluted) mechanic. The fact that promoting a pawn to anything other than a queen is considered "underpromotion" should be indicative on it's own.

You can most decidedly lose a chess game by allowing your opponent to promote 2 pawns (make more superweapons) while you "fail to lose" in another corner of the board. Quite analogous to asteroids in PA, in fact, or krogoths / buzzsaws in TA.

So far we know of several game ending units and/or mechanics:

- Kill the Commander. Win.
- Spread your Commander seed throughout the galaxy. Win.
- Nuke nuke nuke nuke nuke (EVERYBODY gets a nuke!)
- Smashing asteroids into worlds.


There's nothing wrong with those, right? If the game needs more, then more can be added. There's no need to shoehorn more things in when there are already quick and effective kill shots to end the game.

I believe in planetary annihilation, what we think are game enders now will just be planet enders / posession changers

Say we had a nuke that could destroy an entire planets surface, wiping it clean of all life. Not a game ender coz youre probably gonna have more than 1 planet or both be on the same one

1) Point planet-killing nuke at planet containing enemy Commander.
1a) If your Commander is also on this planet, ship him to some other planet.
2) Fire.
3) Game is now over.

Sure, you could be a moron and fire your incredibly expensive sure-kill weapon at a planet not containing the enemy Commander, but why would you want to do that?

Maybe because that planet contains the enemies main production facilities and finding the enemy commander isn't as trivially easy as you think it will be?

Not to mention that Nukes killing off all life on a planet isn't a concern for Robots.

I think that it is too early for such a discussion. Bet that when the alpha goes live then there will be a percentage of players who will try to find the metaphorical "iWin" button. And when they find it then it will be up to Uber to investigate balance.

However when I think about this: a map that has more than one body to fight over then the efficiency of whichever gameender is hindered by the fact that you may not be on that planet/moon/asteroid anymore.

On the other hand: it could get to another superannoyance when players play a sort of "Hide'n'seek". Although I faintly recall that Uber mentioned asteroiding a planet/moon enough should effectively remove the target planet/moon from the game.

So in summary: too early to say IMO.

I hope it is way harder to kill a planet, else the game will turn into 'rush to space, weaponize asteroid, scout for enemy commander, win', which would completely de-emphasize all other forms of combat past the initial economic skirmishing.