长文编译：RTS游戏系统设计的平衡问题

原作者： Brandon Casteel 译者：Vivian Xue

从某种意义上说，战略游戏是一种对玩家进行限制的游戏。玩家在特定时间内花费有限的资源购买单元和建筑、在有限的时间内应对敌方的攻击和战术、管理分配有限的注意力、试图让有限的供应单位发挥出最大的效益，各个单位的功能也具有限制性。

在战略游戏中，玩家们可以通过诸多技巧打破一定的局限性：快速提升经济来尽快建立（或重建）一支大规模的军队（或者用一支军队进行快速打击），在战斗中实现效率最大化，使每个单位对敌人造成尽可能大的伤害，迫使敌人多次消耗资源而陷入短缺状态，这样在他们花费资源重建时，你就可以把他们远远地甩在后面。即时战略游戏（RTS）的核心是纯粹、粗暴的效率，主要目标通常就是尽快对敌方造成尽量大的伤害并将自身的损失和成本降到最低。

这点没错，无论是《星际争霸》（StarCraft）、《帝国时代》（Age of Empires）、《命令与征服》(Command and Conquer)或是设想中的小马宝莉RTS, 这就是这类游戏的玩法。

但是随着这一类型游戏的逐渐成熟，设计者们已经尝试应用了不同的游戏效率模式。你会发现，简单粗暴的输出不过如此而已，这一点我们后面会着重讨论。

那么让我们回归正题，这篇文章的目的是什么？我想通过它说明两点。第一，最好的战略游戏应该包含大约3到4种具有一定独立性的系统；第二，这些系统中至少有一些能够让玩家获得暂时的优势或者挽回劣势。以及游戏中的弹药单位要如何分配才能协助实现这些。

拒绝二元化：多层次让游戏保持趣味

尽管《齿与尾》（Tooth and Tail）有很多值得推荐的地方，但该游戏的战斗结果趋于二元化，这导致游戏在几场快速爆发的战役中立刻胜负分明。《灰蛊》（Grey Goo）也存在同样的问题，游戏中的资源效率是决定比赛结果的首要因素。

从根本上说，我认为战略游戏的设计应实现三种主要并且有点对立的模式之间的微妙平衡。首先，为了提升玩家的技能熟练程度，优质RTS游戏中的每一个系统提供的增量效益应该是递增的：玩家在某一游戏系统上花费的精力越多，受益程度也随之增长（即使奖励有可能会减少）。比如《星际争霸 II》中的采矿系统中，了解扩张的时机或是其它游戏初期能够提高采矿收益的小技巧、部署机枪兵对抗虫族，或者是《命运与征服》游戏中战略性地售卖建造物。大多涉及到细节技巧的战斗属于这一类型，但是经济管理的作用在此也十分重要。

第二，最有趣的战略游戏应该呈现多样化的胜负状态。如果每场交战都是在一方全军覆没而另一方仍兵力充足中结束的话，那么比赛结果在几场交战后就很明显了：这是《星际争霸 2》中的常见情况，一场胜利的袭击经常能决定整个比赛的结果。《齿与尾》的比赛机制甚至更二元化，尤其在比赛的初期，如果某个玩家在生产力和收入上已经取得了稳定的优势，那么他的对手们基本上没机会再与其抗衡。

尽管《魔兽争霸3》（WarCraft 3）中的战役同样具有极强的决定性，但游戏中的其它支撑系统比如工人保护机制、英雄分级和魔法泉，以及发起战斗的难度使战斗的结局十分多样。

让战斗在一种部分成功或部分不成功的状态下结束，使玩家在一系列的战斗和敌方入侵中不至于落的一败涂地，这样可以让游戏继续进行下去并让那些因为一时草率或分神而暂时失利的玩家能够挽回劣势，同时让那些表现得更好的玩家继续保持优势。例如，在游戏《英雄连》（Company of Heroes）和《战争黎明》（Dawn of War）中，玩家在守卫或攻占土地的过程中失败后仍能将队伍撤回到基地中以保存战斗力。

或者是在《魔兽争霸3》中，生命泉水足够大的情况下，玩家可以在周围通过眩晕或身体阻挡来对敌方进行持续干扰，从而削弱对方的兵力。或者还是《魔兽争霸3》中，每个种族都可以选择自己的基地防御系统：兽人的地洞，暗夜精灵的远古守护者，这使得玩家在军队外出而基地面临攻打时能够把损失降到最低，或通过传送门回去保护他们的建筑和单位。

这类系统具有极大差异：比如在Relic公司游戏的撤退系统中，了解撤退的方向和时机，以及队伍逃回基地的路线很重要：因为你的对手可能正带着一支精兵埋伏在途中准备把你打个片甲不留，这使得游戏仍然需要一定的技巧。

关注游戏技巧

在《战争黎明》(Dawn of War)和《帝国时代》（Age of Empires）这类成功的游戏中，夺取胜利的玩法和效率机制多种多样，而不像《星际争霸》那样一定要摧毁对手的一切。额外获胜方式的存在激励了落后的玩家继续尝试，寻找其他的成功途径。

为什么这很重要呢？在战略游戏中，表现得更好的玩家理应获胜，那么为什么要延长其他玩家失败的痛苦呢？为什么要给相对差劲的玩家更多机会来击败（或纯靠运气击败）他们优秀的对手？这得回归到战略游戏的本质和乐趣上，它可以归结为为人们创造有趣的竞争体验，并亲自参与到这些体验中。

坦白地说，我的观点是基于游戏给我的整体感觉，以及我所认为的游戏为了保持战斗的乐趣而做的努力；我知道了如何利用机制来实现更好的战略游戏。请理解，这一切都是出于我对游戏设计的个人偏好，我对“游戏的技巧”甚至“获胜”的定义可能与你认为的大多数战略游戏中的不太一样。我所支持是那些能让玩家们在即使在落后的情况下，还觉得自己可以继续参与下去的游戏设计。我支持赋予玩家们自主性，当然我也承认我的上述观点存在一定缺陷。

与所有竞争性游戏一样，战略游戏中，那些具有更好的策略以及执行力的玩家应当赢得比赛。 但是想想《小世界》（游戏邦注SmallWorld）或者桌游《战锤40K》（Warhammer 40,000），只有在满足游戏结束条件的前提下才会产生胜者，或者《战争的黎明》中Relic式胜利，或者在《帝国时代》中玩家可以通过建造Wonder在经济和防守上称霸，这使得输赢不再单纯取决于任何一方的总体兵力。

在桌游《小世界》中，玩家通过控制各个种族占领土地获取金币，最终的输赢取决于游戏过程中你所控制所有种族产生的资金总量。虽然我不太喜欢桌游，但是这个游戏的确带给了我很大的乐趣。

技巧性的胜利归根还是技巧性的胜利。在基于技巧的竞争游戏中，仍然还会出现压倒性的胜利，这很正常。然而我所认为的战略游戏中的有趣之处是让优秀的玩家保持优势地位，或者让玩家获得胜利的同时也能让失败的一方在游戏中发挥出最大的能力。同样令我感兴趣的是那些能够鼓励玩家通过多种方式掌握或者展示自己的技巧以获得成功的游戏系统。

多样的获胜方式

尽管我很欣赏《钢铁之师》（Steel Division）中的许多机制，但该游戏的胜利仍然高度取决于你能否让你的军队单位存活下来，如果你的对手在战役中获得了领先优势，你很难再有翻盘的机会。

我坚持认为，竞争性游戏中当所有玩家都觉得他们可以继续游戏并且有机会获得胜利时，游戏才是最有趣的。当然游戏过程中肯定会有那么一刻，某一个玩家或者某一个团队会感到成功无望，但是我们可以把这个时刻尽量推后。一旦玩家觉得他们继续玩下去毫无意义，并且没有任何选择或控制权时，他们就会感到沮丧。

那么这些与战略游戏中的弹药、魔法、士气以及其它类似系统有什么关联呢？我很快就会谈到这点。

我想通过以上阐明的是，一个游戏在在达到内稳（homeostasis）或者平衡（equilibrium）的状态（类似于大部分RTS游戏中期）时能给玩家带来最大的乐趣，若将这一状态转向胜利，大多数时候，应在游戏主宰者获得多次压倒性胜利的情况下，并需要以下系统的辅助：

—多种影响胜利的系统，这些系统在某种程度上彼此独立。这样即使玩家们在某一方面的受到压制，他们仍然有获胜的机会。一个简单的例子是在《星际争霸2》游戏的早期到中期，玩家可以对经济扩张、军队规模，攻击方式或技术进行不同的组合来试图压制对手。
—与玩家技术分离的游戏获胜条件。例如，在《星际争霸》中，玩家通过摧毁敌方还击的力量获胜，而在《战争黎明2》则采用了争夺胜利点的模式，当对手的胜利点为0时玩家获胜。
—创建能够让玩家弥补或收回损失的系统，或者让战斗结果不过于两极化。《战争黎明》和《英雄连》中的撤退机制就是一个例子。

下面到了我们谈效率问题的时候了。在战略游戏中，成功的向量（vector）越少，这些向量的效率对成功的影响力就越大。例如，在《星际争霸2》中：矿石和瓦斯是玩家升级和扩张的主要资源，对于玩家扩大收入也是至关重要。但是不同类型的工厂的数量和位置也会影响玩家生产单位的能力，并且某一类型的工厂数量过多或是过少都会导致问题。同样地，尽管《星际争霸》每次的升级变动都不大，但有时额外的防御或伤害可能会为你增加优势：总体而言，游戏为玩家提供了很多可供管理的有趣元素，使玩家们能够制定和施展不同的策略……虽然一些玩家（特别是那些白金段及以下的玩家）通常会完全无视这一点，这些玩家以实现大规模的杀伤为目标，在他们眼中，胜利与否在于谁可以更频繁地杀死另一方的军队并重建自己的军队。

尽管《英雄连2》中的大部分战斗仍然依赖于输出伤害，但诸如机枪压制、战车损坏状态、烟雾弹和互相克制的兵种等系统为玩家提供了许多击溃对手的方法，另外游戏中战略点分布间隔大，使玩家有机会在对手力量微弱之时对其进行打击。

因此除非在高水平的比赛中，这些因素的作用在强大的经济和DPS压制下是很微弱的。但是很明显，能力、部署以及其它战术和战略考虑因素（力场，观察员，根基，暴风雪等）确实能在战斗中发挥作用。

这样分析来看，《星际争霸2》比赛中也有很多因素是可以避免或忽略的，除非在高水平的赛局中。它是一个非常灵活和强大的游戏系统。这也是为什么它如此受欢迎的原因。

对比之下，《齿与尾》游戏中只有一种资源：食物。它可用于建造建筑、单位、农场（经济）和占领新基地/领土。那么最终游戏中的一切都取决于食物。这是游戏的关键。尽管我认为这种极度的专一性值得赞扬，但在多人游戏中，它可能会让游戏变得很无情。当你玩《齿与尾》时，特别是游戏前期还没能扩张的时候，失去任何东西都可能让你陷入《星际争霸2》虫族与虫族对战的那种失控的局势中：那就是，你将永远无法在经济上赶超你的对手。从统计学中的经济增长曲线来看，若你把食物“浪费”在重建单位上，相对的你的对手就有更多的资源建造单位，并且这一优势会像滚雪球一般扩大。

与《星际争霸 2》相比，《齿与尾》更加线性化，因为玩家的可控因素无非是战斗技术、单位选择/部署（有趣的是你可以在较低级别或是较高级别/更强大的单位之间进行选择，但这与你的收入密切相关）和收入，再者由于游戏中玩家只能控制一个领导单位，因此这些因素发挥的作用也有限。

让我进一步说明这点，《英雄连 2》为玩家提供了多样的可操控因素：人力收入大部分是固定的，根据玩家当前拥有的单位数量增加或减少。油料是稀缺资源，并且对表现差的战车采取资源惩罚。弹药一开始很稀少，但随着游戏中后期军队规模的扩大变得充足，而且这时候利用Napalm（凝固汽油弹）造成大范围伤害相对会更容易一点。除此之外，游戏还提供了一个真正的动态步兵战斗系统：包括掩护、士气和机枪压制、烟雾弹和手榴弹以及间接火力武器；游戏中轻型车辆能与间接火力、MG武器以及常规步兵能配合形成不同的作战效果；同样的还有较重型的车辆、AT枪以及游戏后期的超重型坦克。随着游戏进程的发展，这类交互作用的数量和差别会不断增加，玩家们通过提高弹药收入可以更好地掌控局势。

在《英雄连 2》中玩家拥有3个资源（如果你把vp当成一种资源的话就是4个），每种资源都有各自的用途，玩家可以用各种方式使用这些资源；游戏中的单位互相牵制，不过也存在一些回旋的余地：譬如游戏中AT枪是打击坦克的主要武器，但如果玩家们在早期游戏中很难获得AT枪的话，也可以借助MG武器也达到一定的反坦克效果。（显然单位选择是《英雄连》游戏的重要组成部分）。占领据点从一定程度上会影响人力收入，但对弹药（特殊能力）和燃料（用坦克摧毁步兵）的技巧性的使用可以从一定程度上弥补人力不足。尽管它还不是一个完美的系统，但我相信它会日益完善。

让我们把所有的这些例子都归结为一个简明的结论。

为玩家提供多种获胜的途径。这些途径从某方面相互阻碍，但同时又在某种程度上相互独立。一个游戏若不能为玩家提供多种的成功的途径，它将会缺少战略的多样性，那么这游戏将会变得很无聊。

我认为3-5种可行途径是最佳的，我想我还得写一个后续来阐述我的理由，不过让我们暂时把这个想法放一放，现在我要开始讨论弹药系统。

弹药、魔法以及交易成本

《战争游戏》（Wargame）拥有所有游戏中最精致的弹药系统，《最高指挥官》（游戏邦注Supreme Commander）的飞机燃油系统沿用了相似设定，或许还更有趣：当燃料耗尽时，飞行单元的性能会降低，而不是让它们变得完全无用。我认为这是一种令人满意的折中。

好吧，我们已经愉快地谈论完了这个话题的第一部分：为玩家提供多种成功的途径、向量或者可控因素以及它们如何帮助游戏达到稳态或者平衡。或者至少，为什么我认为这些会起到积极作用。

现在让我们谈谈能够暂时破坏这种平衡的方法。

作为一名玩家，你肯定不想暂时打破这种平衡。你想掐住对手的喉咙，碾压他们、掠夺他们直到最终他们徒留悲叹。这很正常，我们所有人都想这么做。我只是希望作为玩家的你，可以为此付出更多一些努力，而不仅仅是使用一些速攻战术。因为同时我也在思考关于你，也许你会是那个被卡住喉咙、被碾压的那一方；关于比赛中失败方经常感到不公平，而另一方似乎注定会胜利；关于如何尽可能经常地为玩家或团队提供愉快的体验，同时还能让技艺更高的玩家/团队取胜。

因此，总的来说，我支持游戏提供这样的一些机制，能够让你暂时地压制对手、采取一些措施让对手陷入困境，这样他们就需要暂时重新部署并且将控制权交到你的手中，但在游戏后期又能让他们恢复战斗，总之游戏在一种互相博弈中进行，而不是像《齿与尾》中让某一方的优势越来越大。

Relic的游戏虽然都比较暴力，但是却充满了这类机制。像引擎爆炸这样的车辆损毁状态限制了一些主宰游戏的大型单元，但是玩家仍然可以在后期通过恢复继续使用它们。游戏中的战车可能损坏，（或者被偷走！希望这不久后RTS游戏都能采用这个模式）让他们恢复到原本的状态需要付出时间和资源，而不是直接买新的替代。

《地球2150》（Earth 2150）中，摧毁敌方的弹药供应单位将让对手在一段时间失去反击的能力，从而改变比分情况。此外，这还能鼓励玩家混合使用武器类型，使一些单位避免受到这种侵扰。

弹药系统的作用与之类似。像《地球2150》、《战争游戏：空地一体战》这类游戏中存在一些系统，一旦受到了干扰，会让你的部分军队无法应对。这是一种当规模和装备处于劣势时的打击方法，一种暂时摆脱敌人的方法，一种抵抗攻击和制造僵局的方法。

在弹药系统中，典型应用是利用一个移动的单元或者静止的建筑为附近的盟军提供储备资源，且一旦该单元发动攻击就会耗尽资源。如果没有持续的补给，军队的弹药最终会耗尽，并且无法进攻。《战争游戏》及其姐妹游戏《钢铁之师》（Steel Division）在这方面做了进一步的改变，单位在地图上移动同样需要燃料。

这些系统，以及我将要讨论的其它系统，暂时阻碍了游戏中效率的积累，为可能落后的玩家提供了重新获得优势的机会。

当然，暴雪公司意识到了这类事情以及它的消极面，并且只对那些施法单位进行这样的处理，让一部分军队在耗尽“弹药”时变得毫无用处——我认为他们已经做出了让步，许多基于法力的单位能通过某种方式重新获得所谓的“弹药”，或者是在没有弹药的时候也不至于毫无用处。

因为，当然，没有了弹药，单位就无法使用武器，变得毫无用处。无论对于领先的玩家还是落后的玩家来说想必都不太好受。

弹药系统的一个巧妙之处是它们并不是即时生效的：与EMP武器或高阶圣堂武士不同，切断弹药供应不会立刻造成影响。弹药或者魔法不是一种二元化的资源，并且随着时间推移会逐渐被耗尽。因此，失去弹药的玩家有机会在局势恶化前做出改变。还记得我之前关于二元和多层次的论证吗？当你知道如果无法尽快重新恢复资源你的军队很快就会开始失去效力时，你会感到恐慌，但是这种恐慌比发现自己的军队根本毫无反击之力时的那种沮丧要好得多。

《钢铁之师》和《战争游戏》通过大量的作战单位避免弹药和燃料系统的二元化，基于不同武器类型的弹药成本也弥补这一点。用炮弹？那供应弹药的战车很快就会被掏空了。步兵或者坦克机枪？那就是完全不一样的情况了。 这鼓励了玩家混合使用单位以及撤退到前线作战基地（FOB）同时精心管理和隐藏有价值的弹药供应武器。

此外，混合使用依赖弹药或独立于弹药的单位（正如暴雪游戏中的施法者那样）是抵消这种影响的好方法，尽管我觉得《星际争霸》对弹药尤其保守——就英雄和施法单位的力量以及涉及法术使用的相互作用的深度（比如法力消耗和反馈这样的反法术工具）而言，《魔兽争霸3》更胜一筹。法术操纵的子经济是我们之前谈论的多种胜利途径之一。

弹药系统存在的一些问题

正如我之前所说，效率是即时战略游戏设计的核心之一。RTS玩家将总是试图寻找获得优势的最有效的方法（并且拥有多个半独立的游戏系统是使游戏的效率更加多面化、有趣的一种重要方式）。任何弹药系统出现的任何问题都足以引起玩家的关注，不过还是要限制弹药系统的整体影响力以避免它破坏了其他重要的游戏系统的平衡。

是的，这是大多数游戏系统的问题，所以这并不奇怪。我只是觉得这里有必要提一下，由于我提倡在战略游戏中增加弹药系统的使用。

实际上，除了传统对输出伤害的关注外，弹药系统当然还存在许多其他问题，这些问题我在上面非常简短地讨论了。其中一个大问题是基于单位的或其它杀伤性弹药供应系统很容易打乱DPS。也就是说，DPS更高的玩家将更有机会打破弹药供应系统。这是战略游戏DPS设计中一个难以解决的问题：输出伤害让玩家们叫苦连天，但是其它的替代方案往往好不到哪去，或是更糟。

弹药系统也会有不公平的时候，特别是如果它们是二元化的或被广泛应用到整个军队中的话。 正如被大面积眩晕和群体控制让受影响的玩家们感到恶心一样，当玩家的战争机器被一连串的死亡单元包围时，《战争游戏》式的弹药系统也会让玩家感到糟糕透了。当然，这就是为什么我认为《地球2150》的系统设计更加明智：不同类型的武器有不同的运作规则，如激光武器需要蓄力但不需要补给，或等离子体武器随着时间的推移会变得温度过高，除非停火一段时间，否则会损坏其承载的装置。

总结

本文从反对即时战略游戏中的二元化结果出发：当然这些结果肯定会产生，我的重点是通过游戏系统可以通过不同的调整和应用让这些结果更像一种偏离（outlier）。当玩家认为他们以有意义的方式参与游戏（特别是在竞争游戏中）的时候，游戏给他们的感觉是最好的。并且即时战略游戏可以通过提供多种半独立的获胜方式来达到一种均衡或稳态。

这个基础之上，我尝试论证弹药系统可以暂时弥补我所希望的平衡状态——切断了基于输出的互动，将焦点从输出转移到保存军队作战力上。这一点和与玩家战斗能力无关的胜利条件结合起来，即：胜利不依赖于剥夺敌人的反击能力，这样的系统提供更多的互动元素并减少了能够对胜利起决定作用的元素数量。

《最高指挥官》或《命令与征服》中的“力量”机制就是如此，本文的初稿试图加入这部分内容。但为了保持简洁，我把删除了关于发电厂或压制系统的讨论（另外一种我所认同的抗衡机制，尽管我认为弹药系统总体更好）。

我希望这篇文章至少能够带给你一些值得思考的东西。这篇文章更多地介绍了我在策略游戏设计方面的个人理念，而不是对现有机制系统的案例分析，因此我认为我最希望的是激发一些关于弹药系统和竞争性游戏玩法的思考和讨论。

感谢阅读，并一如既往：战场上再见！

关于这篇文章和我的写作的一般说明

在我的写作中，我提倡的战略游戏系统，旨在赋予玩家权力，并创造有趣、有意义、有深度和公平的游戏体验。我是一个狂热的战略游戏玩家，虽然技术不是特别好，我的大部分观察都来自于我自己的经验和喜好。

另外，我想向Pocketwatch Games道歉。 我认为《齿与尾》是一款出色的游戏，但它的设计非常适合用来探讨这篇文章提出的问题，所以我在本文中多次把它作为一个有些负面的例子使用。

本文由游戏邦编译，转载请注明来源，或咨询微信zhengjintiao

Strategy games, in some ways, are all about limitations. You have limited resources to spend on units and structures at any given time, limited time to react to enemy attacks and tactics, limited attention and reaction speed to manage as you try to eke the most out of your limited supply of units, each with its own constrained functionality.

There is a huge element of skill in strategy games which revolves around mitigating your limitations: ramping up your economy quickly to build (or rebuild) as large an army as quickly as you can (or on striking with an unexpectedly quick army), maximizing your efficiency in combat so that each unit does as much damage as possible to the enemy, paying for itself multiple times and forcing your enemy into a resource deficit so they need to spend resources on rebuilding while you spend them on getting farther ahead. Pure, brual efficiency is a core watchword of RTS, and generally centers around dealing as much damage as possible as quickly and with the least damage and cost to yourself that you can manage.

This is fine, this is well and good, and this is how the game (no matter if it’s StarCraft or Age of Empires or Command and Conquer or a hypothetical My Little Pony RTS) is played.

But as the genre has matured, designers have experimented with different types of efficiency. Brute forcing DPS can only get you so far, you see. We’ll talk about that, it’s important to the conversation.

So, bottom line, what is this article about? What is the point I’m trying to articulate? First, it’s that strategy games are best when they include roughly 3 or 4 somewhat independent systems for players to pursue success. And second, that at least some of these systems should provide temporary or recoverable advantages for a player over another. And how giving units ammunition can help with all this.

Binary Is Bad: Gradations Keep Things Interesting

and Tail, while it has a lot to recommend, tends to have really binary combat outcomes which leads to few quick, explosive battles that decide the game in one fell swoop. This happens in Grey Goo as well, where resource efficiency is often the primary determining factor in match outcome.

Fundamentally, I see strategy game design as a delicate balance between 3 primary and somewhat oppositional driving mindets. First, quality RTS design dictates that every system in the game should provide players increasing incremental benefits for skillful handling: the more attention and control you execute on a system, the more you should benefit from it, even though there may be decreasing returns from doing so. Some examples of this include mineral line management in StarCraft 2 including knowing when to expand and some of the little tricks you can pull to eke a slightly higher income out of mineral mining in the early game, Marine splitting to counter Banelings, or strategically selling structures in C&C games. Most combat ‘micro’ falls into this category, though economics management is important here as well.

Secondly, strategy games are most interesting when they present multiple success and failure states for any given action. If combat encounters are prone to ending with one player’s force entirely decimated while the other player has a substantial army left, the game will mostly likely feature many situations where matches are decided after a very few encounters: this is common in StarCraft 2 matches, where a single successful attack can and often does irrevocably decide the match. Tooth and Tail features an even more binary system, leaving a player almost without options or recourse if their opponent gets solidly ahead of them in production or income, especially in the early game.

While WarCraft 3′s battles can be strongly decisive, its supporting systems such as worker protection mechanisms, hero leveling and mana pools, and the difficulty of forcing combat encounters to promote tactical combat encounters with a wide variety of outcome states

Creating situations where combat is able to resolve in a partially successful or partially unsuccessful state, allowing players to avoid total and catastrophic failure across the broadest subset of combat and harass encounters, allows the game to continue and a player who might have been temporarily set at a disadvantage through sloppy play or a momentary inattention to redeem themselves, while still putting the player with the better performance at the advantage. For instance, in the Company of Heroes and Dawn of War games, being able to retreat squads back to your base allows you to retain combat ability despite the loss of whatever point of ground those units were striving to hold or gain.

Or in WarCraft 3, where unit health pools tend to be large enough to allow constant poke-and-prodding while players jockey for the stun or body block that allows them to whittle away their opponent’s army. Or, to also use WarCraft 3 as an example, where each faction has options regarding base defense: Burrows for Orcs, Ancients for Night Elves, that allow a player to minimize their losses while their army runs or Town Portals back to defend their buildings and units.

There can be tremendous nuance in these systems: in Relic’s retreat mechanic, knowing where and when to retreat, as well as knowing the fleeing squad’s path back to base is critical: it is possible for a player line up a high damage squads along retreat paths to snipe fleeing units, making the system still one where skill is involved.

Taking A Look At Skill

Relic victories and Wonders in games like Dawn of War and Age of Empires allow different types of play to be successful and different types of efficiency to drive a win than in “destroy your opponent’s stuff” games like StarCraft. And the extra method to drive a win can enable a player who’s fallen behind the opportunity to still try and pull off success along a different avenue.

Why does this matter? Clearly, in strategy games, the player who is better should win, so why prolong the agony of the other player’s defeat? Why give the worse player more chances to accidentally (or via dumb luck) defeat their superior opponent? This comes down to the nature, and the fun, of strategy games. It comes down to creating interesting competitive experieinces for people, and participating in those experiences yourself.

Fundamentally and frankly, I’m basing this assessment on overall game feel and what I see as an effort to keep matches fun and interesting for as long as possible and across the widest possible subset of matches; I see specific mechanical paths towards making that a reality. Please understand that I am advocating for a personal preference in game design and that I’m pushing for a slightly different definition of ‘skillful play’ and even of ‘winning’ than you might see in most strategy games. What I’m looking to do is advocate for game designs that allow players to feel like they’re still able to act and participate in the game even when they’re losing. I am advocating for empowering players, and trying to acknowledge the drawbacks of my recommendations as I put them forward.

In strategy games, as in all competitive games, the player with the better strategy and execution deserves to win. But consider a game like SmallWorld, or like tabletop Warhammer 40,000, where the winner is almost always determined by the end state of the game upon its end condition being met, or in Dawn of War with its Relic win, or in Age of Empires where one can achieve an economic and defensive victory via a Wonder, allows win states at least somewhat independent of either player’s overall ability to wage war.

The board game SmallWorld, where victory is determined by total funds amongst all the races you controlled during the game’s duration. I’m not a huge board game aficionado, but I’ve had tons of fun with it[

Skill-based victory is still skill-based victory. There will always be situations in skill-based competitive games where a crushing victory happens, and that’s fine and good. What I’m interested in is strategy games which force the better player to have more consistency in establishing dominance, or in having them win while still allowing the losing player access to all or most of their ability to act on the game. What I am also interested in is systems which encourage players to master and demonstrate skill along multiple avenues in order to achieve success.

Multiple Avenues For Success

Despite my appreciation for many of its mechanics, victory in Steel Division is still highly correlated with your ability to keep your army units alive, with few ‘levers’ to pull to reverse your fortune if your opponent gains the lead in this regarw

It is my contention that competitive games are most interesting when all players feel like they have some option to continue and succeed for as long as possible. Clearly, there will come a moment when success, and more importantly the feeling of being able to succeed, will be lost for one player or team, but that moment can and should be pushed back as much as possible. When a player feels they are unable to act on the game in meaningful ways, when it feels like they have no options or control, they are frustrated.

And just what does this have to do with ammo, mana, morale, and other similar systems in strategy games? I’m getting to that. Bear with me.

Above, I have tried to lay out my position that a state of homeostasis or equilibrium (analogous to the midgame phase of most RTS) creates the most interesting game state for both players, and that tipping the game from this state towards a win should require, in most cases, repeated successes by a dominant player in order to insure a win. Further, this should be facilitated via systems like:

• Multiple systems that drive player success, each being at least partially independent from one another. This allows players to still go for a win if their opponent hampers their effectiveness in another area. An easy example is in StarCraft 2 early to mid-game, when players can choose a combination of economic expansion, army size, harassment options, or tech to attempt to overpower their opponent.
• Win conditions that are divorced from a player’s ability to play the game. For instance, in StarCraft you win by removing your opponent’s ability to fight back, while in Dawn of War 2 you win by reducing your opponent’s Victory Point total to 0 by holding more VPs than your opponent for longer.
• Creating systems where players can recoup or reatreat from a loss, or where combat outcomes tend to be less binary. One example is the Retreat mechanic found in the Dawn of War and Company of Heroes games.

So here’s where we get to efficiency. The fewer vectors for success there are, the more success is driven by efficiency along those vectors. In StarCraft 2, for example: the core resources of Minerals and Vespene are the primary mechanism for expanding the player’s options and acting on the game state, and expanding a player’s income with both of these is vitally important. But the number and position of each unit factory type also affects a player’s ability to produce units, and having too many or too few of a given factory type can itself cause problems. Likewise, while upgrades in StarCraft can sometimes be subtle, sometimes having additional armor or damage can tip a type in your favor: overall, managing all of these things provides each player with a number of interesting choices for how to develop and implement their strategy… in ways that often are completely undone by players’ (especially those in Platinum league and below) to deathball their massive armies around, where victory often devolves into who can kill the other player’s army more often and rebuild theirs more often.

While a large part of combat in Company of Heroes 2 still comes down to the ‘gordian knot’ of DPS, systems like Suppression, vehicle damage states, smoke, and the interlocking infantry counter system gives players plenty of ways to derail an opponent, and widely spaced control points make it possible to attempt to strike where your opponent might be weaker

So all of those interesting choices, except at the higher levels of play, can definitely be bypassed or subsumed under the weight of income and DPS. Obviously, this is a bit reductive: abilities, positioning, and other tactical and strategic considerations (force field, observers, roots, psy storm, etc) can and do come into play.

Breaking it down, StarCraft 2 has a lot of factors in play, many of which are at least partially avoidable or ignorable, until high levels of play. I mean, it’s a pretty flexible and robust system. There’s a reason it’s so popular.

Let’s contrast this with Tooth and Tail, though. In Tooth and Tail, there is only one resource: food. And it’s used to build structures, build units, build farms (economy) and take new bases/territory. So ultimately, everything in Tooth and Tail comes down to food. And that’s the point of the game. It’s incredibly focused, and I think that’s praiseworthy. But in multiplayer, it can be merciless. Once things start going badly for you in Tooth and Tail, especially in that pernicious early game before you’ve expanded, losing, well, anything can put you into the kind of tailspin that happens in Zerg v Zerg matches in StarCraft 2: that is, you’ll never be able to catch up economically. Economy growth curves make it increasingly statistically likely that food ‘wasted’ on units you have to rebuild will allow your opponent to get that much farther ahead, have more resources to spend on units, and so on rolls the snowball.

In Tooth and Tail, it all comes down to food.

Combat in Tooth and Tail is much more linear than StarCraft 2, since you basically only have the ‘levers’ of combat skill, unit choice/positioning (there is an interesting element here where you’re able to choose between lower level and higher level/more powerful units, but that strongly correlates to income), and income to pull, and they’ve been somewhat dulled by passing all orders through the imprecise sieve of your Leader unit.

To further belabor the point, Company of Heroes 2 gives players a good variety of levers to pull: Manpower is largely fixed income, which inflates or decreases based on how many extant units you have. Fuel is scarce and punishes poor play with vehicle units. Munitions start out scarce but become plentiful in the mid-to-late game, where armies are larger and dealing wide damage with Napalm is a little easier to come back from. Add to this a really dynamic infantry combat system with cover, morale and the suppression system, smoke and grenades and indirect fire weapons. This expands to include light vehicles, which interact differently with indirect fire and MG weapons and regular infantry, then expands again with heavier vehicles and AT guns, and expands again in the late game with super-heavy tanks. Each phase of the game increases the number and nuance of interactions, while expanding Munitions income allows more consistent use of game-changing abilities.

So, in Company of Heroes 2 you have those 3 resources (4, really, if you count VPs as a resource), each with its own purpose, that players can spend in various ways, then there’s the interlocking unit dynamics that has little, but some, leeway built into it – like how MGs can provide light anti-tank in the early game if a player is slow to get out true AT solutions (obviously unit choice is a big component of Company of Heroes games). There’s territory control, which of course influences income to a degree, but manpower (unit counts) are offset somewhat by skillful use of Munitions (special abilities) and fuel (many tanks mow down infantry). It’s not a perfect system, but I see a lot of promise in it.

So, ah, ok. Let’s boil all of those examples back down to a concise statement.

These, uh… aren’t the ‘avenues for success’ I’m talking about, just so we’re clear.
Give players multiple avenues for success. Have those avenues bottlenecked in some way, but somewhat independent of each other, as well. Few avenues for success means less strategic diversity, which means the game will get boring.

…God, this is going to require a follow-up piece where I try to lay out the reasons I think 3-5 main actionable paths to pursue is the best number. Let’s put a pin in that idea for now and finally start talking about ammo systems.

Ammo, Mana, and the Cost of Doing Business

While Wargame has one of the most elegant ammunition systems in all of gaming, the aircraft fuel system from Supreme Commander works along a similar principal, and is perhaps more interesting: reducing the performance of air units when their fuel runs dry, but not rendering them useless. A happy medium, in my eyes.

All right, we’ve had our fun talking about the first part of this topic: the idea of multiple paths or vectors, or levers that players can pull to drive success, and how that can help establish equilibrium or homeostasis. Or at least, why I think such a goal is a positive thing.

Now let’s talk about ways to temporarily upset that equilibrium.

As a player, you don’t want to temporarily upset the equilibrium. You want to grab your opponent by the throat and grind them into the dust until there’s nothing left but pillage and lamentations. Which is all well and good, that’s what we’re all here for. I’m just all about making you, the player, work for it and give a little more effort than executing one decent six-pool or cannon rush. Because I’m also thinking about you, as the player, getting grabbed by the throat and ground into the dust until there’s nothing left but lamentations. About the situations where losses feel unfair and wins seem inevitable. About ways to give both players or teams an enjoyable experience as often as possible while still having the more skilled team/player take the win in the end.

So, in general, I’m in favor of things that offset your opponent’s efforts to wage war, temporarily. Something you can do to put your opponent on the back foot, that will require them to regroup and cede control to you, but which can later be rectified and bring them back to battle without that Tooth and Tail, Zerg versus Zerg style snowball towards irrelevance in the battle.

Relic’s games, as brutal as they are, are filled with such systems. Vehicle damage states, like busted engines, are status effects which limit the operating power of those large, game-altering units in ways that let the player bring them back to bear in the future. Vehicles can be crippled (or stolen! I hope to get into object permanence in RTS sometime soon), and bringing them back to full fighting fury is a time and resource commitment, but far less so than replacing one outright. It’s a temporal, temporary advantage.

Killing ammo supply units in Earth 2150 will deplete your opponents’ abilty to fight back over time, changing the calculus of an otherwise close or ‘won’ combat encounter. Also, this encourages players to mix weapon types, making some units impervious to harassment in this way.

Ammo systems work similarly. In games like Earth 2150, Wargame AirLand Battle, et cetera, there exist systems that, if disrupted, leave a portion of your army unable to respond. It’s a way to fight a larger and better-equipped army, a way to take enemies out of the fight to deal with later, a method for stalling assaults and creating stalemates of victory marches.

In systems where units use ammunition, the typical implementation is some combination of a mobile unit or stationary structure which provides nearby allies with a personal stockpile of a resource which is depleted every time that unit attacks. Without continual resupply, one’s army eventually will run out of ammo, and become unable to attack. Wargame and its sister Steel Division take this a step further, with units also requiring fuel to be able to continue to move around the map.

These systems, and the others I will discuss, are ways to temporarily throw a monkey wrench into the larger calculus of efficiency, and allow space for a player who might have fallen behind to temporarily reset the scales of balance in their favor.

Blizzard, of course, realized this sort of thing as well as its downsides, and relegated this idea solely to spellcaster units, such that a portion of your army becomes useless when out of its ‘ammo’ – and I think they’ve actually backed off of that a bit, with many Mana-based units having a way to recoup said ‘ammo’ or to not be completely useless without it.

Because, of course, without ammo, units can’t fire their weapons, which means in effect they become useless. Not an ideal situation if you are already behind. And, yeah, kind of frustrating even if you’re ahead.

It’s a really crappy situation to have units stuck en route to Point Bravo without fuel; binary situations create excessive player frustration, especially without the ability to, say, airdrop in a unit to replenish them on the field.

One neat thing about ammo systems though is that they’re usually not immediate: unlike an EMP or High Templar Feedback, there’s a bit of a delayed impact to cutting off an ammunition supply. Ammo or Mana is a resource that isn’t binary, and is depleted over time. So the player who’s lost their supply of ammunition has a chance to correct the situation before it bites them in the butt. Remember what I said earlier about binary situations being bad and gradations being good? Well here it is. The feeling of panic you get when you know your army’s going to start losing effectiveness quick if you can’t reestablish your stockpiles soon, that’s much better than the abject frustration of being subject to a root effect and completely unable to respond while your army is obliterated by high-octane area damage.

Steel Division and Wargame attempt to deal with the binary nature of their ammo and fuel systems via the sheer quantity of units that the player is fielding, offset with variable ammo costs based on weapon type. Artillery shells? That’ll empty out an ammo supply vehicle pretty quickly. Infantry or tank-mounted machine gun bullets? A different story altogether. This encourages unit mixing and a combination of retreating to a FOB and careful management and hiding of valuable ammo supply weapons, which are the lynchpin of your ability to operate out on the map.

Pylons and power plants are a similar system, just targeted at production. The ability to take out power and render your opponent temporarily unable to produce new units can be really powerful at pivotal moments, but is much easier to recover from than having your entire infrastructure demolished.

Also, having a mix of ammo-dependent and ammo-independent units or actions (as Blizzard does with spellcasters) is a good way to offset the impact of this, though I feel that StarCraft in particular is too conservative with ammo – WarCraft 3 was a much stronger approach, given the power of hero and spellcaster units, and the depth of interactions involving mana usage (think of anti-spellcaster tools such as mana drain and feedback). The sub-economy of mana manipulation is another one of those multiple avenues for success we were talking about earlier – see how it’s all tying together?

Doing Nothing: Some Problems With Ammo Systems

Protoss High Templar are a huge liability when they were out of Energy. Fortunately, unless you’re literally crazy, you’re probably not building your entire army out of these guys.

Efficiency, as I said, is one of the core watchwords of RTS design. RTS players will always try to find the most efficient ways to gain an advantage over their opponent (and having multiple semi-independent game systems is one key way to make this aspect of the game more multi-faceted and interesting). And the problem with any sort of ammunition system is to make it important enough for players to pay attention while limiting its overall impact so that it doesn’t override other important game systems.

And, yeah, that’s the problem with most game systems, so it’s not really a surprise. I just feel like it bears mentioning here since I’m effectively advocating for the increased use of ammo systems in strategy games.

Practically, there are of course a number of issues with ammo systems that circumvent the traditional focus on pure DPS efficiency, which I touched on above very briefly. One of the big ones is that ammunition systems that have a unit-based or otherwise killable munitions delivery system are prone to DPS-related upsetting. That is, the player with the most DPS delivery power is going to have a better chance of knocking out ammo delivery systems anyway. This is the Gordian Knot of DPS that’s proven very difficult to slice in strategy game design: damage output covers a multitude of woes, and often the alternatives are worse, not better.

Ammo systems can also feel damned unfair, especially if they’re binary or widely implemented across the army. Just as wide-area stuns and CC can feel awful for the impacted player, Wargame-style munitions can suck when your entire war machine grinds to a halt because of a handful of dead logistics units. Which, of course, is why I think it’s smarter to use a system like you see in Earth 2150, where different types of weapon behave according to different rules, like laser weapons need to recharge but don’t need restocking, or plasma weapons overheat over time, damaging the carrying unit unless it ceases fire for a while. That sort of thing.

Wrapping Up

I started by trying to make the case against binary outcomes in real-time strategy game encounters: they’re going to happen, of course, but my point is that systems can be tweaked and implemented that cause them to be more of an outlier. That games feel best when players perceive that they have meaningful ways to interact with the game they’re playing (especially in a competitive game) and that RTS can encourage such equilibrium or homeostasis by, among other things, providing multiple semi-independent ways to succeed.

I tried to build on that foundation by positing that ammunition systems work really well as methods to temporarily offset my desired equilibrium state – losing a supply line provides a cutoff to DPS-based interactions, which shifts the focus from pure DPS to preserving one’s ability to continue to support their army’s continued operation. Tied with win conditions that are themselves partially independent of a player’s ability to fight, that is: win conditions not reliant upon stripping your enemy’s ability to fight back, such systems provide more vectors for interaction and decreases the amount that any individual vector can lead to a decisive win.

The same goes for ‘power’ mechanics like we see in Supreme Commander or Command and Conquer, of course, and early drafts of this article tried to shoehorn that in as well. I decided – for brevity, you see – to exclude talking about power plants or suppression (another temporary offset that I favor, though I think ammo systems are overall better) here. Maybe there’s another 4000+ words on that floating around, waiting to be unleashed.

I hope that, at least, I’ve given you something to think about. This piece is more a look into my personal philosophy on strategy game design than a substantive case or analysis of existing mechanical systems, and as such I think the most I can hope for is to inspire some thinking and possibly conversation about ammo systems and competitive gameplay dynamics.

Thank you for your time, and as always: see you on the battlefield!

A Note On This Article And My Writing In General

In my writing, I advocate for what I see as strategy game systems designed to empower players and to create gameplay experiences that are fun, meaningful, deep, and fair. I am an avid and passionate strategy game player, though not a particularly skilled one, and most of my observations are drawn from my own experiences and preferences.

Also, I’d like to apologize to Pocketwatch Games up front. I think Tooth and Tail is a great game, but its design really lends itself to the particular issues this post raises, so I use it as a somewhat negative example several times in this article. (source：Gamasutra  ）