在游戏开发循环中，关卡设计就是数据输入和布局分配。无论关卡出于何种意图和目标，它们都类似于任务、阶段、地图或玩家互动的其他载体。作为关卡设计师，你主要负责的 是游戏可玩性。这篇文章将让你了解如何为各种类型的游戏设计优秀的关卡，无论是由大批坦克执行的军事任务、在空中遭遇飞行模拟器还是角色扮演游戏中的地下城、解谜游戏 的关卡或某个需要玩家战胜强大敌人的世界的地图。
我会在下文中提出某些关卡设计理论，但开篇会先探讨何谓优良的关卡设计。然后将从象棋和GI Joe的动作中深入探究电脑游戏设计的非电子源头，以及我们能够从它们的成功中 获得哪些经验。最后，我们将彻底剖析故事情节背后的理论，并探讨如何将它们用于关卡设计中。
玩家是指那些付费购买游戏或花长达半个小时的在线时间下载你设计的关卡的用户。如果你正在或有意愿为某开发商或发行商工作，那么用户就是你实现自己梦想的舞台。正如所 有现代营销学校告诉你的那样，只有创造出让用户感到满足并愿意承担其费用的产品才能形成一笔生意。作为关卡设计师，如果你想要获得成功的话，就必须各位关注那些能够让 用户感到满意的事物。
你必须布置满足玩家期望的场景和演员。也就是说，你需要设计的地图不仅看起来与玩家所处的游戏世界相符，而且还必须包含有让玩家不必思考现实生活状况的元素。许多种常 见错误可能会破坏玩家逃避现实的感觉。这可能包括图像差错或冲突等漏洞，但以设计作为出发点来看，不恰当的内容也可能产生影响。比如，麦当劳的金色拱门商标出现在中世 纪的城镇中显然与背景极为不符。同样地，如果角色告诉玩家按ctrl+T来做出动作，这会让玩家意识到他正在用电脑玩游戏而不是存在于某个虚幻国度中。通常来说，为保持玩家 逃避现实的感觉，所有的内容都应该设计成游戏场景中可能看到、说到或做到的事情。
乏味的关卡可能导致游戏失败，尤其当这个关卡是游戏前数个关卡之一。游戏评论员和多数玩家在判定游戏好坏之前，只会给予游戏数个关卡的展示时间。优秀的关卡设计师知道 自己的产品所需达成的目标，他们时常会询问自己“这个关卡有趣吗？”对于许多设计师而言，较为困难的部分在于他们觉得有趣的东西，目标用户或许并不会觉得有趣。作为关 卡设计师，你需要理解核心游戏玩法，这是制作人和首席设计师所所表达愿景的部分内容。你需要努力理解这方面内容，站在目标用户的位置思考问题。
挫败感也可能让游戏失败。速度变慢或图像差错等问题可能会让玩家的娱乐心态消失殆尽。如果关卡设计师能够关注技术限制和设计师对于如何放置艺术作品的指导，可以避开许 多此类的漏洞。当然，设计师也可能制作出令人产生挫败感的漏洞，比如破碎的AI脚本或玩家期望完成却没有完成的任务。更为糟糕的是，设计师可能创造出所谓的“show stoppers”。show stoppers指那些不可完成的任务、无法解决的挑战或不可避开的陷阱，这会让玩家大感挫败。优秀的关卡设计师会察觉到这些问题，在消费者接触游戏之前通过 细致缜密的游戏测试来解决它们。
事实上，玩家不希望购买或玩两款相同的游戏。当然，就像《星际迷航》的粉丝和Gor科幻小说的读者一样，有些玩家可能会买到同样的游戏规则甚至只是情节、场景、角色和艺术 设计有些许变化的相同游戏。这务必要引起关卡设计师的注意，人们不喜欢玩相同的关卡。这不仅会破坏娱乐价值，还无法激发想象力。因而，在关卡中引进新的情节、挑战、场 景和角色（游戏邦注：比如敌人）是至关重要的做法。
玩具总是能令各个年龄段的孩子为之着迷。玩具火车是最具互动性的玩具。你可以扮演设计师、建筑师、油漆匠、乘客和工程师（游戏邦注：如果你乐于制造自然灾害的话，你甚 至可以扮演上帝）。GI Joe和芭比娃娃也都有互动功能，比如可移动的肢体和可更换的衣物和配件，甚至还有车辆等。他们利用孩子的抱负和梦想来赚钱。我最喜欢引用某纸质游 戏行业资深人士和前玩具制造者的话是：谁会知道仅仅一个玩偶就会这么畅销呢？
在游戏中，玩家很轻易便可意识到自己属于英雄这个角色。整个游戏可以视为故事，每个关卡就是对立发生的一部分，即英雄和对手间的较量和冲突。论点的深度仅局限于对玩家 角色和场景的介绍。加入你玩过游戏前篇的话，有些游戏续作会直接建立于原作之上开始引进冲突。合成就是游戏结局所发生的所有事情。无论是胜利或失败，你或许都会看到过 场动画或某些叙事文本。正如好莱坞电影那样，合成从来都不会非常长。
每个关卡都有其自身的故事。作为关卡设计师，你在准备初始情况时便设定了论点。你设定了玩家所处的位置，可能还会指定其最初的武器或部队以及咒语或其他魔法。你通过地 图或谜题来演绎场景。场景和情况可以随关卡的进展而发生改变，因为关卡各部分的内容会不断展示给玩家，比如加入新角色、新元素、能力提升、加入新玩家和新的敌人。由于 游戏具有互动性，你必须非常注意玩家在关卡进行过程中特定时间或地点可能遇到的每种情况。
每次遭遇都有论点，即战斗开始前那个时刻你感受到的害怕、兴奋或期待以及当时对所遭遇情况的认识，这些都会影响到战斗。比如，玩家或许会贸然闯进妖怪的巢穴，而且在战 斗开始之前的那一刻就对当前情况有所认识，在妖怪头领的附近有辆大型坦克，妖怪位于其火炮射程之内。但是，我们不能假设玩家永远都会做出我们预想的事情，从我们预想的 方向进入妖怪的巢穴。关卡设计师必须对此有所规划，并为反常行为提供奖励。以上面的例子为例，玩家或许会偷偷摸摸地从相反的方向进入巢穴，在尽到妖怪之前先看到坦克。 他会选择将其炸毁还是小心翼翼地继续前行呢？加入敌人选择利用坦克来对付他又会如何呢？我们将事态设计得更为复杂些，假设在这部坦克旁边有辆更棒却更易被摧毁的升级版 坦克。玩家应该怎么做呢？在这种情况下，你的选择并非唯一。事实上，设计师也不要将事情设置成选项唯一。作为论点的部分内容，你需要的只是将选项提供给玩家，让他们来 决定怎么做。
对立面是玩家与你的关卡互动的层面。通过设立敌军并规划他们的行为、限制玩家行动的时机和速度或他们必须解决的谜题，你正在制造冲突。而冲突应该与关卡的核心玩法有关 。如若不然，你就会设计出需要在论点阶段制定多项计划的关卡来。换句话说，如果玩家没有在初始阶段就制定完美的计划，那么在关卡中就会很容易失败，如果玩家制定了正确 的计划，就会在关卡中通行无阻。除了战略游戏玩家之外，多数人讨厌这类关卡。玩家希望在冲突出现之时有解决的方式，你不能把他们假设成无所不知或拥有预知未来的超能力 的人。设计师通常犯下的错误是，让呈现出的挑战完全无法战胜，除非玩家之前玩过关卡并知道关卡中的情形。应该让玩家在首次遭遇冲突和挑战之时就可以设法将其解决，这是 关卡设计中的必要想法。
对立面是划分玩家技术的工具。它的存在区分了技术娴熟的玩家和新手以及专注玩家和浅尝辄止者。理想情况下，游戏中应该存在多种胜利条件，因而技能本身并没有绝对的好与 坏。如果只有专业玩家方能打通关卡，那么你就会失去90%的市场，你的游戏也就无法畅销。同样，如果任意普通玩家都可以打通关卡并获得所有的奖励，那么有50%未受到挑战的 玩家便会觉得不过瘾。但是如果你的关卡有个令普通玩家感到满足的胜利条件和奖励专业玩家的可选挑战，那么你就可以令各个层面的玩家感到满意。
故事通过呈现出有价值的内容来保持你的兴趣。人们购买书籍或观看电影的目的不是为了只听着角色们谈论天气，除非天气本身是其中的重要因素（游戏邦注：如《Twister》之类 的灾难电影）。精巧的故事需要包含演绎场景、发展角色或推动情节等细节。虽然书籍中可能不会涉及大量的细节，但是电影不能采取这种做法。电影旨在短期的时间内吸引观众 的注意力，他们想在90分钟或更短的时间内体验整个故事。电影努力将注重点放在最为重要的细节上，而这些细节通常都会涉及角色互动。
关卡设计也是如此，除非你讲述故事的时间非常少。因此，你必须更注重角色互动细节，尤其是那些设计玩家的内容。玩家看到或作出的每件事情都必须使故事进一步得到发展。 玩家所做出的所有努力都应该推动他们走向故事的完结或使他们远离与对立面的冲突。随着游戏进行下去，应该让玩家逐步发现有关他们自己以及对手的更多信息。玩家解锁新天 赋、找到新武器或武器升级、深入理解战略或遭遇敌人新战术和新类型的敌人，这些做法都可以实现上述目标。尽管这些建议或许你看起来显而易见，但你会为设计师经常犯下的 错误感到惊讶，他们将大量时间花在玩家看不到的场景细节之上。
多数玩家只有短期关注点，尤其是那些玩主机游戏的玩家。他们没有足够的耐心去挖掘微小的细节和游戏精妙之处。如果你向他们呈现过多的细节，或者游戏玩法过分依赖玩家对 小细节（游戏邦注：如某个对话信息）的理解，那么你就会失去这部分玩家。对非电脑游戏设计师和RPG设计师而言，很难不再关卡设计中添加各种毫不相干的内容。但这种对细节 的过分关注通常会是游戏可玩性受到损害。如果你制作的不是RPG，那么就必须理解故事细节并非关卡设计首先要考虑的事情。
在非互动细节上花费大量时间无疑是对时间和资源的浪费，但是在这方面投入部分精力也很重要，因为玩家偶尔会注意到这些。比如，假如花上一天的时间来精致某个玩家在前往 参加坦克战斗途中花三秒钟时间路过的农场，这是个很可笑的举动。最好只花数分钟的时间添加些许物体，让玩家觉得这是个农场，比如农舍、谷堆、筒仓和些许牛羊。即便你有 充足的时间来创造出各种非互动细节，你依然不可以这么做。过多的细节会让玩家觉得心烦意乱，感观负载过大。如果他们尝试同那些非互动细节互动，其结果也会让他们产生挫 败感。
让场景中的所有细节都存在某种形式的互动，这或许是个更好的做法。《毁灭公爵》在这个方面做得很好。甚至连厕所都有其存在的意义，即便只是为了提供些许幽默。酒吧有个 正在运转的吧台，街机中有《毁灭公爵》这个游，这会引发你产生“我没空自我娱乐。”这种想法。游戏在这方面的额外付出显得很有价值。互动场景营造出非凡的魅力，使得 这款游戏与所有其他的《毁灭战士》克隆游戏有所差异。
电脑游戏在这方面的掌控较为简单，因为他们的目标市场大多数是那些喜欢科幻小说的读者。那些所谓的“巨作”通常是那些在科幻市场之外设立新题材游戏的作品。《模拟城市 》、《俄罗斯方块》、《文明》以及各种运动类游戏在逻辑或科幻方面并没有很大的进步，它们吸引的都是那些对射杀外星人不感兴趣的玩家。即便如此，今天的大多数游戏都是 面向科幻题材的游戏。
假设你正在制作一款科幻游戏，你确实有某种为玩家扩展事件可能性的义务。但是重点在于了解何时何处以及将现实延伸多远。玩家希望他们能做的事情能比游戏中的其他角色更 多。尽管这听起来像是顾及单方面的想法，但这确实就是玩家想要的东西。如果敌人以某种令人称奇而且超乎他们能力范围的行为来攻击他们，玩家会觉得这是作弊。他们希望对 手以有限的动作来与他们战斗。这样他们就会理解，打不过只是个技术问题而已。
然而，玩家却喜欢以超出AI能力的技巧来将其战胜，比如用咒语轻松击败对手。那么，你就应该给予玩家他们想要的东西。用上帝般的力量让他们感到满足。但是必须注意的是， 如果玩家时时刻刻都有那种能力，游戏就会变得枯燥且失去挑战性。技巧在于取得平衡，这样玩家就不会时常获得此类能力，可以通过限制能力的使用时间等方式来实现此目标。 在理想的关卡中，玩家会面对各种各样的可能性，并且以某种难以置信的方式解决这些问题。这样，他们就会觉得他们做了些不可思议的事情，他们是真正的英雄。
为了让其他关卡设计留有余地，首席设计师可能会限制你使用的设计素材。你应该自己决定如何在维持首席设计师和制作人项目愿景的前提下，在这种局限性中开展工作。如果你 难以做到这一点，可以询问他们的建议。他们也许会提供一些指导，或者使用自己的权力为你提供更多设计素材。有时候你需要充分发挥想象力，最大化挖掘这些设计素材的潜在 效用。如果你发现自己没有足够的元素来填充关卡，可以试试之前没尝试过的组合及布局，没准能想出一些新颖的游戏谜题，从而增加自己的素材储备。
许多设计师常犯的严重错误是制作迷宫。为什么是这个错误呢？迷宫是电脑游戏最早引进的谜题形式之一。但现在很过时了，因为制作迷宫无非就是设置墙体，或者其他可阻碍玩 家移动的障碍，它是最容易制作的游戏玩法。但要慎用这种设计，只有在你真正计穷的时候才能考虑这种元素。最好回避这种设计，要善于以新方式使用现有元素来优化自己的设 计素材，或者让游戏设计师制作更多素材。
索要更多素材既是赢得同事尊重也是易遭鄙视的一种做法。但不幸的是，这是你的责任，你得这么做。但要确保你将自己的出色想法妥当传达给首席设计师。如果对方觉得你的想 法不错，可能就会将其列入计划表中。只是要记住，落实想法通常需耗费大量的美术和编程资源，所以如果你的想法被否决了也不要太沮丧。能够重用现成的美术资源，甚至无需 进行编码的想法，往往才是最佳创意。如果你可以用自己的脚本使这些内容呈现可行性，那就再好不过了。当项目开发进入Alpha阶段时（即编码和多数美术内容已经完工），就不 能再指望添加新游戏元素了。
你构想和落实一个关卡时的快乐心情将传递给这个关卡的玩家。当然，工作的时候难免遇到截止日期将近，关卡编辑器在最不应该的时候崩溃这类令人抓狂的事情。也会出现一些 游戏漏洞和画面帧率问题迫使你投入多个小时重新设置关卡。但当你知道自己做的东西一定会很有趣时，这些沮丧感也就一扫而空了。要记住，会有成千上万玩家将体验你设计的 关卡，他们不会知道你在这个过程中所吃的苦头，但却会体会到你从中注入的快乐情感。
但这并不是说，你得腾出更多时间进行尝试，而是说关卡的核心玩法要足够可靠。最好选择一个可为多种玩法留下发展空间的核心理念。在落实关卡的时候，要先确立核心想法， 使其具有可行性。然后再决定这个想法是否具有优势，你是否愿意进一步推动这个关卡的发展。如果确定无疑，那就继续向其填充更多细节，并对这些细节内容进行试验。通常情 况下，正是这些细微元素和细节成就了杰出的关卡。
为玩家提供多条通往相同目标的路径，这是一种既能给予玩家多种选择以及一种自由感，又能确保他们到达相同点的有效做法。但如果每个选择呈现在玩家面前的都是相同的敌人 ，相同的奖励，相同的风险和成本，那么玩家发现这些选择并无实质上的区别时，他们很容易就会对此生厌。所以向玩家呈现选择内容时，要让这些元素具有非视觉效果上的差异 。这种差异可以体现在不同的挑战上，例如潜行路径、陷阱、隐藏的升级道具，或者更好的战略位置。重要的是不要向玩家多次呈现相同的选择，不然这种选择还有什么意义呢？
向玩家呈现选择、挑战或谜题时，要提供可满足不同玩法风格和技能的多种解决方案。有些玩家的玩法较为保守，而有些则喜欢冒险玩法。有些玩家很谨慎，希望在陷入冲突前先 探明更多关卡内容，而其他玩家则可能拎一把枪就直接进入关卡。有些人喜欢走直线路径，有些人喜欢潜行路径。玩家的游戏风格对你的游戏来说可能极具特殊性，你得及早辨识 玩家的不同玩法模式。确保你的关卡设计与心中的不同玩法风格相吻合，这样才能取悦大众玩家。
不要认为每个玩家都会用同种方式体验你的关卡。要注意，玩家如果没有想出应对关卡的替代性或终极解决方案，他们可能就会陷入困境。玩家处理困境以及掌握游戏玩法的水平 各有不同，学习速度也不尽相同。可以为你的关卡提供一些比较容易，但奖励较少的解决方法，但最好要让玩家清楚自己选择简单的解决方式会错失什么奖励。这样才能激励他们 过后再重玩关卡，挑战更高难度。
玩家喜欢尝试和探索。你在关卡中提供越多解决方法、秘密、可替代性路径等内容，就越能够取悦玩家。作为玩家，没有什么比遇到不甚显眼却能大获成功的解决方案更令人兴奋 了。记住玩家几乎都喜欢“不走寻常路”从而找到其他捷径，甜头或者意料之外的道具。设计关卡之时，要考虑玩家希望尝试的内容是什么，并将这些内容呈现给玩家。如果玩家 有“假如这样做行吗？”的念头，你的关卡设计就应该做出肯定回应“没错，可以！”。
如果设计一个看似挑战、替代性解决方案，路径或隐密地点但实际上毫无奖励的内容，那就太糟了。玩家总是想尝试与所有内容互动，而当这种互动毫无意义时，他们就会非常受 挫。那些没有存在意义的交互性游戏物品（游戏邦注：例如可移动的箱子或爆炸性筒罐）很容易让玩家沮丧。因为玩家此前可能已经尝试了数分钟，甚至数小时去探索与这些物件 互动所产生的结果。所以不可让玩家感到失望。
例如，在《雷神之锤》或《Unreal》关卡中，如果玩家看到一些椽子正处于自己所站平台的跳程内，他可能就会想“啊，这是个挑战，我想看看那里有什么东西”。而假如游戏中 的这些椽子并无存在意义，而玩家却花了一个小时尝试跳向第一个椽子，结果却发现自己一直徒劳无功。玩家可能就会产生退出游戏的念头，更糟糕的是，这也可能激发他的好奇 心，使他产生志在必得的念头。他费了九牛二虎之力终于跳到椽子上时，却发现那里空无一物，他可能就会怨自己为什么要在这个关卡白费许多功夫，也可能对关卡设计师心生不 满。所以，设计和测试关卡时，务必注意清除这些“交互黑洞”，或者最好为探索这些内容的玩家提供奖励。
“节奏”就是引进冲突和紧张感，再加上所谓的“肾上腺素上升”。这和故事、电影领域的“正题-反题-整合”模式（Thesis-Antithesis-Synthesis）很相似。玩家（正题）与反 题相接触时就会形成冲突，这种矛盾会逐渐上升，而后才进入“整合”状态，此时的读者、观众或玩家才开始松了一口气（不过玩家也可能重新开始玩这个关卡，假装失败从未发 生）。
控制玩家的行动速度或行动路程也可以明显影响游戏节奏。虽然你无法简单地在《俄罗斯方块》之类的益智游戏中进行这种调整，但你还可以采用其他策略改变玩法速度。通常情 况下，地形会影响玩家行动速度，例如沼泽地会让你减速，高速公路有助于加速，而阻塞的弯曲小道却会放慢你的进程。此外，还可以通过赋予游戏单位不同的行动速度或者行动 限制来改变玩家游戏速度。例如，为玩家提供一个行动迟缓笨拙的坦克，可以让他的部队放慢速度以配合坦克，而如果提供一辆快速行走的全地形车，就可以迫使玩家加快速度。
也可以通过设置敌人速度来调整关卡节奏。例如，在POV射击游戏中，让玩家追赶一个企图从主人的城堡中逃蹿并且行踪不定的忍者；在横向卷轴游戏中，让boss怪物行动更为迅速 或者缓慢。当玩家必须以自己所习惯的速度更快或更慢行动时，就可以构建游戏的冲突感。通过使用改变行动速度来制造紧张感的不同方法，你就可以较为准确地控制关卡中的节 奏。
要保持玩家对游戏的兴趣感，就要善于巧妙地展示游戏内容。这里的内容指的是游戏的视觉元素，包括地势物体、敌人和同盟部队，升级、谜题等。游戏应该向玩家逐渐推出这些 内容，不要在首个关卡就超载这些内容，以便下个关卡仍有新鲜内容可供玩家探索。首席设计师通常会有一些准则引导你下个关卡应该引进什么新内容，最好让它们成为关卡的中 心内容，使其与核心玩法相得益彰。要以生动的方式引进这些新资产，并且要描述这些内容的独特性。
这种情况提出了一个问题：为了让自己的关卡表现更出色，关卡设计师是否可以用光自己所需要的所有素材和内容，无视游戏中其他关卡的需求？答案当然是否定的。如果某个特 定关卡破坏了不同关卡内容的自然呈现过程，那就会导致其他关卡内容更为单薄。这也会迫使其他关卡设计师返工，而这又会损害项目进程并且浪费时间。依我个人的类似经历来 看，我敢担保这种不当的关卡内容呈现方式势必给项目造成重大影响。
关卡设计师的职责就是给玩家制造挑战。只有当游戏胜利条件尚无定论时，其游戏体验才会给玩家带来快感。所以你得向玩家呈现那些考验他们勇气，并且让他们不知自己胜算的 挑战。采取这种做法时要注意迎合不同玩家的能力（详见第6点），并针对玩家的技能和装备提升而增加难度。你的关卡在游戏时间或“关卡进程”中的出现位置应与其所要求的难 度相吻合。在最初几个关卡，要先让玩家学习如何玩游戏，所以这里的关卡可以适当宽容一点。但出现在游戏尾声的关卡则应最具难度，这样才能匹配玩家增长的技能和资源。
说得容易做得难，今天独特的游戏元素非常罕见。但至少在关卡设计领域，你还是可以用新方式结合不同元素，讲述不同故事。另外要记住，没有哪一款优秀游戏能够完全与前辈 或竞争者脱节，而你也不例外。有时候你可以通过调整他人的关卡设计，从而突出你认为其缺乏的元素，或者体现你的关卡更胜一筹。你常会有一些新想法添加到自己的设计素材 中。你可能会发现自己的关卡设计是前所未有的想法，或者从他人那里得到灵感而进行创新尝试。关卡独特性并不需要完全取决于原创性，你的个人喜好会体现在自己的设计中， 而这一点就已足够显示其独特性。无论你如何落实想法，独特性都可以令你的关卡与游戏区别与其他作品。
例如，《机甲指挥官》中的一个任务要求玩家从一艘敌人护航队的相反方向沿河出发。其任务描述要求玩家通过便利的交叉点，在敌人护航队接近退出点前将其歼灭。在关卡早期 阶段，玩家的位置远离这条河以及敌人。如果你事先没有阅读任务描述，或者查看战略地图/任务目标，你就不会知道自己的目标是什么，当然也不会产生任务紧迫感。你也不会知 道敌人想干什么。而当你探索完毕并且打了数场战役时，你却发现自己毫不知情地失败了。在这种情况下，你有可能重玩一次游戏，并且看看任务描述，但也有可能直接关掉游戏 。
后来这个关卡进行了一些改变，在刚开始时就在玩家视野中呈现河流以起敌军护航队的起点。然后你马上就能看到自己的目标，得知它们处于河对岸。你就可以立即调换河岸的导 弹和激光火力，并发现护航队并不会减速来攻击你，你还会得知自己正处于一个过河竞赛中，迫在眉睫的任务就是阻断敌人去路。这样整个任务目标和玩法无需只字片语，就能在 数秒内一目了然地呈现出来而不至于令玩家困惑——所以最好是为玩家展现可视的揭露内容，位置以及敌人行为。
如果你向他们灌输更多信息，他们在同个关卡中的期望可能就会发生变化。如果你建立了特定期望但却无法将其贯穿始终，这个关卡可能就会让玩家产生困惑或者沉闷无趣。如果 你想通过颠覆玩家期望，揭露其意料之外的内容让他们大吃一惊，那就要确保这种设计对关卡的重要性，因为玩家会理所当然地认为这种元素很重要。例如，你告诉玩家他们正处 于一个工业建筑中，而他们却并没有看到任何一个工业设施，他们就会感到困惑，就会怀疑自己是否走错屋子，或者自己上错楼层了。除非这种意外之举对玩家来说真的很重要， 否则你就应该改变任务描述，或在关卡中植入一些工业机器。与此相同，如果你想给玩家制造意外之感，让他们遇到一种国外技术，你可能就不该将其设置于工业建筑中，因为外 国机器与其他机器并没有什么明显差别。你最还是将这种外国机器放置于一个旧车库中，这样才会吸引玩家的注意力。有时候，只有站在玩家的角度才能知道他们的预期是什么， 从而识别你的关卡中哪部分内容需要改进。
唯一可鉴别玩家技能水平的方法就是，先确定他们技能水平的中间值。玩家开始体验关卡时的中间技能水平，取决于他们玩之前关卡时的低级和高级水准线（游戏邦注：如果是第 一个关卡，这一数值就与他们之前所玩过的同类游戏为准）。这样你可以快速根据玩家完成之前关卡的技能来推断他们的最低水平——即最低水准线。而要确定最高水准线，你就 得从那些尚未玩过其他关卡的用户中收集反馈信息。这一点很难实现，但如果你是根据公司的测试成员，或者那些只为了免费玩游戏而参与测试的骨灰玩家的水准来获取这一数据 ，那么你的高水准线就会严重偏向资深玩家。而这些骨灰玩家不但技能高超，而且还善于在游戏过程中向他人取经，所以比起让他们在自己家中玩游戏，群体测试环境并不能试出 其真实水平。你最好能够辨别出表现最出色的玩家，以这类群体的技能水平为高级水准线，并以推理法确定低级水准线。这样就可以确立玩家接触关卡时的中间技能水平，并以这 两个极端试玩关卡，从而鉴别哪些部分应该调整难度。
每个玩家都有自己的“锦囊妙计”——即解决谜题或挑战的策略、战术。这些妙术包括战斗技巧，侦察方法，偏爱武器，盟军选择，目标选择，建设策略等。设计关卡时，你可以 假设这些玩家会使用自己的一些决策来打败你的关卡。但不能假设玩家已经知道某个特定策略的用法。要注意查看游戏早期的关卡，观察玩家是否已经掌握该策略。如果确实如此 ，那就可以放心使用这种设计，但不要过度使用该策略，因为这样会让你的关卡乏味无趣。如果玩家尚未掌握该策略，那就要慎用，最好不要让它成为你的关卡解决方案。
最好全面了解玩家进入你的关卡中所附带的兵力、武器、口语、技能等级等内容。设计师常会低估或高估玩家开始某个关卡时所带的装备。所以应该研究游戏之前关卡的情况 ，在 资产列表查看它们已采纳的内容（详见第9点）。还要查看游戏测试时的相关数值，估算玩家可能承担或建设的内容。然后据此平衡敌军兵力或其他挑战。
关卡设计师在一定程度上得是折磨玩家的“虐待狂”。你得乐于扮演这种坏人角色，从AI角度出发设计内容。这样你才能制作出更具现实感，更易为玩家所理解的对手。玩家通常 希望AI可以像人类一样采取行动，如果你编写了一个行为像人类的AI，这将有助于玩家成功制定策略，并深度沉浸在游戏中。这也会在玩家心中激发一点小小的恐惧感，因为他们 并没有料到游戏AI居然可以发现他们的脆弱。作为一个坏人，你得让玩家心生恐惧，并充分利用他们的弱点。这样才能让游戏充满挑战性、趣味性和满足感。
就确保关卡设计质量而言，没有什么比测试更靠谱了。虽然我将测试列为第19个原则，但实际上玩法测试是一个持续进行的过程。你在制作关卡过程中就得进行测试。如果你在早 期设计阶段就辨别出重要漏洞或失误，就能够节省下大量返工的时间。另外，许多关卡设计师想出更多提升关卡的念头时，也需要经常进行测试。记住，只有经过严格的测试，你 才能避免自己的关卡出现严重的漏洞，这样才不会在上司或同事面前丢脸。测试关卡也是关卡设计师工作的一部分。
身为关卡设计师最有成就感的一项活动就是看其他人玩你的关卡。此时你不但有机会看到他人的反应（游戏邦注：包括消极和积极反应），还能据此判断他们的体验和你的追求目 标有多大差距。你可以观察他们的玩法风格，看他们如何探索和发现不同技巧、谜题、陷阱和奖励。这有利于判断你的关卡对不了解情况的玩家来说究竟有多大难度。你可以由此 判断哪些环节太无趣或太难，并相应调整其难度。总有些玩家会有一些超乎寻常的举动，遇到这种情形可以向他们询问原因，要知道这些玩家的回答可能会为你提供一个提升游戏 的好主意。总之，观察玩家试玩关卡是你万不可错过的机会。
要始终铭记玩家测试者从来不会有错，尽管他们可能难以清楚解释自己的基本想法，或者提出一些有利于改进关卡的建议。对他们的建议要保留意见，因为他们并不一定是目标市 场的用户。有些测试者可能也不是你这类游戏的粉丝，或者他们对这类游戏已经非常熟悉，可能已经无法提供更有价值的难度调整建议。在调整关卡之前，你应该考虑更多测试者 的说法，这样才能找到这些反馈的共同点。如果只针对一名玩家的积极或消极反馈调整游戏，这可能会让其他玩家对你的关卡失去兴趣。
你投入的时间越多，关卡设计就会越完善。优秀与出色关卡的差别通常就体现在细节上，所以不要吝于投入时间。游戏这种电子媒介的好处在于你可以保存关卡的不同版本并对其 进行试验。你可以根据自己脑中的不同想法，或者测试者的反馈尝试不同想法。永远不要满足于自己的关卡设计，除非你已经体验到了自己最初想象中的那种乐趣。要多花时间去 想想关卡所缺内容，或者找找阻碍关卡实现那些终极体验的因素。只有你才能够让自己的关卡更上一层楼。
游戏设计是一门难以评判的技术，它无形并且不断发展，也没有专业的院校课程。而“面面俱到”设计师就是利用了这一特点来装腔作势，把自己哄抬为位高权重的人物。不幸的 是，这种人在我们行业中随处可见，并且他们总是对你的工作指手划脚。我提到这一点，是希望你不要成为这种人，因为这种人的害处在于其误导性，毕竟让人们觉得你不懂每个 玩家的需求真不是什么愉快的经历。
当你看到其他人的关卡设计，或者自己的早期关卡设计的纰漏时，你就会知道自己是什么时候形成了良好的设计直觉。本系列文章中的原则全部来自我多年制作游戏、犯错误以及 获得启发的个人直觉。而作为新人设计师的你也不可能回避错误。不过，如果你从这些经验之谈中学到知识，就能少走一些弯路。希望本系列设计理论和原则能为你的关卡设计事 来带来帮助。
Beginning Level Design, Part 1
This article is the first of a two-part series covering theories behind level design, establishing some rules for level creation. The intention is to aid those new to the field who want to design levels for pleasure or pursue a career in level design.
Level design is the data entry and layout portion of the game development cycle. A level is, for all intents and purposes, the same as a mission, stage, map or other venue of player interaction. As a level designer, you are chiefly responsible for the gameplay. This article will give you insight into developing good levels for any type of game, whether they are military missions for your horde of tanks, aerial encounters for a flight simulator, a dungeon for a role-playing game, a board for a puzzle game, or a map for a world conquest god-simulator.
I will present some theories behind level design, starting with an exploration of what good level design means. Then it delves into the non-electronic roots of computer game design from chess to GI Joe action figures, and how we can learn from their success. Finally it takes a thorough look into the theories behind storytelling and how we can apply them to level design.
What Good Level Design Means for Players
Players are the consumers who paid good money for your game or dedicated a half-hour of on-line time to download your custom level. If you are working or have desire to work for a developer or publisher, the consumers are those giving you your dream job. As any modern business school will tell you, creating an affordable product that satisfies the consumers is what it takes to make it in business. As a level designer, you must become very aware of what satisfies
the consumers if you want to be successful.
A player buys a game to escape from his or her reality. Good levels and hence good games will immerse the player and suspend their disbelief. From the moment the title screen comes up, you have their full attention. From that point on, they should see and do nothing that reminds them that they are anywhere but in the world you have them in.
You must furnish a setting and actors that meet the players’ expectations. That is, you need to design a map that not only looks like it could fit inside the world they are playing in, but contains elements that help to draw that reality in the players’ heads. A player’s sense of escapism and suspension of disbelief can be ruined by a variety of common errors. These include bugs such as graphics glitches or crashes, but from a design standpoint, these also include inappropriate content. For example, a McDonald’s Golden Arches on the skyline of a medieval town is obviously out of context. Similarly, if a player is told by a character to hit control-T on his keyboard to teleport, then it would remind him that he’s typing at a computer and not in some fantasy realm. Generally, to maintain the players’ sense of escapism all content should be appropriate to what would be seen, said or done in the game setting.
Challenge – Testing the Players’ Mettle
Players buy games to be challenged. If there is no challenge, they might as well be interacting with their word processor or spreadsheet software. Challenge should always come in the form of testing the players’ skills at the core gameplay. A shooter should test their aim and reflexes. A wargame should test their tactics. A strategy game should test their strategic sense. Some games successfully combine forms of gameplay to offer a variety of challenges, such as Command & Conquer, which has both planning/building and tactical gameplay.
Challenge comes from difficulty. The trick to good level design is to present challenges that are difficult enough to merit the players’ attention and make their heart or mind race, but not so difficult as to always leave them failing and disappointed. It’s a delicate balance based on what is perceived as the median player skill, and it is a variable constantly adjusted up until the game ships.
Like a good television show or book, the game must maintain a player’s interest. The introduction of conflict, the revelation of the setting or back-story, the acquisition of new assets, the display of new art, and the increase in difficulty must all be deliberately spaced to keep the player interested and looking forward to the next level.
One boring level can be the kiss of death to a game, especially if it’s one of the first few levels. Game reviewers and most players only give a game that much time before they praise or trash it. Good level designers have learned to be objective about their own creations and when asking themselves, “Is this fun?” The hard part for many designers is that what they find fun may not be what the target market finds fun. As a level designer you need to understand the core gameplay, which is part of the vision expressed by the producers and lead designers. You need to try to understand and become that target market.
Something that helps designers tremendously is to play competitors’ games. Often producers and lead designers will name successful games that they are trying to emulate. Play and study those titles. Make sure your levels entertain, thrill and excite you as well or better than the competition’s levels.
Frustration can also kill a game. Players stop being entertained when they encounter technical problems like slowdowns or graphics glitches. The level designer can avoid a lot of these bugs if they pay attention to technical limitations and to the instructions of the artists on how to place the art. Designers can, of course, create their very own frustrating bugs, like broken AI scripts or door triggers that never trigger, or missions that don’t always end when they are supposed to. Even worse, designers can create what are commonly called “show stoppers”. Show stoppers are unbeatable missions or
unsolvable challenges or unavoidable traps that frustrate players. A good level designer can spot these problems and resolve them with careful and rigorous play testing before consumers get their hands on it.
Player’s don’t like playing, or indeed, buying, the same game twice. Of course, like Star Trek fans and readers of the prodigious Gor science fiction series, some players will continuously buy into the same formulae or even the same game with just slight variations in plot, setting, characters and art. The same can be said for level designs – people don’t like playing the same level twice. Not only does it ruin the entertainment value, it also fails to spark the imagination. It’s therefore incredibly important that levels introduce some variation in the plot, challenge, setting, and characters (i.e. the enemies).
The Roots Of Computer Game Design
Computer game design has its roots in earlier forms of entertainment that predate the joystick and personal computer. Board games, paper and dice games, toys, and the ancient art of storytelling all have methods that continue to capture the human imagination and joy. Level designers can learn by studying these methods and understanding what each form has contributed to the art of computer game design.
Board, Paper & Dice Games
Games predate civilization. Some of our oldest games still survive to this day, like mangala (or stones), dice, checkers, tic-tac-toe and chess. What gives them their lasting power? What can we gain from them as designers of complex computer games? Simplicity and elegance.
These games keep the gameplay and the rules simple. Almost anyone can grasp them and quickly perceive the strategies and skills necessary to achieve victory. Elegance comes from years of refining the rules and components to maximize and balance the gameplay, and provides lasting entertainment value.
Simplicity and elegance should be your goal in level design. So many designers (I being one of them), have fallen into the trap of creating complex games and levels that make it difficult for players to grasp the rules, objectives, strategies and indeed the fun. Designers often fail to play test their level enough to uncover any unbalancing factors and make improvements. So keep it simple, and submit your level to a lot of play testing so you can polish it.
There’s a lot more that can be learned from non-computer games, such as the value of symbolism, statistics, and role-playing, but this goes beyond the scope of level design and should be left for a future article on computer game design.
Toys – Train Sets, GI Joe and Barbie
Toys have always fascinated children of all ages. Train sets are the ultimate interactive toy. You can be designer, builder, painter, passenger and engineer (or even God if you like creating natural disasters). GI Joe and Barbie both have interactive features such as movable limbs and changeable clothing and accessories and even vehicles and play sets. They cash in on kids’ aspirations and dreams. One of my favorite quotes from a paper game industry veteran and
former toy maker is, “Who knew that a doll with tits would sell so well?”
The more you can interact with toys and the closer they get to peoples’ aspirations and dreams, the more they are appreciated. The same can be said about level design. Think of your level as a train set. Consider how can you make it interactive with all the bells and whistles and other special effects. Think about how you are portraying the player and what you are having them do. Let them feel like a general chasing down a retreating tank corps or a squad leader
breaking ranks and charging a hill. Let them feel like a deer hunter chasing down quarry. (Think about the few million units that the latter type game has recently sold.)
The most ancient form of entertainment, storytelling, has riveted mankind since the spoken word. Stories of adventure, triumph and disaster all pull at our hearts. They take us through a ride in someone else’s skin and often challenge our own convictions, illuminate our soul, or simply lighten our spirits. As game designers, you’ll concentrate on the latter.
Thesis, Antithesis and Synthesis
Stories essentially come in three parts:
The thesis, which is the introduction to the setting, the characters and the hero
The antithesis, which is where the conflict and villains are introduced and is what amounts to the majority of the story
Synthesis, where there is some form of resolution, be it triumphant or tragic.
We can see this model followed precisely in the three-act play. We see it in film scripts, and indeed, all the forms of entertainment that evolved from storytelling.
In games, your player is easily identifiable as the hero, and the game as a whole can be seen as one story, where each level is a portion of the antithesis, the interplay and conflict between the hero and the villain. The depth of the thesis may be limited to a cinematic that explains who you are and what the setting is. Some sequels will gloss over the thesis and jump right into the conflict, assuming you’ve played the prequel. The synthesis is everything that happens at the end of the game. You’ve either won or you’ve lost, and you may see a cinematic or read a few lines of narrative before you see the credits. Like Hollywood movies, the synthesis is never very long.
Understanding and Developing the Thesis in Level Design
Each level in itself is its own story. As level designers, you set up the thesis by preparing the initial situation. You position the player and perhaps indicate his initial arsenal or force or set of spells or pieces. You render the setting with your map or your puzzle board. The setting and the situation can change over the course of the level as portions of the level are revealed to the player or new characters or other elements are introduced such as power-ups or new player or enemy forces. As games are interactive, you have to be very conscious about every possible situation a player can be in at any given time or place over the course of the playing the level.
Each encounter has its thesis – that moment right before the battle when your fear, adrenaline, and anticipation kick in, and how observant you are of the situation right then and there will influence the fight. For example, a player may stumble into a rats’ nest of bogies and realize at that moment right before the fighting begins, that just next to the head rat is a large fuel tank within easy range of a missile salvo. But we cannot assume that players will always do the same thing and come from the same direction. A level designer has to plan for that and reward that behavior. Using the previous example, a player may come from a sneaky route from the opposite direction and see the fuel tank in his line of sight before he sees any of the bogies. Does he blow it up now to be cautious or walk on through? What if the enemy chooses to use it against him? To complicate matters, let’s say that there is a nice but destructible power-up right next to the fuel tank. What should the player do? In this situation, you don’t have to make it a single choice. Indeed, you
really don’t want to make it a single choice. As part of the thesis, you need only present options to the player and he’ll decide what he wants to do.
Introducing and Refining the Antithesis in Your Level
The antithesis is where the players interact with your level. By positioning enemy forces and scripting their behavior, or by setting the timing and speed of the bugs they have to zap or the puzzle pieces they have to place, you are creating conflict. This should be where the core gameplay of your level is. If it ’s not, then you’ll have a level that requires too much planning in the thesis stage. In other words, if the player doesn’t plan things out right from the beginning then the level is over before it began, and if the player plans correctly then there’s not much to it. Most people, with the exception of strategy wargamers, hate this kind of level. Players need the ability to resolve conflict as it arises – you can’t assume they are omniscient or psychic. A common mistake designers make is presenting challenges that are absolutely unbeatable unless you’ve played the level before and know what to expect. It is essential that players be capable of resolving the conflict and tackling challenges the first time they encounter them.
The antithesis is where you present the knife’s edge to the player. It divides the good players from the bad, the experts from the novices, and the dedicated from the dilettantes. Ideally there is more than just one victory to be won, because indeed the division of skills is not black and white. If only experts can beat your level, then you’ve lost 90% of your market and your game won’t sell well. Likewise, if any mediocre player can beat your level and reap all the rewards, then it’s not satisfying to 50% of the players who weren’t challenged. But if your level had a satisfying victory for the mediocre
players and optional challenges to entice and reward the good and expert players, then you’re presenting multiple edges to challenge and satisfy a diverse group of players.
Synthesis – Making Your Levels End in a Satisfying Tone
Synthesis is the result of an encounter or the entire level. It’s a moment of reflection for players to evaluate the encounter or level and what they got out of it. Whether players fail or succeed, they should be able to recognize why and how they might do better next time. This keeps them interested in trying again or just replaying for a better score or reward.
Victory or failure should be obvious. Players should understand why they lost. Victories should come as the direct result of the final acts of the player, not as the result of something the player does midway through the level (the latter tends to make players bored). Ending the mission on a big, satisfying note leaves a player feeling good.
Stories maintain your interest by presenting worthwhile content. People don’t buy a book or see a movie just to hear characters talk about the weather, unless the weather itself is the villain (as in disaster movies like Twister). All the details that a well-written story contains are those that render the setting, develop the characters or move the plot. While books can get away with including an awful lot of detail, films cannot. Films are aimed at short- attention span people who want to experience the whole story in 90 minutes or less. Films try to focus on the most important details and these usually are the ones involving character interaction.
The same can be said with level design, except that you have an even shorter amount of time to tell your story. As a result, you must focus even harder on character interaction details, especially those that involve the player. Everything the player sees or does must further the story. All of the players’ accomplishments should move them toward the completion of the story or pull them further into the conflict with the villain. As the game is played, players should discover more about themselves and their opponents. This can be achieved when players develop new talents, find new weapons or upgrades, gain insight into strategy, or encounter new enemy tactics and new enemy types. All of these suggestions may sound obvious to you, but you would be surprised how often designers make the mistake of spending a lot of time working on setting details that are rarely, if ever, seen by the players.
Most gamers have a short attention span, especially those who play console games. They don’t have as much patience with minor details and game subtleties. If you present them with too much detail, or if your gameplay hinges on the player understanding the significance of minor details (like a single dialogue message), then you will lose them. It’s very hard for non-computer game designers and RPG designers to not populate levels with all sorts of irrelevant content. Often this focus on details works to the detriment of gameplay. If you’re not making an RPG, then you have to understand that the finer details of the story come second in level design.
Spending a lot of time working on non-interactive details can be a waste of time and resources, although it’s important to put some effort into it because the player will pay some attention to it. For example, it’s ludicrous to spend a day reating the details of a farm that a player will pass in three seconds on his way to a tank battle. It’s better to just take a minute to sprinkle a few objects that give the player the feel of a farm, like a farmhouse, barn, silo and a few cows. Even if you have all the time in the world to create all sorts of non-interactive details, it’s still not a good idea. Players get distracted and suffer sensory overload from too many details. They also can get frustrated as they try in vain to interact with non-interactive details.
It would be even better to make all the details of the setting interactive somehow. Duke Nukem did an excellent job of this. Even the toilets had some purpose, if only to give a little humor. The bar had a working pool table and the arcade had a Duke Nukem machine that prompted you to say, “Hmm, I don’t have time to play with myself.” The extra effort it took was well worth it. The interactive setting created a great allure and set this game apart from all the other Doom clones.
Verisimilitude – When to Stay within the Realm of Probability
Verisimilitude is the technical term used by writers to describe the readers’ acceptance of the facts and events within the story. When the story steps out of the realm of probability, the readers get frustrated.
Works of fiction must suspend the readers’ disbelief if they want to keep the reader. Readers are only willing to accept so much. How much varies with the reader, which often separates the readers of classical fiction and literature from those of fantasy and science fiction.
Computer games have it easy because their target market is much more likely to be readers of science fiction and fantasy. Though the so-called “break through” titles which establish new genres of games often go beyond the sci-fi and fantasy market. Titles like Sim City, Tetris, Civilization, Deer Hunter, and sports games of all types don’t make any grand leaps of logic or fantasy, and they entice players who’ve never shot a single alien. Even so, sci-fi and fantasy oriented games are the vast majority of games made today.
So assuming you are working on a sci-fi or fantasy game, you do have certain latitude (or indeed, a certain obligation) to extend the realm of possibility for the players. But it’s important to know when and where and how far to stretch reality. Players like the realm of possibility extended more for themselves than for other characters. While this seems one-sided, it’s what players want. Players feel cheated if the AI enemy kicks their ass by doing something amazing and beyond their capabilities. They prefer to have their butt kicked by an opponent who’s limited to what they can do. Then they can at
least be impressed and comprehend that it is just a skill issue.
On the other hand, players enjoy pulling off amazing feats beyond the scope of the AI capabilities and romping the AI for a spell. So give the players what they want. Let them enjoy themselves with a little god-like power. But be aware that giving that ability to players all the time can lead to a dull, unchallenging game. The trick is to balance it so that players don’t always have that edge, either by limiting the use of the ability or by countering it with enemy powers. In an ideal level, the players will face overwhelming odds and overcome them by leaping beyond the apparent realm of possibility. That way they can feel like they have done the impossible and that they’re real heroes.
The lead designer should describe the boundaries of the game reality to level designers. This will give you a concept of how the fantasy world works and what you can do.
Additionally, this reality often evolves as the core gameplay is balanced and new ideas are introduced, because preconceptions often fail when the game is complete enough to play. The so-called “fun factor” outweighs the unsubstantiated premise every time. However the boundary is set, it should be maintained throughout the game. Having one level that distorts players’ sense of the game’s reality and their own limitations can break the verisimilitude and
potentially ruin the game.
Armed with this understanding of level design theories, you can begin creating your own levels with greater confidence and a clearer insight into what will make them successful. Next week I’ll present a set of rules for level design and offer advice to aspiring professionals.
This article is the second of a two-part series that covers theories behind level design and suggests a set of design rules. The intention is to aid gamers who want to design levels for pleasure or pursue a career in level design.
Level design is the data entry and layout portion of the computer game development cycle. A level is, for all intents and purposes, the same as a mission, stage, map or other venue of interaction that the player in. As a level designer, you are the presenter of all the labors of the programmers and artists and chiefly responsible for what most believe to be the most important part of a game, the game play. This article will give you insight into developing good levels for any type of game, whether they are military missions for your horde of tanks, aerial encounters for a flight simulator, a dungeon for a role- playing game, a board for a puzzle game, or a map for a world conquest god-sim.
In last week’s article, I discussed the theories behind good level design. This article formulates a set of rules for level design and offers some parting advice to aspiring professionals.
20 Rules to Design By
1) Maintain the vision.
The “vision” is the core idea of the game design. It’s what the producer and lead designer express when selling the game and what they impart in the so- called “concept document.” It’s also what they expect you, the level designer, to understand when building your level. It’s very important that this vision is communicated to you very clearly. If the producer and lead designer have not expressed to you what they want, then you need to coax it out of them. It will save you a lot of time and grief in the end.
When designing your level, you must maintain the game designers’ vision. If you deviate from it you risk rejection. While designers cannot always describe specifically how to accomplish their vision, you must try to figure out ways to truly express the vision they are looking for. If you cannot maintain and express the vision, then either the vision is imprecise or unpractical, the design tools and palette are insufficient to the task, or your skills are not up to it. In any case, you need to address those problems if you hope to construct a successful level in a timely manner.
2) Learn the design palette.
One of the first things you need to establish before you begin your machinations is the design palette. The design palette includes all of the art and game play elements at your disposal.
Knowing what elements you have to work with and how you are to use them is imperative for good level design. Get instructions from the artists (if you can) and play around with the art in a test level to establish the look and feel you want. Talk to the programmers and find out what the technical requirements and limitations are, like what data arameters need to be set, what scripts need to be written, and what to do in order to keep within memory and processing-time constraints.
The design palette goes beyond art and code as well. It includes all the player and enemy forces and their behaviors, game play objects such as power-ups, switches and weapons, buildings that perform a game function such as turrets, power stations and walls, and game play puzzles and possible solutions (the so-called “bag of tricks”). Ideally you will have time to learn how to place all of these elements with your design tools (such as an editor) and play with them before you begin a real level.
The lead designer, in order to save elements for other levels, may restrict your design palette. It’s up to you to figure out how you can work with what you have in a way that will maintain the vision of the lead designer and producers. If you cannot, ask them for advice. They may provide some guidance or use their power to give you some more design lements.
Sometimes it takes a fresh look and imaginative effort to use design elements to their maximum potential. When you find you don’t have enough design elements to fill a level, experiment with untried combinations and layouts. You may stumble upon some new game play puzzle that you can add to your design palette.
For example, you may run out of ideas for using turrets, and after considering your options, you might discover a that particular combination of fixed turrets and enemies in a certain placement presents a balking defense to the player unless he takes advantage of ranged weaponry or provokes the enemies to pursue him beyond the range of the turrets. Once you’ve introduced this scenario into your level, the design of the subsequent levels could include that particular puzzle.
One grave mistake that all designers make at some point is to create mazes. Why is that a mistake? Mazes are one of the first forms of puzzles introduced in computer games. It’s old now.
Because all it takes to make a maze is placing walls or other terrain that blocks movement, it’s the easiest game play to create. It is sort of a last resort when you are fresh out of game play elements and ideas. When you get to this point, stop. Try to improve your design palette by coming up with new ways to use existing elements or by pushing the game designer to create more.
Pushing for more design elements is a good way to earn both respect and disdain from coworkers. Unfortunately, it’s your job. But make sure you do present your good ideas to the lead designer. If an idea has merit, he’ll try to get it in the schedule. Just remember that implementing ideas often involves the commitment of both art and coding resources, so don’t be surprised to hear “no” for an answer. The best ideas are often the ones that reuse existing art and involve little to no coding. If you can make it all work with your own scripts, that’s even better. When development reaches the alpha stage of the project (when all the coding and most of the art should be done), don’t expect any new game elements.
I’ve seen producers make the time for particularly good ideas as a project nears alpha, but it usually comes at the expense of the artists’ and programmers ’ sleep. That’s the reason why pushing for more design elments can also earn you the disdain of coworkers. Try to understand that new ideas take time to evaluate and develop. Don’t make a jerk out of yourself by getting insistent. Instead, keep those ideas on the back burner for the data disk or the sequel.
3) Have fun while you work – it will show.
The joy you experience when conceiving and implementing your level will convey to the person playing it. Sure, there will be frustration when deadlines loom and level editors crash at the worst possible time. There will be game bugs and frame-rate issues that will force you to rework levels and strip out what took hours to place. But it’s easy to ignore all of that when you are doing something that you know is going to be fun. Remember, there are thousands of people who will play your level and never know what you went through, but they will certainly feel the joy that you put into it.
4) A level will only ever be as good as you imagine it.
A great sculptor doesn’t begin chiseling a block of stone until he envisions in his mind what the completed sculpture will look like. The same is true with level design: there’s no point in beginning to design your map if you can’t truly see what you’re working towards. You might have a vague idea about what you are trying to make, but to start designing away without a clear vision can lead to a lot of wasted time and effort. Bosses aren’t really keen on wasted productivity, so try to get your level nearly right the first time, so you don’t have to toss it all out and start afresh.
This isn’t to say that you should leave some time to experiment, but the core idea of the game play for your level should stand on its own. It’s also best to choose a core idea that leaves a lot of room for a variety of game play. When you implement the level, establish the core idea with broad strokes, and just make it work. With that done, decide if the idea has merit and whether you want to go further with the level. If so, fill in the fine details and experiment with subtle game play details. Often it’s the subtler elements and details that make the difference between a good level and a great one.
5) If there’s no difference, what’s the point?
Having multiple routes to the same goal is a good way of giving players choices and a sense of freedom while still ensuring they end up at the same point. Yet, if each choice exposes the players to the same types of enemies, the same rewards, and the same risks and costs, then players will only get frustrated and bored when they discover that there is essentially no difference. When presenting choices to the players, there should always be some non-aesthetic
difference in game play. The difference might be the introduction of different challenges, a sneakier route, traps, hidden power-ups, higher elevation for better map revelation, or just better tactical position. It’s important not to present the same choices to players multiple times. Otherwise, what’s the point in offering them a choice at all?
6) Cater to different playing styles and abilities.
When presenting options, challenges or puzzles to players, try to offer multiple solutions that cater to different player styles and abilities. Some players play conservatively, while others like to play it risky. Some people are cautious and like to reveal as much of the level as possible before proceeding into conflict, while others just jump right in with guns blazing. Some take the straightforward route, while others look for the sneaky way. Player styles may be completely unique to your game or type of game, and you should try to identify those modes of play early on. Make sure you design your level with all the different play styles in mind, so that everyone has fun.
Don’t assume that every player is going to play your level the same way. Be conscious of how difficult it can be if a player doesn’t figure out alternate or ultimate solutions to your level. Players’ abilities at handling conflict and mastering the game play vary, and people learn at different rates. Offer easier but less rewarding solutions to your level, but make sure the players know what they’re missing if they opt for the easy solution. This encourages them to replay your level and try harder next time.
7) Reward player imagination and efforts.
Players like to experiment and explore. The more solutions, secrets, alternate paths, and so on, that you provide in your level, the more satisfied players will be. It’s a great feeling when, as a player, you come up with a not-so-obvious solution that succeeds. Remember that players almost always go off the main route hoping to find shortcuts, hidden caches of goodies, or other unexpected items. When designing a level, try to think about what players may want to try, and give that to them. When they say, “What if…?” your level should respond with, “Yes, you can.”
Nothing is worse than designing what appears to the player to be a challenge, alternate solution, route or secret place that offers no reward. Players try to interact with everything, and when the interaction is pointless, frustration results. Interactive game play objects (e.g., moveable crates or exploding canisters) which serve no purpose tend to frustrate players.
Players may try for minutes, or even hours, to figure out what they are suppose to do with these objects. Don’t let players down in this regard.
For example, in a Quake or Unreal level, imagine if a player saw some rafters just at the edge of his jump range from a narrow ledge and said to himself, “Ah, a challenge. I wonder what’s up there.” If those rafters served no purpose in the game, the player might spend an hour trying to jump out onto the first rafter, only to repeatedly fail in his efforts. The player might quit and feel let down, or even worse, this might pique his curiosity even more, and his resolve to get out there might harden. If he ultimately made it and realized that there was nothing up there, he’d get annoyed both at himself for wasting time playing the damn level, and at the level designer. So, when designing and testing your levels, look out for these “black holes of interaction” and get rid of them. Or, better yet, give them purpose by rewarding players who expend the effort to figure them out.
8) Pay attention to level pacing.
Pacing is the introduction of conflict and tension, plus what some like to call the “adrenaline rush.” This follows closely the Thesis-Antithesis-Synthesis model that we know from stories and films. The tension builds as the player (the thesis) interacts with the antithesis, and it crescendos right before the synthesis, where the reader, watcher or player breathes a sigh of relief. (Or, alternatively, the player may get grim from his failure and restart the level, pretending it never happened.)
Because games are interactive, forcing a certain pace into the level can be difficult. What if the players don’t do what you want them to do? What if they take too much time? What if it’s too easy and unexciting when it’s played slow or too intense if played too fast? There are some things you can do to remedy this without taking all the interactivity out of it.
Time limits add tension that’s immediately perceptible by the player. A time limit can force a player to move more rapidly, or adopt tactics that you want him to use, such as splitting forces to achieve multiple objectives. You can put in an artificial time limit – like a mission clock, a puzzle-solver clock, or a turn time limit. You can institute a realistic time limit into a level, like the time it takes a certain enemy or ally unit to move to its exit point, or the time before enemy reinforcements arrive to overwhelm the player.
Controlling the movement speed or distance a player may traverse in a turn drastically affects game play pacing. While you cannot just arbitrarily change this in your level unless you are doing a puzzle game like Tetris, there are other ways you can play with speed. Often terrain affects movement speed, such as swampy ground that slows you down, a highway that permits you to speed up, or an obstructed and twisty route that slows your progress. Giving units different movement speeds and/or movement restrictions can slow or speed up the players, if they have to travel with that unit. For example, giving the player a slow, heavy tank will encourage his forces to slow down to the tank’s movement speed, and making the player guard a fast-moving ATV will force him to speed up.
Pacing can also be set by the enemies’ speed. For example, in a POV shooter, the player may have to chase down an elusive ninja who’s trying to escape from a lord’s castle. In a side-scroller, the boss monster can be made faster or slower. Whenever a player has to move and act faster or slower than he’s accustomed to, it builds tension. By using these various methods to manage tension through movement speed, you gain precise control over the pacing in your level.
9) Reveal assets carefully.
Keeping the player interested in the game requires careful asset revelation. Assets are the game’s eye candy, such as terrain objects, enemy and friendly units, upgrades, puzzles, and so on. All but the simplest games try to reveal these assets gradually to players, so as not to overload them on the first level, and to keep them interested in going on to the next level.
The lead designer will usually have guidelines for what new assets your level will introduce. Try to make these new assets a centerpiece to your level, somehow associated with the core game play. Their introduction should be dramatic or significant, and ought to portray the uniqueness of the asset.
For example, if you are introducing a new power-up that makes the player invisible, then make that invisibility a pivotal part of the solution to the level. If you are introducing a new enemy that flies, set up an encounter where this creature alone attacks the player in an environment that demonstrates the benefit of flying. If you are introducing a scattergun, make the gun available somewhere in the middle of the encounter with the flying enemy, so the player can see the dramatic difference in the effectiveness between his rifle and the scattergun against flyers.
The position of assets within the level is extremely important. Positioning power-ups, booty, and other loot – commonly called “gimmes” – establishes goals for players to move towards.
Gimmes are often the reward for the challenges you put between them and the player. Careful spacing of enemy encounters and game play objects, such as turrets, bridges, fuel drums, and so on, keeps the player interested in exploring and completing the entire level. A lull in the introduction of assets can encourage the player to turn the game off.
A good example of careful asset revelation within a level is shown in Heroes of Might and Magic II. At every turn, your heroes reveal a little more terrain and more assets to investigate, acquire or conquer. This revelation is what some call an “event horizon,” because it triggers and inspires players. New assets that appear on the event horizon keep players interested.
Unfortunately, an example of bad asset revelation can be seen in the same game. Heroes of Might and Magic II sacrificed its diversity of assets to make an individual level interesting, but in so doing, nothing new was left to be revealed in subsequent levels. With nothing new to reveal in later levels, the designers merely tinkered with the quantity and alliances of enemy players.
This scenario raises a very good question: Is it okay for a level designer to ignore the other levels in a game and use any and all of the assets he wants in order to make his level better? The answer is no. If the natural progression of asset revelation from level to level gets broken by one particular level, then the other levels seem weak in contrast. It also forces other designers on the project to redo their levels, and that causes havoc and wastes time. The next thing you know, that one level has set a precedent that the lead designer did not intend. Having just finished a game project on which this happened, I can vouch for how much a level that breaks the asset revlation can screw everything up.
10) Challenge the player.
Your job as level designer is to challenge the player. A level isn’t truly satisfying unless victory is at times uncertain. So you have to present challenges to players that really test their mettle and make them uncertain of their victory. When doing so, you have to cater to different player abilities (see rule #6) and to increasingly skilled and equipped players. Where your level is positioned in the game timeline or “level progression” should indicate how difficult it needs to be. In the first few levels, players learn how to play the game, so these levels should be a little forgiving. Levels at the end
should be the most difficult to coincide with the increased skill and player resources.
There will be times when you find that your level, although it plays really well, doesn’t quite fit into the progression. It may make the levels before it or after it seem too easy or too hard. There are a number of solutions to this problem.
•You can scale up or down the difficulty in your level without grossly changing the game play or the fun factor.
•You can ask to reposition your level in the game. This isn’t always an option if you have a tight story line, however.
•You can make your level a sort of “change-of-pace” level. Change-of-pace levels are usually easier than the previous level but subject the player to an unusual limitation, so they remain difficult in the fact that the player is using untested skills. An example is the “Tanya” mission in Command & Conquer: Red Alert, where you no longer control a large number of tanks and troops, but instead one super “Rambo” soldier.
In some games, levels are grouped together into modules, like missions within an operation, floors in a dungeon, or regions on a planet. While the subsequent modules should generally increase in difficulty, the last level within a module may be more difficult than the first level in the next module. This is because there’s a natural pause and release of tension that players experience when they’ve achieved very important objectives in the last level of a module. Players are not ready to jump right into the intensity again and often appreciate an easier mission to catch their breath.
Although it is easier said than done, the ability to create unique game elements is very rare these days. Yet at least in level design, you have a chance to combine elements in new ways
and tell different stories. And besides, no good game completely ignores its predecessors or the competition, and you shouldn’t, either. Sometimes it’s useful to play the competition in order to identify aspects of your level that you think are lacking, or spot where your level is better. You often come up with new ideas to add to your design palette. You may find that your level idea has never been implemented before, or you may get inspired to try something new. A level doesn’t have to be completely original to have uniqueness. Your individual tastes will emerge in your design, and that alone will make it unique. Hopefully, the differences will fill a void in your own and other players’ experiences. However you do it, uniqueness sets your level and your game apart from the others, ideally in a positive way.
12) If the player didn’t see it, it didn’t happen.
Don’t assume all players will read dialogue or mission descriptions, and don’t rely on their observation skills, powers of precognition, or capacity for logical deduction to understand what is going on in the level and what they should do. Players must see what is happening to understand it. The old proverb “a picture is worth a thousand words” is entirely accurate in level design. To a certain extent, you are bound by the art and animations, but a lot can be accomplished with observable AI behavior, enemy and object placement and settings, and the revelation of terrain.
For example, a mission from the recent hit MechCommander starts off with you on the opposite side of a meandering river from an enemy convoy. You have to destroy the convoy before it reaches the exit point by racing to convenient jumping or crossing points, as the mission description tells you. In its infancy, the level started the player far from the river and the enemy. If you had not read the mission description or bothered to look at the tactical map or mission
objectives, you would not have a clue as to your objectives, and you certainly wouldn’t perceive any sort of urgency. You wouldn’t know what the enemy was up to, or why. By the time you got through exploring and fighting a couple of battles, you’d lose without any idea as to why you lost. Maybe you’d try again and read the mission description, or perhaps you would just turn the game off.
The level was changed to put the river and the starting point of the enemy convoy within your line of sight as the level started. Right away, you see your target and perceive the problem of them being on the opposite side of the river. Soon after you exchange missile and laser fire across the river, you realize that the convoy will not slow down to attack you, and you find yourself in a race to cross the river and cut them down before they get away. The whole objective and core game play of the mission is revealed in seconds without any words or confusion – just with insightful revelation, ositioning, and enemy behavior.
13) See through the player’s eyes.
Players usually watch most closely those objects that appear on a level’s “event horizon.” The event horizon is where new terrain is revealed and where enemies are engaging the player.
Changes in the event horizon often trigger a reaction from players or influence their decisions, and changes elsewhere may not get noticed immediately.
For instance, if an enemy unit suddenly appeared in the middle of previously revealed terrain, it may not attract the player’s attention, at least until a blip appeared on the radar or the new unit attacked one of the player’s buildings. However, if the enemy unit appeared where new terrain was being revealed, it’s likely that it would be noticed right away. Likewise, a building isn’t really looked at except when it’s initially revealed.
While some players spend time examining previously revealed terrain, most people do not, and it becomes even less likely when the game takes place within a 3D environment. Players usually only observe what is in the “here and now,” and you should put yourself in their position to ensure that you don’t put imperceptible events in your level.
14) Fulfill player expectations.
Players will have certain expectations about your level based on what they may have already seen or been told. While it is fun and challenging for a player to experience the unexpected, you have to be aware of their initial expectations. This makes it easier for you to ensure that you are either meeting those expectations, surpassing them, or tossing them out altogether.
Players’ expectations can change throughout a level as you feed them more information. If you build up certain expectations and fail to follow through on them, the level can seem confusing or barren. If you elect to surprise the players by tossing out their expectations and revealing the unexpected, be sure it ’s important for your level, because the players will certainly perceive it as important. For example, it you tell the player that they are in an industrial
building and they don’t find any industrial equipment, they’ll get confused. They’ll wonder if they are in the right building or if they’ve missed any floors. Unless it’s important to the plot to surprise the player, you should either change the mission description or insert a few industrial machines into the level. Likewise, if you want to surprise the player with the existence of alien technology, you probably wouldn’t want to put it in an industrial building, because alien machines wouldn’t necessarily look much different from other machinery. You would be better off putting alien machine in the cellar of an old barn, where it would really grab the player’s attention. Sometimes, it’s only by taking the player’s perspective that you can perceive their expectations and identify aspects of your level that need to be improved.
15) Balance the difficulty for the median skill level.
Players of varying skill levels will play your game. While you can try playing your level as a bad player and again as a good player, you will probably not draw any significant conclusions about your level in this way. You’ll probably just conclude that “players who are bad should expect to lose.” The problem is that “good” and “bad” are vague terms.
The only way to identify what skills players will really have when they begin your level is to determine their median skill level. The median skill level of a player starting your level can be determined by using low- and high-water marks that previous levels have established (or, if it’s the first level, from previously played games in the same genre). You can quickly deduce what minimum skills a player has based on what it took to complete the previous levels – that’s your low-water mark. To determine the high-water mark, you have to gather feedback from people who haven’t played any level beyond yours. This can be difficult, however – if you are basing the high-water mark on the abilities of individuals in your test department or the extreme game geeks that show up to the focus groups for a free game and pizza, your high-water mark might be skewed too far towards the extremely talented players. These hardcore players are not only talented game players in their own right, they also tend to learn from one another while playing, so no one is ever going to play as badly in a group testing environment as they would if they were playing the game by themselves at home. You’re better off identifying the best player, setting that person’s skill level as the high-water mark, and using deductive reasoning to determine the low-water mark. This establishes the median skill of players approaching your level, and with this knowledge, you can play test the level at both extremes and identify where it needs to be made easier or harder.
16) Know the players’ bag of tricks.
Each player has his own “bag of tricks” – strategies and tactics for solving puzzles or challenges that are put before him. This bag of tricks includes battle tactics, scouting methods, preferred armament, their choice of allied forces, their choice of targets, their construction strategies, and so on. When designing a level, you can assume that the player will use some of the tricks from his bag to beat your level. However, don’t assume that a player knows a one particular trick yet. Look at the earlier levels in your game and see if players have been taught the trick yet. If they have, feel free to use it, but be careful not to rely on an overused trick, as it makes your level boring. If players have not been taught the trick yet, then be careful not to base your level’s solution on its use.
17) Learn what players may bring to the fray.
Have a thorough understanding of what players bring with themselves to your level, in terms of forces, weapons, spells, skill ratings, and so on. It’s not uncommon for designers to underestimate or overestimate what players will be equipped to do as they begin the level. Study the previous levels in your game. Look at the asset revelation schedule (see rule #9). Examine play testing statistics. Estimate what players may be able to afford or build. Then balance the enemy forces and other challenges accordingly.
As the game evolves over the course of time, keep an eye on the design of previous levels and make sure that they don’t change significantly – that can throw off the balance of your level or spoil your core game play. For example, if a designer working on the level prior to yours arbitrarily threw in a jet- pack, and you had already created a treacherous, 20-foot wide river to coax the player into a cool bridge encounter, it would ruin your whole level.
Be a watchdog over the design of other levels, because it will protect the integrity of your level. Worship the asset revelation schedule so that you don’t ruin someone else’s level, and nobody can spoil yours.
18) Be the adversary.
To a certain extent you have to be sadistic to the players. You should enjoy being the adversary, and think from the AI’s perspective. This will help you make much more realistic opponents that a player can understand. Players naturally put a human face on the AI, and so they expect the AI to behave like a human. When you script the AI to behave in a human fashion, it helps players successfully strategize and often draws them deeper into the game. It also evokes a little fear in players, as they don’t expect a game AI to recognize their weaknesses. As the adversary, you need to provoke fear in players and prey on their weaknesses. It’s what makes the game more challenging, fun and fulfilling.
19) Play test, play test, and play test some more.
Nothing surpasses play testing when it comes to ensuring quality level design. Although I’ve listed it as the19th rule, play-testing should be an ongoing process. You need to test your levels as you make them. It will save you a lot of time reworking your level if you can identify a significant bug or flaw in your thinking early in the design process. Plus, play testing is often where many level designers come up with some of their best improvements to levels. And don’t forget that only through rigorous play-testing you can spare yourself the embarrassment of your boss or your coworkers finding some really heinous and obvious bugs in your level. Testing your level is part of your job.
One of the most rewarding activities in level design is watching other people play your level. Not only do you get an opportunity to see their reactions (both positive and negative), but you can gauge how close they come to the experience you strove for. You can observe their play styles, see how they explore and discover the various tricks, puzzles, traps and rewards. It helps you see how difficult your level is to people who don’t already know the solutions and don’t necessarily have your play skills. You can identify where your level is too boring or difficult, observe solutions to puzzles that you didn’t expect and thereby make them easier, or harder. There’s always a player who will do the unexpected, and when you come across this situation, don’t be afraid to ask them questions like, “Why did you go there?” The player may provide you with a great idea for improving your level. Watching a player test your level is definitely an opportunity you should never pass up.
Always remember that play-testers are never wrong, though they may not be able to clearly explain the basis for their opinions or offer good suggestions for improving your level. Take their advice with a grain of salt, because they are not always the target market or the target skill level. Some of your testers may not be big fans of your type of game, or they might have played the game so much that they’re no longer good sources of advice when it comes to the game ’s difficulty. You should get input from as many play testers as you can before you change your level, so that you can see if there’s consensus in the feedback. Reacting to only one player’s response, whether positive or negative, can spoil your level for the other players.
20) Take the time to make it better.
The more time you spend working on a level, the better it can get. It’s often the subtler details that separate a good level from a great one, so take some time to put them in. It’s one of the finer pleasures of the level designer’s job to perfect a setting or the choreography of a battle. The beauty of the electronic medium is that you can save different copies of your level and experiment with them. Try out different ideas from your own twisted mind or based on feedback from play testers. Don’t ever be content with your level until you’ve experienced the fun you originally envisioned. There’s often something you can do in your level to get that vision across. Take the time to figure out what’s lacking or what’s preventing you from having that ultimate experience. You are the only one who can make it better.
The Myth of the “Every-Man” Designer
The “Every-Man” designer is the person who thinks that he or she knows what every person wants in a game. Being human and of only one mind and heart, this is a very pretentious assumption. You should have the humility to recognize that your tastes differ from others and that you are not always right. Keep your mind open to feedback and fresh ideas, and consult with people who may have more experience than you. If you do not, your games will miss their intended
Game design is a very hard skill to judge, being intangible, evolving, and not taught in any school. The “Every-Man” designers take advantage of this by putting on airs of great skill to put themselves ino positions of power. Unfortunately, our industry is full of such people and they are often in a position to judge and change your work. I hope that by mentioning this here, early in your career, that you will not become one of them, because it can be a very unpleasant realization for you and your company that you don’t know what every player wants.
Developing Level Design Instincts
Level design instincts are what employers look for when they interview you. To a certain extent, employers assume you have some of these instincts if you have designed any levels at all, for they only come from practice. They are what you take from game to game and project to project, and they’re what make your job so special. It’s these instincts that let you immediately apply design theories and rules on the first pass of designing a level.
You’ll know when you have developed good instincts when you can look at someone else’s level, or an early level of your own, and the mistakes will glare at you. All of the rules in this series of articles came from my own instincts which I developed over years of making games, making plenty of mistakes, and having plenty of realizations. You, as a beginning designer, will make plenty of mistakes. However, hopefully you will learn from these experiences and you will stick with it. Hopefully these level design theories and rules will get you a head start on a satisfying hobby or career in level design.