作者：Chadi EL-ZIBAOUI

我是《刺客信条》（以下简称《AC）的多人游戏地图的首席关卡设计师，我作为关卡设计师的第一张地图是Mont Saint Michel。

当我4年前第一次接触关卡设计测试时，我一点都不了解项目细节，我唯一知道的只是《AC》是Annecy工作室所开发的一款多人游戏。与FPS多人游戏模 式不同的是，除了《AC》本身，我没有可依赖的现有游戏参考。那时候我玩了几个小时的《AC1》，如果我没记错的话我完成了Damascus。因此我找来 了《AC2》，认真投入游戏并最终几乎完成了游戏。我不只是在游戏，我还特别关注于“幕后”元素并尝试着搞清楚其资产制作，攀升设计，关卡设计以及等等参 数的方法。当我得出自己的结论时，我开始进行测试，这一切都帮助我在2010年5月能够加入该团队中——即在游戏出现在E3大会之前

1.地图设计原则和过程

让我先介绍我们工作原则中的一些理念，这将在我谈论Saba Island时帮助你更好地理解。

专门的地图团队组织

地图的团队组织是我必须熟悉的第一件事之一。根据授权和项目的需要，团队必须在一开始就确定好，在我们的情况中也就是，设计地图将整合游戏玩法需求和强大的历史背景与城市规划和建筑基础。

举个例子来说吧，我们有一些地图“突击队”是由2个关卡设计师（负责地图的游戏玩法元素）以及4至5名美术人员（负责地图的图像内容）所组成，他们在地图理念阶段以及完整且完善的地图的最后发行中都保持着密切的合作。他们不仅会交换许多想法，利用队友的专业技能强化自己，同时他们还具有不同的知识水平以及关于各自工作的实践经验。

例如关卡美术师非常熟悉关卡设计成份和限制元素，相反地，关卡设计师则非常熟悉3D软件包。

地图起源

每次当我开始创造一款《AC》多人游戏时，我们在地图团队中的主要目标便是提供给玩家最棒的多样性以及最高的质量。显然，当谈到多样性时我所指的是关卡设计和视觉识别。

视觉多样性将源自环境和时期本身，而关卡设计的多样性将受到对于之前《AC》项目的数据分析的影响（显然我们是着眼于最常游戏的地图并尝试着明确它们最受欢迎的特定成份），并确保它能够应对你在《AC》多人游戏地图中能够体验到的两种主要游戏风格，特技vs潜行。

就像在导航局限于地面的地图（如《ACB》中的Castel Gandolfo，《ACR》中的Souk，《AC3》中的Brewery以及《ACBF》中的Pirate Headquarter）会被贴上“潜行”的标签，游戏将鼓励玩家待在较低的地方，混在人群中去欺骗目标，而运行导航出现在不同屏幕的地图（不管是地面还是屋顶）则被贴上“特技”的标签，它将提供更多高空导航的机会（如《ACB》中的Mont Saint Michel，《ACR》的Galata，《AC3》的Animus Core以及《ACBF》的Saba Island）。

牢记着这些限制，团队便可以开始收集地图概念方案，我们将对此进行研究并选择最适合游戏的一个。一旦确定了这些概念，团队便将进一步获得细节内容，创造每个地图区域的视觉识别，提供包含2D Ariel视图和3D原型的第一个关卡设计，如此我们便可以体验并评估粗略的卷轴与导航。

两个强大的支柱：导航和方向

显然，导航和方向是我们在地图的整个开发过程中都将关注的两大内容。

关于导航，地图必须通过提供给玩家导航友好型环境去支持我们的核心游戏玩法。在多人游戏模式中，我们希望玩家能够专注于他们的驻澳目标（也就是其他玩家）而不是与环境较劲。因此地图不应该是一个难度成份，而应该是能让玩家利用并从中获益的工具。对于玩家来说，因为地图的设计缺陷而被杀死或错过目标是很受挫的（例如在攀登过程中受困）。因为我们投入了许多努力去完善并修改地图导航元素的漏洞。

第二大关注焦点是方向。地标非常重要，他们能够帮助玩家确定方位并让他们的队友知道自己在哪里，我们总是特别关注于这点。存在一些不同类型的地标：

独特的建筑/遗迹（罗马万神庙，Mont Saint Michel的修道院以及Northwest Passage上失事的船舶）

典型区域（市场，花园，池潭）

自然元素，如朝阳，从而产生阴影

这非常接近于我们在现实生活中所做的，你经常依赖于能够让自己轻松地为别人指点方向或在位置的地方找到前进之路的地方。

历史是精神食量

《刺客信条》是具有积极性与丰富价值的授权游戏，每次开始一个新的《刺客信条》项目时，我们都会深深沉浸于各种历史元素。作为关卡设计师或关卡美术师，我们需要完全沉浸到历史，城市规划，建筑，生活方式等等有关游戏所发生的时代和区域的内容中。在地图团队中，我们特别关注于通过研究阶段而提供给玩家现实且可信的环境。历史是一大品牌支柱。除了提供历史准确性，这也将促成关卡设计理念的诞生。

有时候玩家将注意到我们花了许多时间去优化的细节内容，而有时候他们不会直接感觉到它，但不管怎样这都将提供给他们顺畅的游戏体验。

2.SABA ISLAND的例子

现在我将使用《ACBF》中的Saba Island作为例子去解释地图的生命周期。因为有太多话要说了，所以我决定简化内容从而让你能够大概了解我们是如何面向《ACMP》设计地图。

早期的概念

Saba Island的历史是始于为《ACIII》进行DLC地图研究期间，但那时我们认为这并不适合美国革命背景。启发这一地图的区域是科罗拉多的Mesa Verde。这是塑造我们关于地图最初的视觉和游戏玩法想法的参考对象。

如果你有机会尝试我们的多人游戏模式，我敢保证你将能够清楚为什么这个地方是Saba Island之母。实际上，这一地图是地图内容总监的幻想，他渐渐地喜欢上这个地方，当我们开始着眼于它时，我们也会因为其潜能而受到诱惑。

我们开始编写地图的早期概念，其主要理念为：

在有房子的岩石下建造村庄。

高地起伏的地域。

地图中间一个荒凉的区域。

此外，我们也很早就提出了在地图中间设置一个光井的想法。

村庄中的特技游戏玩法带有许多地面到高出以及高度下到地面的机会（这里的参考是Mont Saint Michel，这仍然是带有最强大的下坡的地图）。

地图中间更平坦的区域将鼓励潜行/基于地面的游戏玩法。

我们收集了一些视觉参考，历史和地理信息。这个理念暂时被搁置了，不过并未耽搁太久。

当我们开始创造《ACIV》的Black Flag时，我们真的对海盗主题和Mesa Verde的理念感到兴奋。

显然我们将创造一个秘密的海盗巢穴。

不同的地图，不同的现实

当我们开始创造一张地图时，我们是依赖于4个不同的现实并进行适应，从而让它匹配我们的游戏玩法：

现实本身。当我们发现一个与游戏环境相一致的地方并且它已经提供了一个有趣的布局，我们便会开始收集更多信息和蓝图，然后为了支持游戏玩法而做出适当调整（就像《AC3》中的Saint Pierre或者《ACB》中的San Marco）。

《刺客信条》的个人现实。当我们想要呈现一个个人战斗已经呈现的地方时，我们便需要从它们已经接收的现实开始。这并不是一次简单的修建，即我们可以整合自己的游戏模式，我们需要进行一次重要的关卡调整，一次去适应我们的游戏玩法—-不同于他们的游戏玩法（如《ACB的Roma，《ACR》的Galata，以及《AC3》的Boston Harbour）。我们经常提供类别的一两张地图从而与游戏的单独部分保持有效的联系。

真实的地方，虚假的地方。在这种情况下，我们拥有一个强大的主题，但是我们却并未找到一个能够有效呈现它的现有地方。让我们以New York Brewery为例。当我们将其呈现在游戏中时这个地方并不存在，但我们通过收集了有关啤酒厂的信息从让它看起来非常真实。

幻境。这是我们知道并不存在于现实中，但是存在于流行文化或集体想象中的地方。就像《AC3》中的Animus Core（与Animus环境和品牌的DNA联系在一起），以及《ACBF》的Saba Island。

Saba Island属于最后这一类，事实上我们不能带着实际的城市规划去创建我们布局的基础。虽然基于实际的城市规划具有一些优势。

首先，城市规划是过去几个世纪里专家用于处理街道，场所，房屋等等分配的众多参数的结果。这意味着事情的存在具有一定的理由，因为我们不想重新创造城市规划规则，所以我们需要理解原因。

其次，基于这些规则，每个人都拥有他们自己关于城市及其代码的看法，这是你在每天生活中所体验到的。举个简单的例子来说吧，因为宽度的不同，你可以创造出街道或大路间的差别，并且你可以猜测在富裕或贫穷的邻居间你是否能够创造出一条路（只因为你周边的材料和公共设施）。城市中有很多东西能够帮助你更好地了解环境，这也是我们在设计不是受到现有场所直接启发的地图时所关注的焦点；我们依赖于这些代码去行程布局。

从概念到原型

这将我引向了下一步，即基于源自概念的视觉和游戏玩法目的，我们将创造第一张详细的地图。我们的工作主要分成两部分，但最后它们也会紧密地联系在一起：一方面定义定图上各个领域的视觉识别，另一方面创造第一个2D布局原型。

然后团队将提供一套有关所有信息的文件并开始创造地图。

每个区域的识别将提供有关不同氛围的信息，同时还会提供有关哟徐i玩法目的的信息。关于Saba Island，团队识别了4个区域：

早前版本的区域识别

早前版本的游戏玩法识别

监管区域被当成是后勤和供应区域，它们能够储存其财报；荒凉的区域是地图上唯一没有任何建筑的区域；海港将具有任何与船只以及修理车间相关的内容；生活区域是海盗们吃饭喝水的地方。

设计文件中详细描述了每个区域及其特性的相关内容。

此外，我们还收集了更多有关全局拓补的信息进一步明确地图的外观和可能带给人们的感受。

因此，关卡设计师绘制了第一张2D地图并在此识别了主要街道的场所，不同的地面关卡，拥挤的网络，碎琅区以及隐藏区域。一旦确认了所有的这些信息，我们便开始在编辑器中使用占位符资产创造第一个3D原型。这帮助我们快速对地图产生第一引擎然后能够开始定义下一个步骤。

你可以注意到上面的俯视图与你能在游戏中体验的地图是不同的，但你已经找到了主要的成份。这是质问并迭代每一张地图所创造的结果。这是我们关于Saba Island以及每张地图所提出的一些主要问题，我们团队通过创造原型去测试这是否是一个值得遵循的方向。

首先，我们想要衡量在可游戏区域中设置船只的利弊，然后基于不同的船只大小和方向创造不同的原型。其优势在于能够提供给整张地图一个好的意境；秘密的海盗巢穴是不能没有船只的，对吧？同样地，作为一个地标，这能够帮助玩家更好地识别方向。具有挑战的一部分是关于形状妨碍了海港两部分间的简单且快速的秘密导航。关于每个这类型的决策都将有不同的参与者，因为我们需要考虑美术，关卡设计，技术限制等，然后选择一个最合理的解决方法。我们决定整合船只并关注于它是如何通过楼梯与其余的地图部分联系在一起，并将其与船只北部的丛林连接在一起。

关于丛林，挑战是来自关于像树木和石头等形状的使用。我们在创造Northwest Passage便拥有类似的体验，我们知道这是具有挑战性的，所以我们需要付出许多努力去提供一个清楚且经过优化的导航。如此，这将创造一个带有更多潜行游戏玩法的独特场所。然而，对于那些喜欢特技导向型游戏玩法的玩家来说，这并不是什么优势，因此我们在这区域上方整合了一些自由奔跑的内容。关于船只，这也是一个可辨认的区域，总是作为我们选择背后强大的动机。相信我，我们在这区域中完成了许多形状的原型创造与清除，但最终，团队想出了可适应的解决方法，并且每个人都对最终结果感到满足。

团队必须专注的另一个主题是城市/建成区中的建筑设计。我们想要评估在岩石中设立房子并以洞穴上沿作为自然的屋顶的好处，就像Mesa Verde。再一次地，我们开始创建不同理念的原型，如使用较低建筑的屋顶作为NPC的地面，深入岩石挖掘狭窄的街道，并在地面与洞穴的上沿间建造房子。这一点是非常微妙的。如果我们是基于这种方式建造房屋，那么当你开始攀登其前方区域时，你便不知道自己不能触及屋顶，因为它根本就没有屋顶，所以我们不能让玩家发现这样的“惊喜”。于是我们便在完成房子的建造之前创造了一个中间屋顶，然后调整攀登路线让玩家能够最终到达屋顶。基于此，我们将这一特别的建筑作为Chief的房子，即成为该区域的地标。

现在团队将继续创造并专注于我们所创造的这些强大的方向，而原型创造，游戏测试和分析将开始一个新的循环，最终我们将再次明确下一个步骤。

关于我们在地图上所作出的修改的质量的一大重要组成部分便是玩家本身。很早之前，当我们拥有第一个可游戏的地图版本，即带有大多数成份和游戏模式时，我们便开始向外部游戏测试者呈现地图，让他们玩几个小时，除了我们的服务器在其游戏过程中所记录的数据（游戏邦注：玩家位置，致命的场所，隐藏区域等等）外我们还会收集他们的看法。而当我们能够在其源头获得信息时，这便是非常幸运且宝贵的内容。

显然，我们需要在某个时候停止这一过程，因为如果不这么做，这便会是一个永无止尽的工作。然而，我们不能一下子停下来，如果在适当的时候我们继续向前，那便不能做出多大的改变。当我们满足于地图游戏玩法的感觉时，我们便能够专注于游戏模式平衡与调整，优化导航和资产等等。

（本文为游戏邦/gamerboom.com编译，拒绝任何不保留版权的转载，如需转载请联系：游戏邦）

Designing Multiplayer maps for Assassin’s Creed : Saba Island Case

by Chadi EL-ZIBAOUI

Hi, my name is Chadi EL-ZIBAOUI I’m the Lead Level Designer for AC multiplayer maps, I’ve been working on ACMP since ACB, my first map as a Level Designer was Mont Saint Michel.

When I first took the LD test 4 years ago I didn’t know anything about the project’s details, the only information I had was that a multiplayer for AC franchise was under development in Annecy studio. Contrary to an FPS multiplayer mode, I had no existing game references on which I could rely on, except AC solo itself. At that time, I had played AC1 a few hours, and if I remember correctly I’ve completed Damascus, and that was it. I thus grabbed a copy of AC2, and I played intensely, completing almost 100% of it. Not only did I play it but I paid particular attention to the “behind the scenes” aspects and tried to figure out what were the recipes behind the assets production, climbing design, LD, and metrics amongst others. Once I’d drawn my conclusions, I jumped into the test, and gave it my best, which allowed me to join the team in May 2010 during a very dynamic period of the production – just before the game was revealed to the press at E3.

I.MAP DESIGN PHILOSOPHY AND PROCESS

This said, let me introduce a few concepts that are part of our general work philosophy and that might help have a better understanding when I’ll be talking about Saba Island case.

A dedicated map team organization

The map’s team organization was one of the first thing I had to get familiar with. The team has been shaped from the beginning to be tailored with the franchise and project needs, which in our case means designing maps that fairly merges gameplay requirements and strong historical background with urban planning and architectural foundations.

For example, we have several map ‘strike teams’ made up of 2 level designers (responsible for the gameplay aspect of the map) and 4 to 5 level artists (responsible for the graphical content of the map) working closely together from the map conception phase  to the final release of the fully polished map. Not only do they exchange a lot of ideas and nourish themselves with their teammate’s expertise, but they have different level of knowledge and practice of their counterpart’s jobs.

For instance, Level Artists are familiar with Level Design ingredients and constraints, and inversely LDs are familiar with 3D software packages.

Map Genesis

Every time we start an AC multiplayer game, our main objective in the map team is to provide our players with the greatest variety we can deliver on schedule and with the higest quality. Evidently, when talking about variety I mean level design and visual identity.

The visual variety will stem from the context and period itself, whereas the level design variety will be influenced by analyzing data tracked on previous AC projects (obviously we look at the most played map and try to identify their most popular specific ingredients), and making sure it addresses the two main gameplay styles you can experience in ACMP maps, Acrobatic play versus Stealth play.

As an example, a map in which navigation is constrained to the ground-level for the majorioty (Castel Gandolfo in ACB, Souk in ACR, Brewery in AC3, Pirate Headquarter in ACBF…) is labelled “stealth” and players are encouraged to stay in low-profile and play with the crowd to outwit their target, whereas a map that allows navigation on different levels (whether it be grounds or roofs) is labelled “acrobatic” as it will offer more aerial navigation opportunities (Mont Saint Michel in ACB, Galata in ACR, Animus Core in AC3, Saba Island in ACBF).

With these constraints in mind, the team can start gathering map conception proposals which we will then study and select the ones we think best suited with the game. Once these concepts are validated, the team can start getting further into the details, developing the visual identity of each area of the map, providing a first level design which will generally consists of a 2D Ariel view and a 3D prototype so we can experience and evaluate rough volumes and navigation.

Two strong pillars, Navigation and Orientation

Obviously, navigation and orientation are two things we will focus on during the whole development of the map.

Regarding navigation, maps have to support our core gameplay by providing a navigation-friendly environment to the players. In the multiplayer mode, we want players to be able to focus on their main objective (that is, the other players) rather than struggling with their environment. Thus maps should not be a difficulty ingredient but rather a tool, which players should try to make the most out of, and use at their benefit. It would be very frustrating for players to get killed or to miss their target because of design flaw in the  maps (i.e. getting stuck while climbing), Therefore we put a lot of effort on polishing and debugging the navigation elements of the map.

The second big focus is orientation. Landmarks are very important to allow players to get their bearings or to let their teammates know where they are,  we always pay special  attention to this. There are different types of landmarks:

Unique buildings/monuments (the Pantheon in Roma, the Abbey in Mont Saint Michel, the wrecked ships in Northwest Passage…)

Typical areas (a market, a garden, a pond…)

Natural elements, such as solar orientation, and resulting shading

This is very close to what you would do in real life, you usually rely on places which are easy to identify to give directions to someone or to find your way in an unknown place.

History as food for thought

AC is a very motivating and enriching franchise for it forces us to be immersed into historical aspects every time we start a new AC project. As Level Designer or Level Artist, we need to fully immerse ourselves into history, urban planning, architecture, lifestyle and almost anything related to the period and locations in which the game is taking place. In the maps team, we pay particular attention to the research phase (which can sometimes involve taking courses with lecturers from Museum’s) to provide players with realistic and believable environment, History being one of the brand’s pillar. Besides providing historical accuracy, this will also give birth to level design ideas.

Sometimes players will notice the details we spent time polishing, and sometimes they won’t directly perceive it, but this consistent whole will provide them with a smooth game experience.

II.SABA ISLAND CASE

I will now explain the life cycle of a map using Saba Island from ACBF as an example. There is a lot to say, thus I tried to simplify the content so you can have a general idea about how maps are designed for ACMP rather than going into too many details.

Early Concept

Saba Island’s history started during DLC maps research for ACIII, but at that time we agreed that it wouldn’t fit easily with the American Revolution context. The location which inspired this map was Mesa Verde in Colorado. These are the kind of references that shaped our first visual and gameplay intentions for the map.

If you had the chance to try our multiplayer mode I’m sure you can point out why this place is the mother of Saba Island. Actually this map was the Map Content Director’s fantasy, he literally fell in love with that place and when we all started looking at it we were seduced by its potential too.

We started writing a kind of pre-concept for the map in which the main ideas are resumed as:

A village built under the rocks with houses dug into it.

A terrain with differences in height.

An organic and wild area in the center of the map

Additionally, the idea of a light well in the center of the map was put on the table very early.

Acrobatic gameplay in the village area, with a lot of ground-to-roof and roof-to-ground opportunities, (the reference here was Mont Saint Michel which is still the map with the greatest downslope).

A more flat area in the middle of the map to encourage a more stealth/ground focused gameplay.

We gathered some visual references, historical and geographical information, and that was it. The idea was put aside, but not for long.

When we started ACIV Black Flag we got really excited by the Pirate theme and Mesa Verde’s idea naturally came back to surface.

It became obvious to us that we would build a secret pirate lair (which was actually the name of the map during its production).

Different maps, different realities

That said, when we start a map we rely on 4 different kinds of realities to adapt it more or less so it fits with our gameplay:

Reality itself. When we find an appealing location which is consistent with the game context and that already provides the roots of an interesting layout, we will start gathering as much information and blueprints as we can, we’ll then make the adjustment needed to support the gameplay (as an example, we can mention Saint Pierre in AC3, or San Marco in ACB)

AC Solo Reality. When we want to represent a location the solo campaign already showcases, we need to start from the reality they already adapted. It’s not a simple crop in which we can integrate our gamemode, there’s a quiet an important level of adaptation work that is conducted to fit with our gameplay which is different from theirs (Roma in ACB, Galata in ACR, Boston Harbour in AC3). We usually provide one or two maps of that category to have a better connection with the solo part of the game.

Real places, fake locations. In this case, we have a strong theme we find very interesting but we didn’t find an existing place friendly enough to represent it faithfully. Let’s take New York Brewery example. This place didn’t exist as we represented it in the game but we gathered enough documentation about breweries at that time so it looks credible and inspired from reality.

Fantasy. These are locations which we know don’t exist but they exist in pop culture or collective imagination. This was the case of Animus Core in AC3 (which was linked to the Animus fantasy and closely linked to the brand’s DNA), and Saba Island in ACBF which is the map we are talking about.

Saba Island belongs to this last category, as a matter of fact we couldn’t start with real urban plans to build the foundations of our layout. Starting with a real urban planning does indeed have some advantages.

Firstly, urban planning is the result of a lot of parameters that experts dealt with to organize streets, places, housing, etc. over centuries. This means that things are where they are for a reason and we need to understand why because we don’t want to reinvent urban planning rules.

Secondly, everyone has developed their own perception of the city and its codes because off all these rules, this is something you experience in everyday life. As a brief example, you can make the difference between a street or an avenue because of its width and you can guess if you’re making your way in a rich or poor neighborhood just because of the materials and the street furniture around you. There are a lot of things in a city that will help you understand your environment and this is what we focus on when we design a map which is not directly inspired by an existing location; we rely on these codes to shape our layout.

From Concept to Prototypes

This leads me to the next step, based on the visual and gameplay intentions from the concept we will make a first detailed map design of it. The work is split in two parts but in the end they are closely linked to each other: defining visual identity of the various areas of the map on one hand, and prototyping a first 2D layout on the other hand.

The team will then provide a set of documents with all the information needed to start building the map.

The identities of each area will give information about the different moods and also give information about gameplay intention. 4 areas were identified by the team for Saba Island:

Area Identities early version

Gameplay Intentions early version

The intendance area was supposed to be the logistics and supplies area where they could store their treasures also; the wild area was the only part of the map where there wouldn’t be any construction; the harbor would host everything related to boats with repairing workshop; the life area would be the place where the pirates can have a drink and some food.

Each one of these areas and its characteristics are detailed and illustrated in the Design Document.

Alongside this, more information about the global topology is gathered to help us figure out further, what will the map look and feel like.

Thus, the Level Designers draw first 2D plans where they identified the main streets and places, the different ground levels, the crowd network, the chase breakers, and the hiding places. Once all this information has been validated, we start building the first 3D prototype in our editor with placeholder assets (simple shapes or assets from previous games). This is what helps us to quickly have first impressions about the map and define the next steps.

2D top view and resulting 3D prototype

You can notice that the above top view is quite different from the map you can play in the game but you already find the main ingredients. This is the result of a lot of questioning and iteration every map goes through. These are some of the main questions we raised about Saba Island and for each one, the team built prototypes to test if it was good directions to follow.

First, we wanted to weigh up the pros and cons about having a boat inside the playable area, different prototypes with different size of boats and orientation were made. The good point was that it provided a great mood to the whole map; a secret pirate lair wouldn’t be without a boat, right? Also, as a landmark it was very useful for player’s to be able orientate them. The challenging part was concerning the shape impeding easy and quick stealth navigation between the two parts of the harbor. As for every decision of that kind, there are different people involved because we need to take into consideration artistic, level design and technical constraints to choose the right solution. We decided to integrate the boat and pay extra attention to how it’s linked to the rest of the map through stairs deserving the upper deck, and free run to connect it with the Jungle part to the north of the boat.

Regarding the Jungle, the challenge was coming from the use of organic shapes such as trees and rocks. We had a similar experience with Northwest Passage, we knew it was challenging and that it would need a lot of effort to provide a clean and polished navigation. Beneficially, it would create a unique playground with a more stealth oriented gameplay. However, it should not disadvantage players who like to play it acrobatically, thus, we integrated a few free runs above the area. As for the boat, it would also be a very recognizable area which was always a strong motivation behind our choices. Believe me, a lot of shapes prototyping and cleanup has been done on that area, but eventually, the team came up with adapted solution and everyone was satisfied by the final result.

Another topic the team had to focus on was the building’s design in the city/built-up area. We wanted to assess the benefits of having houses dug in the rocks and linked to the natural ceiling of the cave, as in Mesa Verde’s references. Again we started prototyping different ideas, such as using the roofs of lower building as a ground for NPCs, narrow alleys dug into the rock and houses built from ground to the ceiling of the cave. This particular point can be delicate. If a house is built that way, when you start climbing its frontage, you don’t know that you won’t be able to reach its roof since it doesn’t have one, so we can’t let the player land on such a surprise. What we did is create an intermediary roof before the end of the building and we tweaked the climbing so you always reach it wherever you start climbing this building. On top of that we identified this particular building as the Chief’s house that would be the landmark for that area.

The team could now go further and focus on these strong directions we built with them, the work of prototyping, play testing and analysis will start a new loop, at the end of it we will define the next steps again.

A key ingredient to the quality of the modifications we do in the maps is the player himself. Very early, when we have a first playable version of the map with most of the ingredients and game modes integrated, we start showing the map to external play testers, we ask them to play several hours and gather their impressions in addition to data recorded by our servers during their games (player position, kill locations, hiding places usage…). We make maps for players and when we can get the information at its source, it’s a godsend and very precious to us.

Evidently, we need to stop that process at one point, because otherwise, it’s a never-ending-story and you can always improve a map. However, it can’t be stopped in one go, so the more we move forward in time, the smaller the modifications get. When we are satisfied with the map gameplay feel, we can focus on game modes balancing and tweaking, crowd adjustments, polishing navigation and assets, I will detail that part in another post.(source:gamasutra)