Design on the Brink.

To coincide with QuakeCon, last week saw a number of Bethesda published games receiving discounts on Steam, part of that promotion was a ‘Free Weekend’ for Brink. Having heard mixed things about Splash Damage’s “mingleplayer” title, I took this opportunity to investigate for myself. I play Team Fortress 2 and occasionally Halo Reach when the mood strikes and I’ve spent time playing and designing levels for Unreal Tournament 3 however prior to that I hadn’t spent much time time playing online multiplayer games since Quake III Arena, though even now I can close my eyes and play a game of Q3DM17 ‘The Longest Yard’ in my head. So I started Brink not as a complete neophyte but also not as a veteran of Splash Damage’s Enemy Territory series; though I did spent several hours with the original multiplayer portion of Return to Castle Wolfenstein.

Once through the menus and character customisation phase my first impressions of Brink were not favourable. The layout of the levels, combined with the UI, often made it difficult to find particular Objectives. Markers on the UI would point straight ahead even when it was frequently necessary to not only turn away from the Objective itself but then head either up or down in order to reach it. This was highly counter-intuitive the architecture of the levels guided my movement in a particular direction and the UI was confirming that I was nearing the Objective yet in all but a few instances I would run into a wall and have to turn almost ninety degrees to locate the way to progress. Once I reached them the Objectives were frequently in locations with few uniquely identifiable landmarks which made orientating myself with relation to the environment and other players a challenge.

I’ve no doubt that repeated play would have helped me learn the layout of each level, but for a player approaching Brink for the first time the damage had already been done. My first reaction was one of getting lost and frustrated.

Even for a new player this is rarely an issue in Team Fortress 2, or Halo Reach, and the big differences are in the use of environmental cues both obvious and subtle to guide the player through the level. Within Team Fortress 2 players are either BLU or RED and areas of the level occupied by one team or the other are coloured appropriately. In addition more blunt and obvious signposting directs players to ‘Capture Point A’ or the BLU ‘Flag Room’.

The two factions in conflict within Brink don’t really give themselves to the use of colour coded environments. The Rebels and the Security have individual colour schemes in the broadest sense but the variation within each can make it difficult to determine exactly which side is which. This is part of a larger problem with the visual design of Brink in that it is difficult to tell at a glance enough information about other players to make useful tactical decisions. Additionally the highly dynamic nature of the objectives within Brink creates further problems when considering colour coding the environment.

The use of direct signposting however would make a lot of sense. The Ark is an environment build for people to live and work in so human readable signs would not be out of place. Such signposting would require careful design to ensure that it remained readable in Brink‘s visually busy environments and also kept to the aesthetic style of the different parts of The Ark. In fact the Reactor level does feature such signposting in the form of coloured lines running along the floor guiding players to different areas of the level. These have two main problems, firstly the nature of the Objectives within the level mean it’s not always clear which of the named locations I wanted to be heading toward at any given moment, and secondly, and possibly more importantly, identifying where those lines are leading required that I find the point on those lines where the location is written on the floor. In order to know which direction to go in I needed to walk through the level looking down at the floor until I could work out which coloured line lead where, suffice to say that didn’t end too well.

Even when implemented superbly these measures are aids and shouldn’t be the only way to guide players through a level. Ideally the very structure, the architecture, or the level itself should do the heavy lifting. How would you go about designing a level for a game like Brink so that the architecture itself helped to guide players, in particular new players, through the space in the most efficient way for the given Objectives?

The difficulty comes in balancing the needs of new players with those of experienced players. The former need to understand where they are going and how to get their while the latter benefit from a more nonlinear layout where their skills and knowledge of the level can be put to use. Brink effectively caters to the second group of players, with levels that allow various opportunities to use knowledge and skill to gain a tactical advantage, but it is frequently unsuccessful at providing the type of guidance required by new players.

With that in mind, if I were designing a Brink level this is the approach I would take.

With Objectives and Spawn points decided upon I would set each of the Primary Objectives in or near an identifiable and named landmark. Such locations will naturally become focal points of action by dint of them being Objective locations, additionally it is just as natural for the corridors and hallways leading to such locations to be designed from an architectural perspective to lead people toward them; offering different perspectives of them from different angles during the approach. Each Primary Objective location would then be linked by these logically constructed spaces, and since each location would be associated with a particular named landmark simple in-context signposting could be used to add an additional layer of guidance. Furthermore players could refer to these locations using their names in order to coordinate their movements with the rest of their team.

New players would be informed of their Objective and its position in reference to being near or at the specific landmark. They could then follow the obvious route to the Objective in order to rapidly enter combat. With approaches designed to offer glimpses of the final location these players would have the opportunity to survey their Objective before engaging the opposing team and if necessary push ahead or wait for support. From the defenders perspective these glimpses of the approaching team would allow them to estimate numbers and make preparations accordingly.

Consider the Boneyard level from Halo Reach, when played in ‘Invasion’, the attacking team approach the first Objective across an open area that provides both cover and a line of sight to the Objective and the area around it, allowing both attackers and defenders time to form plans and take preparatory actions before reaching engagement range. Pushing further through the level attacking players approach subsequent Objectives by moving through the half destroyed hulk of the USS Commonwealth. The ship again provides cover from attack while allowing glimpses of the area immediately around the next Objective. Compare this to the Security Tower level in Brink, where for either team it is all but impossible to even see the first Objective until you are on top of it, and there’s very little time for preparations before entering combat. Obviously Boneyard is a vast map, larger than any in Brink but the ability to catch a glimpse of what is going on before you enter a particular encounter is something common to all levels within Halo Reach, even ones as compact as Zealot. There are instances in Team Fortress 2 where you are not offered a sight of what lies ahead but these encounters almost exclusively occur as the climax of a level, the final Capture Point of the Second Stage of Dustbowl being a prime example. Still the area surrounding the first Control Points of each Stage of that level are visible from the the Spawn location and players are never asked to enter these first Stages unprepared.

With the majority of new or inexperienced players taking these direct routes it would be easy for an experienced team to lay an ambush for them, this is countered by the ability of attacking players to observe the Objective before committing to any engagement. Giving them the opportunity to wait for support before pressing ahead, encouraging teamwork by making it feel like a choice on the part of the players rather than a stricture imposed by the game.

With this basic layout in place I would now start to layer in additional pathways between the Objectives, along with shortcuts through these direct routes. Indirect, and often requiring the use of SMART to navigate, these additional routes could take experienced players over, under, or through the routes used by newer players. This would allow experienced players to access Objective locations from different directions. Moving via access corridors, across roof spaces or through basement rooms these paths by their very nature would not be as obviously signposted and would allow experienced players to exploit their skill and knowledge of the map to gain an advantage. A well coordinated team could them time an attack so that the defending team were hit from multiple directions simultaneously.

Newer player could even be encouraged to seek out these alternate routes by placing Secondary Objectives along them, or simply by electing to follow the more experienced members of their team.

From the perspective of the defending team, newer player would see the obvious routes for attack and know where to expect the likely attack from whereas experiences players would know to cover the additional entry vectors or to actively enter these secondary routes to engage the attacking team directly.

More complex map layouts could make greater use of secondary routes and non-linear approaches to Objectives, but there’s no reason to abandon the obvious paths, having a navigational safety net to fall back on helps to encourage exploration. The existence of such direct paths also adds to the authenticity of the levels themselves. Whether it’s a place of work, industry or accommodation areas of importance need to be accessible.

A direct route through the environment is not the mark of failure for a multiplayer level consider Team Fortress 2, it’s possible to walk from the Spawn point to each Objective in turn almost on autopilot so good are those levels at subtly directing movement with colour coding, signposting, and above all the physical structure of the level itself. Players who are invested in a game will spend hours mastering the intricacies of each level, but if their first experience is getting lost and frustrated it’s unlikely they will stick around long enough to develop that investment.