Article - Detail vs. Design

[MC]

This article was written by David Johnston, a mapper on the Counter-Strike team who created the maps "Dust", "Dust 2" and "Cobble".

Part 1:

It seems to be a common misconception that adding as much detail to game 'maps' today improves the ambience the player can experience. These are small details like individual handles on doors and drawers in game environments where the player isn't likely to pay much attention to them, let alone care if such detail was present or not. Yet designers continue to flood their maps with these 'improvements'.

The problem is that even maps with such minor additions rarely get the ambience boost they need. Will a player really think an environment is much more fun to play in if they can outline a handle on a drawer some 20 feet away? Can someone be so boring as to get excited about such a small detail?

It seems 'fun' is no longer such a factor in games today. Appearance and polycount is what matters more. With the amount of screenshots and images released from games today as publicity, more attention is turned to what the engine can display and how fast it can do this than how this technology can be used to create a game experience for the player. We see it everywhere - PC and console leisure magazines are infected with a compulsive will to judge games on images alone - even when the images show nothing of the actual gameplay such technology can be used for.

Let's take a step back. This happens with current and ageing games too. More focus is being put on appearance and eye candy than gameplay and flow - and if you want to be a designer, you need to understand how to use both. Graphic designers not only know how to use their tools, but how to use them in the way they were intended. Artists know that they can dip a paintbrush into a paint-laden bucket and draw something across a canvas - the good artists realise this can be used to create 'art'.

I take an example - the 'Avanti' map from Valve Software's TFC. When I first walked around that map in the game, I was in awe. It was amazing. I felt like I was not only walking around in a possible Team Fortress II, but also a real version of this place. I thought it must be the details - but on closer inspection details weren't overpopulating this map. It was its design. Simple and basic to an extent, but used in such a way that it 'felt' incredible. In addition, it played well too. Not that we didn't expect that from a company of such calibre.

Looking at some other TFC maps... can anyone honestly say Rock2, 2Forts or the assortment of other maps would be much better if details were scattered across them? Would they be better? Probably not, no. (I can see Mr Sawyer giggling to himself a little here... Casbah had a few small details on it - handles and things. I'm a hypocrite!)

There are too many level 'designers' out there who believe that just because they can use the tools, set up a system of entities (unless they're required for the gameplay, like TFC entities) or functions or two, perhaps create some interesting shapes, they know what makes a good map. To be frank, they don't. Granted, many do, and this is obvious from their released work. However, screenshots show little about gameplay when it comes to map design.

I always thought such attention could only benefit a map - but it doesn't. My current projects have taught me that. Often, an ignorance helps make a map feel right. Creating an environment where everything is perfectly aligned and cut straight feels 'wrong'. Another example - look at maps in Half-Life, specifically those underground, or with strong 'mechanical' presences (such as On A Rail). Designers who know Half-Life and are obsessed with getting perfect edges in their levels will cry when they see some of these maps from the game that literally made our heads explode in delight (well, some of us). There are unaligned textures all over the place - 'glass' textures used where glass would never be used in real-life, and ignorance of some 'rules' of texture placement. Yet, does this matter? If the same person was to have made the map instead of the original author, would it be better with his ideals and perfections?

Probably not, no. Ignorance when it comes to design is what created the ambience Half-Life gave. Maybe some of it was time constraints or forgetfulness, maybe ignorance, but it worked. The time they could have spent 'improving' these things has gone into thinking about the gameplay beforehand. And even if they did have time to 'improve' the map by aligning everything, that would destory some of the atmosphere it gives.

So when someone (an amateur, perhaps someone with little knowledge about what makes a game fun) sees two maps, alone, in the game environment, with no true gameplay progressing - an environment perfect to explore the map - what usually sways them? Small details. Lightbulbs created with 50 polygons compared to the measly norm of 10. Carpet which is perfectly aligned to corners of rooms. Trims separating every area from the next. 'Neat' placement of architecture in line with curves overhead and bevels below.

That would be what a bad judge would do. A good judge would take a different approach. Sure, they'd notice details and small, minor enhancements, but they'd gladly put some of it aside to focus on gameplay too - how would this area benefit the player alone, and how about his team? How could it be abused by an opposing team? What would the gameplay be like without this area? Could this area be changed to focus attention elsewhere?

Those sorts of questions don't usually pass some 'mappers' minds till they get past testing. They haven't considered the gameplay implications of their map beforehand, and as such have much work to do after testing.

The problem is an inbalance of time use. Some 'designers' put far too much time into the appearance of their maps - making sure textures and surfaces are perfectly aligned, architecture looks impressive, and small 'enchancements' like that. However, with this time being used for such purposes, no thought is developed for the gameplay side of the map.

It's a shame - gameplay works on common sense. Everyone has that. Some have slightly twisted versions of common sense, but that makes maps all the mpore individual when the gameplay is different to what our common sense tells us to expect next. What's disappointing is how designers seem reluctant to fall back on basic human thought to prevent looking outmoded by technology enhancements.

Part 2:

The problem with detail is deciding when to add it, how to add it, and what to add (of course, detail doesn't need to be an additive process - it could be modification or even subtraction). The details mentioned previously are quite inane - aligned textures on surfaces where textures needn't be aligned, and trims absolutely everywhere. I'm not saying a designer should be ignorant of these factors - indeed without them, some maps would lack the ambience needed. Yet sometimes, such attention to detail acts against the map.

Almost religiously, we can again look at Half-Life. Thinking back to the "On A Rail" chapter, one can most probably remember where some switches, crates and stairs were at some point during the whole experience. The chances are you ignored trims, or at least consciously ignored them. You wouldn't stop in a run from a pack of grunts just to gaze at the wonderfully crafted trim separating one area of flooring from the next. You don't notice these things, sometimes even at a sub-conscious level. Yet, aside from trims, there were many other details in this chapter.

It seems important now to define what I call 'detail' in a map. To me, detail constitutes the addition of geometry that doesn't add to gameplay directly, at least not compared to what it adds to the ambience.

Back to "On A Rail", there were/are lots of 'details' carefully scattered across the chapter. No area is boring or cumbersome, but to the unsuspecting designer it's never obvious why. I refuse to describe what it is here. It isn't attention to detail - lining up small textures on minature parts of the map, in fact sometimes its the opposite. The ignorance of complete visual perfection. It is the tapping into the players subconscious through design, and more. That's for designers to discover. In the process of discovering what makes a map you learn much more than I could possibly remember and/or describe.

I know many budding designers out there will hate me for not describing it here, but discovery alone is not only educational, but improves self-confidence. I've been there.

What I'm desperately trying not to suggest is that complete mis-alignment of textures is the way forward. That is underkill. I've almost said aligning of textures to edges isn't important. That's a lie. It is. What is important however is knowing when to do it, when it is worth the time, and when it isn't.

Generally, the problem with some 'designers' today is their obsession with tiny things that in reality, don't matter. Maybe its jealousy. One shall look at another map, be awed at how well designed it is, and in order to attack it, criticise how textures aren't aligned - assuming the original designer doesn't know how to align textures properly. In order for a map to be perfect among mappers (notice the distinction between mappers and designers), everything has to be done to make the look perfect. Edges have to be sharp and concise. You'll never find paving slabs ending halfway into a wall. Pipes use designated pipe textures, nothing else. Cables and supports use textures you'd never see used on anything else. Everything has a single use.

That's not the right approach. That is just a shallow way of sucking away confidence from people who could be excellent game developers. I have seen people with so much hope and promise turn their back on what they enjoy simply due to unneeded criticise from others (those who are considered 'friends', but actually act in competition - I am guilty of this crime in the past, and I admit it).

Keeping an open mind is what designers need. Not only is the understanding of game flow and artistic design important, but also how to use it in unexpected ways. Realising how one thing, no matter how unconnected it may seem, could work extremely well for something else. Being prepared to be different, and willing to practise this is needed.

As an overall look at the detail issue, let me give a basic way to determine when 'detail' is needed and when it isn't. It's actually common sense. It works on the basis that players will actively notice some things about an environment, and will ignore others. Players notice if walls aren't aligned to the floor or ceiling properly. They notice if paving is at a weird or unconvincing angle to adjacent walls. They'll notice if a door has edges in its middle. They don't notice if a glass texture has been used to seperate two floor types, and even if they did, their involvement in the game - due to its gameplay - should determine if they are convinced by it or not. An unconvincing game leads to amplified criticism of its environments.

I know, after my rant here, some people will argue I am complaining about things I am guilty of, in essence, being hypocritical. Yes, I have been 'guilty' of some of the acts I have mentioned, including over-attention to silly detail that doesn't matter. I realise it doesn't matter. I know I could have done these things better. I have said before in interviews I am my own best critic. I criticise everything I do. If other designers did the same, the quality of maps released today would increase substantially. Some people just need to take a step back, look at what they are doing, and realise that everything has imperfections. Then they need to put their overexaggerated ego aside and realise that these things need addressing. On the other side, those with potential need to understand how it should be used, and how it will be criticised.

Source: http://hosted.planethalflife.gamespy.com/davej/detail_versus_design.htm