Task 24: Personal review of the second year: Where do you want to go, and how do you get there?

This year has been a learning experience in many ways. I’ve expanded my artistic skills and been introduced to several new types of software, but I’ve also learnt about how important it is to manage my own creative process. I’ve had a fairly hands-on lesson about time management, and what happens when it goes wrong, and I’ve come out of it with a better understanding of how to approach tasks in future.

A lot of the time, it was down to my inexperience with a certain program, or underestimating how long something would take. Obviously, on the other side of that, I can now reflect on what happened and use the experience to plan better for project I do over the next year, and even further on.

Last year, over the summer, I’d intended to learn how to use UDK. I put it off while working on other project, and as a result the building project during the first term was the first time I ever used the software. While I did manage to learn what I needed, I still felt very unsure of what I was doing, and that had a big impact on my ability to work efficiently. I had exactly the same problem after Christmas – I’d had no prior experience with Z-brush, and therefore I wasn’t able to anticipate what would be involved when incorporating it into my workflow.

A lot of the time, I was simply overwhelmed with the amount of information there was to sift through. Mike Pickton’s video tutorials were often very helpful, but I still felt quite lost at times. There are plenty of tutorials out there, but most of them go into much more detail than I needed for what I was doing, and it was often difficult to find the small piece of information that I was looking for. I also tend to have trouble working up the confidence to ask for help when I really need it.

Therefore, if I want to get the most out of the projects we’re set next year, then I need to be better prepared. I’m intending to spend the summer better getting to grips with both Z-brush and UDK. I want to work through some of the more extensive online tutorials that I didn’t have time to go through when I had a deadline.

I also want to get more confident at using Photoshop for digital painting. I’ve found myself using pen and paper much more this year, simply because I find it so much easier, and I’d like to be able to produce digital sketches and thumbnails with the same level of confidence.

One thing that’s surprised me this year is how much I enjoy character design. I wasn’t much interested in it previously; I was much more interested in environments and buildings. However, despite all the problems and frustration I had with the Mortal Engines character I very much enjoyed the project, and I’d like to follow through and improve on the skills I picked up.

I have a tendency to be a perfectionist and to obsess over detail, and from what I’ve experienced, that works much better for characters than for environments.  I also enjoy drawing people much more than environments. At this point, I’m strongly leaning towards doing a character-based project for next year’s FMP.

Next year, I want to make more of an effort to keep on top of my work, not just to reduce stress but so I can make better use of the formative feedback sessions. This does mean I need to be a lot stricter with myself about deadlines and how I use my time, and also possibly be less overambitious. I am terrible at estimating how much time a given task will take, and as a rule of thumb I should probably always allow twice as much time as I think a task will take. That way, if I do finish a task to schedule then it just means I have time to improve on it further. When I do come across problems, I need to address them sooner, instead of just hoping they will solve themselves and letting the work build up.

I feel that I’ve gotten somewhat better over the last part of this year, especially with the group project where I managed to keep on top of the work quite well. I’m much more aware of the need to manage my time, and more recently have had more success sticking to a schedule I set myself. I’ll try and continue this over the summer, to stay in practice and just to make sure I spend the time efficiently.

­Ultimately, universities are for self- development, not just learning set skills, but finding out how to improve on what skills you have, and to build on what you already know and enjoy.

Task 18: Game Engines

Game engines are software that provides development tools for the creation of video games. Game engines are widely used, by both professional studios and indie developers, and the variety of engines available is huge. They provide a framework for the developers to build on, with common features including rendering, physics, collision, sound, scripting, artificial intelligence and networking.

Game engines allow someone with little or no knowledge of coding languages to start building a game level. For professional companies, they provide a structure for the to build on, giving them tools to efficiently port games to consoles, or an platform in which to test models and environments while their own custom-built engine in still in development. (link)

The most commonly used type of engine is the FPS (first person shooter) engine. These engines produce a fully 3D environment the player typically sees from the point of view of their character, although plenty of third person games use FPS engines, too.

There are also engines which cater to much more niche types of games – for instance RPG engines that use tiled or isometric graphics, and 2D engines for point-and click games.

There are a lot of factors to consider when choosing between engines; while most of them share an array of basic features, they may have individual pros or cons such as the quality of the rendering technology, the ability to add middleware, or the adaptability to different consoles.

Unreal Engine

The Unreal engine is one of the most popular game engines, and quite likely the most widely used. Unreal Engine 1 first appeared in 1998, and the Engine is currently on its third iteration, with several games currently in production using the upcoming Unreal Engine 4.

Side-by-side comparison of a character from different versions of Unreal Engine, from http://en.wikipedia.org/wiki/File:Unreal_Engine_Comparison.jpg

Unreal Engine is the only game engine I have actually used, apart from playing around with TES: Oblivion’s construction kit, which is a heavily modified version of the Gamebryo engine.

The editor is free to download, and the engine itself is free to use for educational and non-commercial projects. Using it for a commercial project requires a licencing fee of $99, plus 25% royalties after the first $50,000 earned. The low entry fee makes Unreal Engine a viable choice for indie developers, although the high royalties may put off larger projects with much higher predicted earnings. (link)

One this that might put developers off Unreal are its graphical limitations. While they’re not bad, necessarily, the style of the rendering is fairly recognisable. There is also no real-time lighting, meaning lighting maps have to be used, which not only can slow down production time but can also mean lighting might not ever look as good as you want it to. Thankfully, unreal engine 4 is introducing real-time lighting.  That said, UE3 can still look pretty good, and there’s a whole host of games out there that prove it; Bioshock, Mass Effect 3 and Deus Ex: Human Revolution, to name but a few.

Unreal engine is also very customisable. There is plenty of  middleware available and the engine itself is very customisable. Most commercial games will be using a version that is heavily modified.

Cryengine

Cryengine is another popular engine. Developed by Crytek, the engine was first used by the first person shooter Far Cry and its various remakes and sequels. Cryengine 2 was then used for the game Crysis, before breaking the theme-naming streak by licencing out the engine for the development of several unrelated MMORPGs.

The big advantage that’s usually put forward in favour of Cryengine is it’s graphics. Cryengine is pretty, very pretty, especially when it comes to natural lighting. Since for a large proportion of gamers, appearance of the game is a top priority, there are predictably multiple videos and image sets demonstrating Cryengine’s graphical superiority.

Personally, I think it’s just a matter of taste. Cryengine has pretty bloom lighting and fog effects, and will help a lot with something that’s striving for photorealism, whereas UDK would work best for something that’s going to be a bit more stylised, especially since you can customise the shaders.

There’s more to an engine than graphics, however. Cryengine is a lot less flexible than other engines. It does have a lot more inbuilt capabilities when compared to UE, but much fewer optional add-ons. It also requires plugins to import assets, and online connectivity to use both the engine and editor. Cryengine also lacks vertex painting, which I the main reason my group chose not to use it for the queens building project. However, if you want to focus on your assets, it is a nice way to make the most of your 3d models.

Other notable enigines

  

The Unity engine is sometimes recommended as an alternative to Unreal. It’s not as good-looking as Unreal, but it’s more user friendly, especially when working in a team. Assests can be imported in blender or photoshop file format, so it saves time that would be spent converting them. Also - like UDK - it has numerous plugins that can streamline it further, and as a result can be used to produce content much more efficiently that in other engines. However, it does lack the primitives and BSP brushes that UDK has, so placeholder meshed will have to be used to start off with.

The Source Engine, created by Valve, is the engine used for the Half-life and Portal games. Like UE and Cryengine, it is frequently used by individuals bothe to mod pre-existing games and for non-commercial projects.

The Anvil Engine (formerly Scimitar engine), an in-house engine used by Ubisoft for the Prince of Persia and Assassins creed games, currently in its third iteration as of assassin’s creed III. It is specifically geared towards character animations and interaction with the environment, as would be expected from titles with such a heavy emphasis on platforming. The engine is not available to licence for companies, not is it available for private use.

Task 15: Visual composition

Composition, as a general term, can extend to cover every aspect of an image, including the use of colour and tonal values. More specifically, it describes how things are arranged within a picture, relative to each other and relative to the edges of the images.

Depending on how you define them, there’s any number of different elements to consider when composing an image, but one way the key components can be broken down is into line, shape, colour texture, form, value and space.

Line is an illusion created by the edges of objects. Many drawings styles use lines to define the boundaries of objects within an image, as even though lines don’t actually exist it’s the easiest way to convey form. Whether the lines within an composition are actual lines or merely the edges of contrasting tones or hues, they’ll be one of the most easily identifiable features.

Shape is self-explanatory, and refers to both individual objects and combined objects. Colour is deceptively difficult to use properly, it does little to define objects, rather creates mood, warmth or cool. It can be used to bring attention to certain areas of an image, but not as well as tonal value.

Tonal value describes how light or dark a part of an image is. Tonal value is arguably the most important part of an image, since it’s the most visible part. Form describes the 3-dimensional qualities of an element, which is expressed through shading and perspective.

Texture is similar to form in that it can describe 3-dimentional qualities within an image, except instead of describing the overall structure it describes the tactile qualities. Texture is the difference between a reflective surface and a matte surface or a smooth surface and a grainy surface. Mostly texture adds detail and realism to an image.

Space is the proportion of an image that objects occupy; conversely, negative space is the portion of the image surrounding them or between them. Usually the focus of an image with be the positive space – the object or person the picture is focusing on – but sometimes the negative space can be the subject, for instance if the image is focusing on a piece of sky between trees or buildings. 

So, how do these all work within an image?

There are eight elements of composition to consider: unity, balance, movement, rhythm, focus contrast, pattern and proportion.

There are established principles for composition; some things work and some things just look ‘wrong’. The easiest principle to follow is the rule of thirds. Elements within an image will look best placed a third of the way up an image, or a third of the way along. By dividing an image into nine equal sections, you can identify the four focal points.

Focal points are places in an image where you intend the viewer to spend the most time looking. There are certain things people’s eyes will naturally gravitate towards, such as faces and hands. More detailed areas will also draw more attention, as will bright colours and contrasting hues.

If there is more than one focal point, they need to be arranged so that they are balanced. To some extent, symmetry in an image makes the composition feel grounded, but can also make it look unnatural.  

 Asymmetrical balance, which uses elements which are equal, but different, looks much more natural.

Task 19: Sound for Games

Sound has been present in games from the very beginning of their history, and in some ways has evolved parallel to the graphics as the capabilities of the systems have improved. The ability to synthesise sounds has been around for quite a while longer than video games, but for a long time was primarily stored in an analogue format, which required delicate physical components and was therefore greatly limited by practicality. Digital sound was much more affordably, but for early handheld game devices, sound was limited to synthesised music with very few notes and subsequently basic tunes. PC games were also limited, because producing sound took a significant amount of processing power, and even in modern games audio is demanding.

However, better technology does not equate to better music. In fact, sometimes the simpler tunes are the most memorable. Plenty of basic tunes have accompanied game along their development, becoming synonymous with the games themselves. They evolved alongside the graphics into theme tunes which in some cases are still used in those games. Theme tunes are an easy shorthand to represent a game even if it goes through major graphics reinventions, and create some amount of nostalgia for long-time players.

My own gaming experience has been solely limited to PC games, which for some don’t seem to produce the same scope of memorable theme tunes. The nearest thing I’ve probably experienced is hearing the Elder Scrolls theme song in the Skyrim teaser trailer and thinking “well, about time”. The main theme music for Skyrim purposely uses the same melodies as the previous games. Subsequently, the lack of the TES theme music in the recent Elder Scrolls Online cinematic trailer has been one of the most common complaints – as people have rightly pointed out, without the signature music it could be advertising any generic fantasy game.

While sound for games is still system-intensive, the main limitations for quality are mainly down to skill and budget. This is part of the development of games into a large industry, with increasingly large budgets and production teams. There’s been a demand for the quality of the music to increase alongside the graphics, to a point where it’s no longer feasible for a single person to produce sound for an entire game.

Sometimes it’s not just ingame music or theme tunes that get associated with a game title. The Assassin’s creed: Revelations cinematic trailer featured a vocalised song by a group called Woodkid, which has become irrevocably associated with that particular game. You also get songs written specifically for the game, such as the songs during the end credits of the Portal games, which were written specially for the games.

Composing for video games is now not all that different from film, and some composers working in games, such as Michael Giacchino, also work in multiple other genres such as film and TV.

Music for games is increasingly gaining the recognition it deserves. In 2012, for the first time, the Grammy awards included video game music as contender for Visual Media awards,

For a large part, sound in games is used much the same way it is in film; to convey mood and to create tension. This happens right the way from the title screen – action games have heroic or upbeat title music, horror games will play something creepy to set the mood.

This works within the game too, on a more specific basis. Thematic or more neutral music might play when the player is just travelling around the map, but when the player meets an enemy more fast-paced ‘combat music’ will start playing, drawing them into the action. As in films, the choice of music and sound has a huge effect on players’ experience of the games, and really shouldn’t be underestimated. Amnesia, for example, is much less scary without the player character’s rapid breathing, but with the sound on that alone can get your heart racing, even when there’s nothing to be afraid of. 

Sounds can be used to convey information to the player much more quickly than visual information can, especially during situations where the player has to focus on multiple things at once. The combat music mentioned above also serves to alert the player to the prescience of enemies that the character would have noticed, even if the player did not. During combat, there will be different sounds for the player getting hit, and for the player landing a hit on an enemy, and even for different types of attacks. Having different sounds leaves the player free to focus on aiming and moving around, and only checking their stats when prompted.

Sounds might also inform the player about things happening in their environment, such as telling them a gate has opened, whether a lever or button has worked or not. Footsteps of enemies alert the player to enemies they can’t see, especially in first person games. Sometimes musical cues are also used to tell the player they’ve completed a task or finished a level.

Voice acting in games is a fairly recent development, although it’s quickly become expected in any large-budget title. It’s not particularly vital, since there’s little that can’t be conveyed through text, but it adds another level of realism and depth to the game. On the negative side, it limits the amount of dialogue and greatly adds to production costs. If done badly, it can be worse than having no voice acting at all. This is more a problem in games with dialogue options than in linear games where voice acting is scripted. It’s very difficult to make conversation seem natural, and if you play a game for 40+ hours anything the characters say more than once will become noticeably repetitive unless it’s only said once.

One way games get around this is to create a made up language, since if the player doesn’t understand what’s being said it doesn’t become repetitive. The most well-known example is The Sims, where all the characters speak ‘Simlish’ –  which has since been coined as a general term for any conlang used in games for the purpose of remaining ambiguous.

Other examples of this include Magicka, where – unlike in The Sims - the (mostly) nonsense words are accompanied by English subtitles. In this game, the use of simlish is mainly a way to reduce voice acting costs and to make the game easier to translate into other languages, although it also adds a certain charm.

Task 17: Documentation (part 1)

Overview – Space-themed Platforming Game

This brief details the creation of assets for a game with a ‘retrofuturistic’ theme. Retrofuturism is 20th century aesthetics applied to science fiction; it creates a style that looks both space-age and old-fashioned, and is instantly recognisable.

The game will be for PC, and be aimed as casual gamers and younger players. It will primarily be a platforming game with an emphasis on puzzle solving. The gameplay will involve interacting with the environment via various tools such as grappling hooks, magnetic boots and jet-packs.

The overall look of the game should be clean and bright, with emphasis on primary colours to emulate the ‘pop art’ feel of retrofuturism artwork.  It should aim for a simplified ‘cartoony’ look rather than gritty realism, but retain realistic character and environment proportions. To create and maintain a unified stylised appearance, all textures will be primarily digitally hand painted. The game will take place in a variety of different themed levels, so assets that will be present throughout the game - such as reoccurring characters, need to be versatile enough that they are suitable across several different colour palettes.

The game will be produced for the Unreal Engine, using the Unreal Development Kit. The models with be produced primarily in 3Ds Max, with Z-brush or similar software used to produce normal maps.  The textures will mainly be produced in Photoshop.

 

 Lead Character – Space Explorer

The lead character is a female space explorer. She will be wearing a space-suit with helmet that obscures the face. The space suit doesn’t have to look realistic by modern standards – in fact it is preferable that it is obviously influenced by science fantasy. The suit can show minor wear and tear, but should for the most part look clean and new.

The character model should value readability over detail, and be realistically proportioned - although some exaggeration is allowed to aid readability - with a primarily blue or grey colour theme and plus a highlighting colour such as orange if needed.

Since the character will be in the foreground of the screen consistently during gameplay, only one Level of Detail is needed.

The character mesh should be under 20k tris, and the mesh should prioritise the silhouette and use appropriate polygon density. Since the character will be very visible and will constantly be moving, the topology needs to be built so that it deforms smoothly when the model is rigged.

The mesh should be textured using the following maps:

2 1024x1024 diffuse map with alphas

2 1024x1024 colour specular map

2 1024 normal maps

The diffuse map should be hand painted in Photoshop, with photo references used sparingly to convey the different textures of the materials such as metal and plastic. Mirroring the textures should be avoided as should obvious symmetry, except in areas where it will not be noticed. 

NPC – Robot Helper

This character will be present in various levels, to be directed by the player. It will be mostly humanoid, although not necessarily with accurate proportions, and has to be visibly mechanical. Most importantly, the character needs to look like it could realistically move, with jointed segments where the limb bend.

The character mesh should be under 15k tris, and the mesh should prioritise the silhouette and use appropriate polygon density. Since the character is mainly made out of metal, the majority of the mesh will not deform when moving, so the topology should take into account where extra geometry will be needed for it to deform cleanly and where it will not.

The mesh should be textured using the following maps:

2 1024x1024 diffuse map with alphas

2 1024x1024 colour specular map

2 1024 normal maps

The diffuse map should be hand painted in Photoshop, and the normal and specular maps need to realistically display the different materials – metal, plastic, rubber, etc – that the character is made of.

  

Part Two 

Task 16: Level Design

At its most basic, a game level is the space a player moves around in.

In most games, the path of the player is actually fairly linear, and in simplest terms the level involves guiding the player from point A to point B. This means that when designing levels, the most important aspect designers have to think about is how the player will move through the environment, including how their movement will be directed or impeded by obstacles such as secondary objectives and enemies.

Many games only have a single path the player can take if they wish to advance the game. Sometimes this linear gameplay is obvious because the player is funnelled towards their goal by the scenery, whereas other times a seemingly open environment will have invisible constraints that force the player to visit certain sections in a set order with - most this often simply the player’s limited ability to jump.

Linear gaming like this is not necessarily bad, and such railroading is generally judged differently depending on the genre.  In platforming games especially, discovering the limits of these constraints is a large part of the gameplay, and with many combat-orientated games the player doesn’t want to be constantly getting lost while fighting enemies. (Conversely, some puzzle games will use combat specifically to make the puzzles arbitrarily harder – the puzzles in Amnesia take very little brainpower, but are difficult to complete while your character’s curled up in a corner crying.)

Non-linear gaming can take several different approaches, with different types of dynamic gameplay. Games such as sandbox games make every effort to give the player freedom, at least within the set parameters, while some games simply provide more than one path for the player to take.

Sometimes this is a feature built into individual levels, with multiple ways of solving a given puzzle. For instance, in Deus Ex each level environment is fairly open with multiple paths the player can take depending on their preferred play style, whether they decide to sneak around and pick off enemies one by one, or to charge in head first.

1)  Kill now or 2) Kill later

It could be argued that - by nature of giving the player control - any game is non-linear, as no two players will do exactly the same thing. However many games actively work against this, for instance by blocking off areas the player has already been through or killing the player if they try to go off course, so from the point of view of a level designer there is certainly a conscious decision.

It’s difficult not to think about the context of a level when you think about level design, partly because the environment will have a large impact of how the level feels to play.  A good demonstration of the difference between level design and environment design can be seen at the beginning of Portal 2.

The first few levels – the puzzles and the positioning of key elements - are often identical to those at the beginning of the first game, but with the overgrown scenery, open spaces and organic lighting, the environment is very different. Subsequently, while in Portal these levels felt almost claustrophobic, in Portal 2 the exact same level designs give you a brief sense of freedom and healthy dose of déjà vu.

Portal 1 vs. Portal 2

I only just discovered you can add captions.

Portal is in fact often taken as a parody of the idea of level design over story; highlighting the lengths games sometimes go to explain level design and puzzles within the context of an environment. The game makes little effort to disguise the purpose of the ‘testing chambers’ as anything more that the puzzles they are, leaving the developers open to fully explore the possibilities of the genre without having to justify everything. In fact, the developers published some tutorials detailing the process.

I’ve found that games often fall into one of two extremes – although that might be due to my own polarizing preferences for genre, since I tend to go for either RPGs or platformers.

On one hand you get games like Portal, where the puzzles seem to have been given the most leeway and the environment has been slotted in around it. In particular, I remember one of the Prince of Persia games where there are some very conspicuously placed tree branches. Sadly I don’t have a screenshot, but take my word for it that their attempt to incorporate the environment into the level design failed spectacularly. Games like Uncharted are also occasionally guilty of this, although they do seem to be getting better at it, perhaps because of more flexible game engines.

At the other end you get games such as Skyrim and other RPGs, where a large proportion of the gameplay mainly comes from the context of the story and the visual aspect of the environment. Not to say that the environments in these games aren’t somewhat concerned with playability; in my post about environment design, when discussing how level design is blended into the environment I talked about how most Skyrim dungeons provide a shortcut back to the entrance. However, in these games level design often takes a backseat to put narrative behind the wheel.

Following the path to the right and jumping down takes you back to the entrance.

That’s not to say context is entirely removed from level design. Some portion of the feel of a game still comes from the layout of different elements, and both environment and level design can lean on one another. Narrow spaces can create a feeling of urgency, or give the player a sense of objective, and large spaces can either give the player a feeling of freedom or leave them feeling vulnerable. 

Vertical space can also have an effect on the player, climbing up and climbing down invoke different emotions, and can be used to support the narrative. A player will probably perceive moving upwards as more effort than climbing down and expect a larger pay-off, even if the gameplay is almost identical.

Often the context of the environment inspires the level design, for instance Assassin’s Creed takes the concept of climbing buildings and weaves it - seemingly effortlessly - into the environment, the gameplay and the level design. Rather than using buildings simply as props or background scenery, the game takes the concept of a sprawling city and tailors the gameplay towards the environment. There are some conspicuous gameplay elements, but these are mainly minor things, such as boxes and flagpoles placed to help you move fluidly.

Much of level design is simply about balancing different interactions, constantly giving players something do without the tasks becoming repetitive or boring. This is where predicting the path the player takes becomes important, as it allows the level designer to make a vague timeline of player activity. Long distances between points translate to long periods of the player just moving around, and forcing a player to backtrack through the same place multiple times becomes tedious.

Of course, it’s impossible to fully predict how a game will feel without playing it through, which is where blocking out levels and play-testing comes into play. Creating fully fleshed out environments is expensive and time consuming, so to get a feel for the gameplay early on the development designers create a white-box, a very basic outline of the level without all the visual aspects. 

White-boxing levels allows play-testing almost from the very beginning of production, and most importantly the environments can easily be changed to improve the player’s experience. The developers get a better idea of what the game will be like to play, and when the art team begin to flesh out the visual part of the game, they know all their assets will actually be used.