Q1 SP design theories 1.

Introduction:
These two pages contain some detailed design theories, which suggest how to make a high quality Q1 SP map, by analysing the various aspects that contribute to high quality Q1 SP maps. They are written as general guidelines (which the exception of Theory 0 which is an absolute rule), and it's the overall quality of the map that is important, rather than fitting in with any specific suggestion here. They are also written solely from a player's (and a critical reviewer's) point of view, not a mapper's point of view.

Choose a theory:
Theories page 1:
» Theory 0: The basic golden rule of mapping.
» Theory 1: What makes a proper Q1 SP level???
» Theory 2: Common flaws to be avoided in Q1 SP levels.
» Theory 3: What makes a map look good???

Theories page 2:
» Theory 4: What makes a map play well???
» Theory 5: Other aspects of Q1 SP map quality.

Theory 0: The basic golden rule of mapping.

The other theories are guidelines, suggestions, advice, possibilities...this one is not. This is a rule - a rule that applies to getting reviews on this site, but also a rule that should be (but isn't) obeyed by any mapper anywhere. It's this simple:

    If you publicly release a map, that map must be good quality and worth playing.

A very simple rule that shouldn't even be necessary to state - really this should be taken for granted. But it isn't, it needs to be stated clearly, and there are some related details/questions that are worth considering.

» How can map makers know what good quality is???
There are a few possible answers to this:
1. Everyone has seen and played Id's own maps, so everyone has seen and played good maps. Id's maps are often flawed, and have been bettered by some custom maps, but most of them are still fine examples of good quality maps: the gameplay is invariably good, the architecture is often good. So map makers should be able to compare their maps to Id's maps, and if their maps are equal or better, then they would be good quality.
2. There are plenty of resources around the net to learn from. For example, the vast number of custom maps, both good and bad quality, the help tutorials and technical guides to mapping, and general map quality articles such as my own Theories. So if mappers are unsure what good quality is, there are resources they to learn from - as well, of course, as getting people to beta-test their maps.
3. If mappers don't know what good quality is, then perhaps they should not be releasing maps at all. This may seem harsh, but making a map is one thing, making a good map is another. If a mapper can make a map, but just doesn't know how to make a good one that will be enjoyed by the public when it's released, then perhaps they should not release it.

» What about first time mappers learning to map???
There are two aspects to consider here:
1. When a mapper is learning to map, and indeed learning to make a good map, they shouldn't publicly release a map until they've made a good quality map. When they make test maps to learn mapping, or to try to make a proper map, those maps should be kept on their hard drive, or perhaps privately released to acquaintances and more experienced mappers for testing and feedback, until they manage to make a good map - the good map is the first one they should publicly release.
2. Being a first time mapper is no excuse whatsoever for releasing a bad quality map. Some mappers mistakenly think it is, and try to make such excuses in their texts - sorry, but that excuse simply isn't acceptable. Firstly, the reason above, bad quality maps that one makes while learning to map should be kept on the hard drive until a good quality map is achieved. Secondly, there are plenty of people, from all genres of mapping, who have released good quality maps as their first released maps, some of these have indeed been excellent maps.

This rule may seem a very stern and judgemental one, but it is working towards a positive goal: people enjoy good maps, the better the map, the more they'll enjoy it. People don't enjoy bad maps - at the best it will bring them no enjoyment, and worst they'll be offended by the mediocrity and low quality. So any mapper should strive to release the best maps they can, for their own satisfaction, and for other peoples' enjoyment - strive for quality!!!

Theory 1: What makes a proper Q1 SP level:

And for my next trick, I shall <drum roll> attempt to... </dr> ...define the undefinable!!! Now I just need a volunteer from the audience... Okay, so trying to define what makes a proper Quake1 single player level seems a pretty fruitless task, but I have had someone ask me what he would need to make for it to be considered a "proper" SP level. So I've thought of some guidelines, that may even help people who are wondering if they are capable of making a Q1 SP level. Of course, these guidelines aren't actual *rules*, and some proper (and even very good) levels do fail to meet one or two of these, but I think they are pretty good "basics".
Please be aware as to exactly where these guidelines lie: they go beyond defining a mere "level that can be played in single player", but do NOT go as far as defining what makes a GOOD single player level. That is, any level with a single player spawn spot in and a room for the player to fit in can pedantically be classed as a single player level, but not by any standards a "proper" level - I would say the "proper" means it must offer single player gameplay and design. Equally, it is possible to make a proper single player level according to these guidelines, that is designed for and playable in single player, but is of very low quality and not worth playing in the slightest. Guidelines defining what makes a single player level "good" are another matter entirely and will be detailed in further theories.
The guidelines:

1. The level, in it's single player version, should be specifically designed for single player:
This is an fundamental guideline which means that the single player version should be intended to provide proper single player gameplay and design as detailed by the other guidelines, and further should succeed in this intention. This does mean that the level can be based on or converted from a different sort of level, for example a deathmatch level or even a "test" level, but that when it is released as a single player level it should comply with the guidelines below.

2. The level should contain an entrance and an exit:
That is, you should enter the level at a specific point, preferabley marked as an entrance, and ideally realistic (like a teleporter, door, canyon, tunnel, for example), and you should exit at a specific point, usually a different point to the entrance, and again marked as an exit, again realistic if possible. The level should require some travelling and/or task accomplishment to progress from the entrance to the exit.

3. The level should attain a degree of complexity, and require progression through it:
This expands on 2., and means that the level should contain some (i.e. more than one) areas (i.e. rooms, corridors, outdoor or underwater sections), connected together, that you must progress through to complete the map. Although there are a few maps that consist of single, tall, and multi-leveled rooms, these do have some (but not much) complexity in the room design. In general, the map should achieve complexity via multiple areas and complex area designs, and should require the player to visit a majority of those areas to progress from start to exit. Further, the areas and their layout should be distinct and varied, i.e. not just a series of identical rooms placed in a line.

4. The level should contain a "reasonable" number of monsters:
Where "reasonable" is errr, ummm... Okay, I personally would say a bare minimum of 20 medieval monsters, or 30 base/small monsters. The latter would include Rottweilers, Grunts, Enforcers, Fish, and possibly Zombies, and the former would include everything else. This is to ensure there is some worthwhile and continuous single player combat, which is of course the essence of Quake =). Also, these monsters should be used and positioned throughout the level, and not just dumped together in one room - this is all part of the level being reasonably complex and requiring progress through it. It may be possible to make a proper and even worthwhile level with less monsters, but they would have to be chosen and placed almost perfectly to be effective.

5. The level should be completable without cheating or altering console variables:
This means exactly what it says, the level should not require any movement, weapon, or protection cheats, nor should it require the player to change the game settings via the console such as "r_fullbright", "r_wateralpha" and such like. I'm sure no-one would think of doing a level where you had to cheat to finish. Related to this, there should be no areas in the level that the player can reach with normal movement and not escape from without using movement cheats. Also, there should be no areas where player can get trapped on the wrong side of a door or barrier, and not be able to open it again to proceed.

6. The level should be completable without "trick" moves or gameplay:
This really means that it shouldn't require the player to perform and moves that are not in the normal single player repetoire, such as rocket jumping, grenade jumping, monster boosts, bunny hopping or hard jumps off scenery. This is because these are specialist moves, and though fun, it is unreasonable to expect all players to be aware or capable of them. Equally the map should not require any obscure types of gameplay that are more suited to secrets than map completion - for example, hidden ledges to walk on, keys hidden in secrets, buttons in completely dark areas - what the player has to do next should be logical and reasonabley clear, even if it does require exploration. All of these tricks are fine for secrets of course.

7. The level should be free of serious technical problems that make it unfeasible to play:
For example, the map should not crash repeatedly, there should be no areas in the map that will automatically cause it to crash, the player must not be able to fall out of the level, the map should run at a reasonable framerate (I would say a minimum of 20 FPS), any triggers required to exit should work without fail, any bugs the level might have should not make the level unplayable or impossible to complete, and ideally the level should run smoothly in Quake, WinQuake and GLQuake.

8. The level should be completely, well, completed:
This means the level should be all completely made, with no missing or unfinished sections, it should have been fully complied and "VISed", and should be ready to play. Also it should be adequately lit, and all items, weapons and monsters required for both gameplay and completion of the map should be present.

9. The level should be adequately designed and themed:
This basically means the level should not consist merely of some unrelated and disparate rooms connected together. There should be a noticable theme (even if that theme is very vague) that runs throughout the level and shows the each area is part of a whole level (unless there is a clear reason why an area deviates from the theme). The design itself should indicate that the level was designed or at least put together as a whole level.

10. The level should not contain and major sections copied from other maps:
This means sections or designs should not be directly copied (i.e. unmodified) from any other custom or id maps, SP, DM or other. This is basically to protect the property of other map authors, therefore the author's own work could be copied into the map - probably not advisable though. Tentatively, it might be possible to have a proper SP level that includes sections of other maps that are noticably modified and therefore not identical to the sections in the original map - I would not recommend this though, and originality should be considered important.

Theory 2: Common flaws to be avoided in Q1 SP levels:

Just a few things that are common flaws in single player levels, and should be avoided, roughly in the order of importance. Obviously this is just bad stuff to avoid, and doesn't cover good stuff that will really enhance the map, and further, it is a limited selection of the most common and off putting flaws - map makers should strive for quality in all areas of their maps.

1. Level is has no/poor theme:
Extremely important flaw that can put players off right from the start, and alas, fairly common. This can vary from an properly themed level containing some disparate elements, to a completely unthemed made from seemingly random styles. Theme basically involves the overall look and style of the level, and in particular how convincing and coherent it is as a "fantasy" place. Thus, without a strong theme to hold the level together, it will often look like an incoherent mess, with no sense of "place". And who wants to play through an incoherent mess??? A strong theme means that all the architectural, design, detail and textural elements of the map work well together. One thing it does NOT mean is having to stick to standard or recognisable Id themes - a strong theme can be any sort of theme, no matter how unusual or vague (pseudo-medieval cybergothic, anyone???), as long as the level does that theme well and sticks to it.

2. Level is too hard to start and too easy to finish:
Hugely common flaw that affects at least 90% of levels, to a varying degree. The thing is, when you start a level, you have very little weapons, ammo and armour, and usually little room to move and flee. While at the end of a level you will be tooled up with some serious firepower and armour, as well as usually having most of the level to manoeuvre in. Additionally, when you start a map, you need to not lose too much health, while at the end of the map, that doesn't matter, as long as you survive. All in all, it's essential to take time to ensure the level starts off fairly gently, and progresses smoothly (with a few mini-climaxes) to a big, spectacular (and challenging) ending.

3. Poor and ugly architecture/texturing:
It's too common that people think they can get away with ugly looking levels clad in shoddy textures. Well, they can't - people need levels that look good, because the enviroment the gameplay occurs in is intrinsic to the enjoyment of the level. Therefore it is essential to ensure the map looks good (see Theory 3: What makes a map look good for more details). Flaws to avoid include: unthemed and random looking designs, overuse of right angles, inconsistent use of angles and curves, mismatching sizes and proportions through the level, ill fitting textures, inappropriate textures, textures and lighting that obscure designs, visual glitches and bugs - in general anything that makes the look of the level fall below the standard set by most Id levels and better custom maps. Additionally two other architecture related flaws to avoid are: slow running and choppy areas (which indicates little concern for playability), and designs and details that hamper player movement. All in all, poor architecture should be avoided at all costs.

4. Low quality and inappropriate modifications/extras:
Specifically, new monsters, weapons, sounds and skins. Some mappers like to include such extras to spice up their levels, however there are many bad quality modifications that are both noticably inferior to Id quality, and very clearly and painfully out of place with the Quake themes and in-game monsters and weapons. Thus when bad quality modifications are used, they spoil the overall thematic effect of the level, and reduce it's quality, as well as being unpleasant to use/face in themselves. Thus it is crucial to ensure any modifications used are (like maps) at least Id quality, and fit in perfectly with the visceral, grimy, oppressive style of Quake and it's features. See Theory 5: Other aspects of Q1 SP map quality - Part 5: Modifications and Extras, for more guidelines.

5. Lack of skill settings:
As mentioned in Theory3, the balance between challenge and fun is a very fine one, and because players have different playing skills, skill settings are very important. If a map does not have skill settings, it is quite likely that it will be too hard for some players, or too easy for others, or both - and there will be nothing the players can do to make the level more suitable for them. Therefore distinct skill settings should be included.

6. Gameplay situations that are hard/impossible to survive first attempt:
Dying in a situation that you couldn't possibly react to first time, and having to replay that section to survive it, sucks, and sucks bad. It's very easy for level authors and beta testers to become used to a map, know everything that is going to happen, so can react to everything perfectly, while for the player facing the map for the first time, it can seem nearly impossible. Ways to avoid this include paying close attention to the comments of first time beta testers, giving good indications of surprises and traps, and making the map challenging in other ways that don't rely on surprises and reaction-dependant situations.

7. Instant death traps:
They suck, and they suck bad. Without exception, instant death traps that kill you without warning, or without a lot of time to react, spoil the gameplay of a map. They force tedious restarts, repeated saving, and often cause you to have to replay sections over and over again. If you use a trap, either make it so it damages you a bit (no more than 50% health as an absolute maximum, preferably much less), or gives you chance to react and escape, or releases monsters right in your face - the latter can be a lot of fun.

8. Obscure progression or trick gameplay:
This includes any gameplay that deviates from the normal reptoire of single player gameplay (i.e. running, shooting, dodging, finding keys and switchs, opening doors and minor exploration), in such a way that is unclear and obscure enough that it is likely to cause frustration to players. Examples to be wary of could include: excessive exploration, complex puzzles, necessary keys or switches hidden in secrets, hidden ways to progress, reliance on trick moves (grenade or rocket jumping, for example), repetitive jumping or movement, overuse of awkward scenery to hinder the player. It's worth noting that some of these, when used well, can add interest and thought to gameplay, but they can also be frustrating, so they need to be used with great care, if at all.

9. Monsters not used well:
All monsters have clear strengths and weaknesses, and thus they can often be used ineffectively, or sometimes, far too effectively. Some problems to avoid would be: Any close combat monster that cannot reach the player (obviously rather important in Quake, particularly with Fiends getting stuck in cramped areas), monsters that are guarenteed to fight each other placed in close proximity (especially monsters with area effect weapons), Rotfish that the player can see before entering water, Ogres above or below the player where their grenades cannot hit. Monsters being too efficient to be fun could include: Most monsters placed right next to player in cramped areas, Spawn placed where the player cannot avoid their attacks, Vores and perhaps Shamblers in areas with no cover (depending on weapons), and Grunts used in excessive groups (their shots can be very hard to avoid). So level authors need to be well aware of each monsters characteristics, and use them carefully so that they are challenging and effective, but still fun to fight.

10. Secrets either not present or used not used well:
Secrets are a fun part of the map, it can be cool hunting around and then feeling really clever when you find one. Also, they reward exploration, and encourage players to use their eyes and brains as well as mouse1, so maps without any secrets are missing out a bit. The best kind of secrets require some exploration and involve some travelling, rather than just a hole in the wall with a different texture over the top. Also, what is in the secrets is very important: Firstly, it must not be anything that is necessary to complete the level, like keycards, as that would just be stupid. Secondly, the items should be chosen with care, they should give the player an advantage, but not make the level too easy. For example, items such as Quad, Ring of Shadows, and 666 should be made really hard to get, while stuff like MH, extra ammo, bits of health etc, can be made quite easy to get without ruining the gameplay balance.

11. Levels being just too hard in general:
Challenge is good. Exciting gameplay is good. Situations that you have to struggle and try hard with are good. But levels that are consistently too hard, with too little health and ammo, too many monsters and traps, and awkward situations, are bad. They are just not fun, and gameplay that is too hard can really spoil a level. There is actually a reason behind this - the reason why a level is better being too easy rather than too hard: When a level is too hard, the player is struggling so much to survive, that they won't notice other good stuff - like designs, clever missions, architecture, atmosphere, lighting, enemy placements, etc etc, while if the level is too easy, they might not be having as much fun with the gameplay as they should, but they will still be able to notice and enjoy all the other good stuff. Of course, being too easy is a problem too, and the ideal is to get the gameplay just right, so they will have be having a thoroughly fun time that is both challenging and reasonable, as well as soaking up all the other good stuff in the level.

Of course all of this theory only focuses on negative aspects to be avoided in levels, and packing levels with positive aspects is far more important - so the next theories will cover that in depth...

Theory 3: What makes a map look good???

A simple question I was asked the other day - with a complex answer. The answer is: a LOT of different aspects of the level's overall architecture, all working in unison. In this theory, I'll present all the aspects *I* think make a map look good - but all of these have been confirmed by several players and map makers. However, I don't make maps, so this is purely from a player's point of view, though I have gone into a fair bit of depth (but obviously not as far as stating my personal architectural preferences!). I've divided the aspects up into various sections, but these aren't all distinct sections, and there is some crossover.

1. Theme:
A sometimes overlooked aspect of architecture, but an extremely important one for the overall feel of a map. Basically, theme means that whatever style you chose for the map, the map sticks to that style, with no areas that contradict that style unless there is good reason. For example, a base map should be consistently base, a gothic map should be consistently gothic. Theme is about making each area of a map look like part of the whole map, rather than an added extra. However, the theme is not restricted to standard or clearly defined themes - a unique, blended or vague theme can look great, as long as the map sticks to it.
» Consistent colour scheme and textures:
A limited palette of colours and textures should be chosen for the map, and shouldn't include any that are glaringly different or out of place (except for details). It also means chosing textures and colours that look good together. It does not mean chosing a single colour or texture for the map, of course, but using subtle variations instead of clashing ones.
» Correct and consistent proportions:
Each section of the map should be made at an appropriate scale - not too big and not too small. Of course, a map can include very big and very small sections, but each section should feel right for it's designs, details, and purpose. It also means applying the correct proportions throughout the map, so you don't have two areas with similar designs but wildly different sizes.
» Consistent structures and designs:
A certain style for the structures and designs should be followed throughout the map - not using exactly the same designs, but having some link between them so they look as if they all fit into that map.
» Appropriate designs:
Correct designs should be chosen for what the map, and each section in it, is supposed to represent or be. For example, a cavern should have appropriately angled and random rock, a bastion should have an oppressive structure and appearance of solidity, a base should have hi-tech designs. Of course, some maps may have no real world correspondence, but for whatever strange reality they are part of, they should be designed appropriately for it!
» Appropriate fluids and sky:
A small point but worth paying attention to. It means choosing the right sky and liquids for the style of map, and for the atmosphere you want to create. It means avoiding such faux pas as clear blue water in a grimy dungeon, lava surrounded by wood, and such like.
» Doing what Quake does best:
An important and underrated aspect: the Quake engine and overall game design does certain things a lot better than others: Quake does gothic, grimy, brooding and solid designs very well, it does airy, cheerful, bright, and delicate designs very poorly. Additionally, Quake does not do real life™ designs at all, they just don't work. So sticking to what Quake is best at will make a map look it's best.

2. Designs:
This is concerned with the main physical structures within the map: i.e. the rooms and the major designs they contain. This is the body of the map, and although simple structures and designs can be compensated for by excellent themes and textures, it's really important.
» Structures:
Structures are what map designs are made of - hard to explain, but it's basically *stuff*, pillars, archs, buttresses, sticky-outy and sticky-inny bits, ledges, roof designs, stairs, altars... all sorts of things. As long as these fit into the map, AND comply with all the other guidelines given, the more the merrier, as they provide a lot of visual (and combat) interest in a map.
» Setpieces:
These are the impressive bits of architecture, the stuff that makes you go "wow!". A setpiece is basically a main room of a map, with the best architecture and most impressive structures (and often important combat). Most good maps have setpieces, with plainer sections inbetween - simply because designing and making impressive structures throughout the entire map is hard to do. So having some major setpieces is good, but also having lots of minor setpieces is equally good - in both cases the setpieces should fit in with the overall style of the map.
» Angles and curves:
Right angles and square designs suck - that is a proven scientific fact. Well, okay, it's not that bad, but angles and/or curves should be considered essential for a good looking map - there is just something inexplicabley better about well angled structures. This includes such aspects as angled off corners, angled joins where walls meet floors and ceilings, angled corridors, angled buttresses, well angled outdoor sections (very important) and many more. And of course, in a lot of these areas, curves can be used too, also on pillars and arches. Additionally, the angles and curves should be consistent, there should be some similarities between the angles in each area, what they are and how they are used, to fit with the overall style of the map.
» Varying room dimensions and designs:
All the rooms and the designs they contain should not be too similar. There should be strong links to ensure they look like part of the same map, but equally there should be many varying room layouts, dimensions and designs, to provide continual visual interest.
» Complex room designs:
Most of the rooms and sections of the map should have some complexity, and not just be all square 2D rooms linked by corridors. There should be some 3D designs, plenty of structures that contribute to complexity, and floor plans that are more complex than a square or rectangle. Again this is to provide continual visual interest.
» Solidity of structures:
This relates to "Doing what Quake does best", and means that structures should look appropriately solid for their style and purpose - and for most Quake designs and maps, this will be noticably solid and heavy, rather than delicate and light. This makes the structures look more imposing and impressive.
» Clean and well finished designs:
An important aspect to do with the way designs are finished: they should be neat, they should be free of areas that look obtrusive, mismatched or stuck on, they should look part of the map rather than just placed in it, and they should be well made - this applies to all the designs in the map. Good designs that are finished scrappily just won't look good.

3. Details:
This is concerned with smaller stuff than the designs - basically what gets added to the major designs and indeed the rooms as a whole to spice things up, add more eye candy, and just make things look more detailed. This includes stuff such as lights, vents, trimming and edging, icons and "pictures", insets, grilles, machinery and computers, and unidentifiable stuff that is just there to look good.
» Fine detail to large detail balance:
A balance should be struck between structures used as large scale details, and small scale details that go on those structures to enhance them. Both of these sorts of details need to work in harmony, without the large details overshadowing the small ones, or too many small details obscuring and detracting from the large ones. Once the structures are in place, fine details should be added carefully and slowly until a balance is acheived.
» Overall detail balance:
The overall level of detail should be fairly consistent throughout the map. Although there should be some variation in details used and their amounts, there shouldn't be any areas that clash due to one area being hugely detailed and an adjacent area being completely plain. Additionally, thematic links through all the details used are good, giving the map some consistency.
» Appropriate details:
The map should have the details appropriate to it's overall theme and style, for example, hi-tech details in a base map, gothic details in a metal map, primitive or no details in outdoor areas. Of course, by blending styles and themes, the boundaries of what details are suitable get blurred or removed, but they should still look fitting for the map.
» Textural details:
A vast amount of detailing can be done through texturing, and this can be hugely important in making a map look good - rather than having details "stuck" on to larger structures, using textural details can make the details seem like a varying part of the structures. All the guidelines above apply to textural details, and texturing in general is covered below.

4. Texturing:
If the designs of the map are it's body, the texturing is it's skin, and since it covers everything in sight, it needs to be done good for the map to look good. The relationship between textures, structures and details is also very important, and these 3 aspects must work together and compliment each other for the map to look good, rather than clashing or obscuring each other. Equally, texturing is very much related to the theme of the map.
» Carefully chosen textures to suit the map:
Going along with the theme and overall style of the map, all the textures should be chosen so they create/enhance/match the theme. In fact, textures are arguably the main aspect of a map's theme, so to make a map look good, they really need to work well together. Of course, as with theme itself, this doesn't mean sticking to standard or stereotypical texture combinations, but just ensuring all the textures work well together and given the map a strong, unifying theme.
» Properly applied and aligned textures:
Related to a map being well finished, but going even further than that. Basically, all textures used should neatly fit the structures they are used on, and that their edges should line up exactly with the edge of those structures, and any nearby structures or details. This also includes the important aspect that the textures should be chosen to fit the structures: textures that are of a clearly fixed size (crates for example) should never be chopped off or stretched, but only be used on area where they fit.
» Plain textures for large areas, detail textures for smaller areas:
A good balance needs to be struck between details and plain areas: large areas covered with a very detailed texture look hideous, while large areas with small and well done areas of detail look much better. This goes along with the "Fine detail to large detail balance" point - plainer textures on large structures and areas allow those structures to look more impressive than when they are obscured by a very detailed texture. So detailed textures are best used for just that: details.
» Trim textures around edges and door:
Something small that adds greatly to the overall look of the map: if surface (i.e. wall, floor or ceiling) textures are just chopped off when the surface stops, like at the edge of a ledge or doorway, it looks crude and unnatural. Lining all such areas with a trim texture enhances the quality and detail of a map. This also applies to detail areas - often these look best with a border around them.

5. Lighting:
Lighting wraps up the whole level, everything in the level is affected by lighting, so although it's perhaps not as important as the above aspects, it really can make a good looking level look great.
» Appropriate lighting:
Another theme related aspect, the lighting should be chosen and used to enhance and compliment the maps theme. So for example, futuristic maps should have brighter and more regular lighting, medieval maps should have darker flickering lighting, outdoor daytime sections should have fairly bright and constant lighting, while outdoor nighttime sections should be much gloomier. Obviously the physical type of lights used should match the maps theme (this falls under "Details" though). As with all theme aspects, creating a unique lighting style to match a unique theme is great, as long as the lighting fits in.
» Realistic light/shadows given light source:
Following from the above point, the lighting should produce an appropriate level of light, and appropriate shadows, given it's source. Really this is just a matter of common sense - it's clear how light and shadows look in everyday life, so they should look pretty similar in a map. Getting this right adds greatly to making the map look convincing.
» Ambient and sourced lighting:
Two different sorts of lighting, and taking advantage of both makes a map look it's best. Ambient lighting is a general and non-specific lighting, that doesn't seem to come from anywhere in particular. It's not particularly realistic, except perhaps in open rooms or outdoor sections, but it is essential to allow the player to see. Sourced lighting is lighting that clearly emenates from a source, like a neon strip, a window, or a torch, and casts distinct patches of light onto surfaces and structures. Used correctly, it adds detail, and looks realistic and dramatic.
» No fullbright or fulldark areas:
These suck in general: they are unrealistic, they don't enhance the maps architecture, and fulldark areas are annoying or impossible to play. There are very few uses for them, but occasionally a small section of temporary fulldark can be used well for surprises and drama.
» Use of lighting for drama and ambience:
Going beyond just having good lighting to suit the map, lighting can be used as a tool to really enhance the feel of the map and it's gameplay. It's hard to specify what makes good dramatic lighting, but carefully used contrasts between light and shadow, and use of light to highlight certain aspects of the map, would come into it.

6. Other aspects:
There are a few aspects that aren't really architecture, but can affect the look of a map, so should be taken into account.
» Monsters usage:
Well, the monsters are a visual aspect of the map: you see them before you ruthlessly slaughter them =). A few things that can help the monsters make look good would include: Choosing the right monsters to go with the theme, using symetrical and organised monster combinations (for example, squads of Grunts, or having two Ogres on a ledge instead of one Ogre and one DeathKnight), and avoiding custom monsters unless they look excellent and fit in perfectly with the Quake style.
» Originality:
Well, strictly speaking, as long as the architecture is as good, an unoriginal map will look as good as an original one. But aside from originality being a pretty essential feature of a map anyway, an unoriginal map will not have visual interest for the players as they will have seen it before - thus an original map will have greater visual impact.
» Atmosphere:
Actually, a lot of the architectural aspects above contribute to atmosphere, but also atmosphere can enhance the overall experience of a map, including the visual aspects, just by immersing the player in the map more, making them feel as if they are in a place - and if they feel that, they may well pay more attention to that place.

Continue to Theories 2 page.