Joined: 19 May 2005
Posted: 02 Feb 2007 03:01
I've already posted this over on tracker - it's to give all of you thinking of making a map some (hopfully) useful tips for before you even open an editor.
I thought I'd repost this here for exposure and also to see if the devs agree or disagree.
Americas Army Mission Design Philosophy
Ok the first thing you're going to want to know is who am I to be offering advice when my first map isn't even in beta-testing stages yet? So allow me to take a few words to fill you in on my background and why I feel comfortable relaying to you information I've gathered, as well as adding some of my own thoughts to this tutorial in how to approach making a dev-quality (or better) mission.
Although I work in a legal firm as Head of IT, my longest-term friend and best man at my wedding is lead game designer for a game studio in Brighton (where I live) that has had several No.1 AAA titles on the PS2, and is currently developing their franchise for the PS3 and beyond. Also, the guy we used to share a flat with went on to become head 3d modeller for another game studio down here in Brighton, who released a series of popular games across multiple platforms.
When we got together one of our favourite pastimes was drinking whiskey and throwing around ideas for new games, new levels for existing games and similar. This has taught me an awful lot about the philosophy of game design and map design in particular. A lot of what follows stems from the coversations I had with these professionals.
To add to that, myself and a group of friends spent over a year methodically play testing *every single* custom map that came out for the game Call Of Duty and it's expansion pack United Offensive. By that I am including every release from pretty much every site, including foreign language ones. I saw some great custom maps, some pretty ones, and an awful lot of very very bad ones. After this we both began playing AAO and haven't really looked back since. It's also why we knew that to begin with after the editor was released we shouldn't expect too much in early days.
Now, I've played a good number of the AAO custom maps that have been released now, and some of them are great, some of them are terrible yet almost all of them have a great deal of potential.
There appears to be, however, on most (but not all) maps out so far a critical element missing that for all their faults, the devs have got right, and as a mapper you could consider more before releasing your map as a final. That thing is environmental realism.
There is, I think we would all agree, a balance to be struck in a game like Americas Army. It's not a simple 'scales of justice' 2 way balance either, but a three way balance. Another way to think of this is that at every stage in your map design and building process, every decision should be weighed on three scales, and if any one of those three is a great deal heavier than the others, the decision should be re-thought.
Those scales are 1) Realism, 2) Gameplay & 3) Technical Achievability.
Let me cover each one in more depth and then explain what I mean about weighing them up against each other.
This is the part which whilst not exclusive to Americas Army, is a lot more important in this game than in, say, Unreal Tournament, Quake, or even Battlefield 2. It's also one of the harder parts to get right, because it stops you from throwing some ideas together and saying 'tada!' to get a quick map out. Research and common sense are your allies here.
By research I mean both the 'concept' of the map and the design. For instance - if you wanted to set a map in a school (an idea I've heard discussed several times on this forum and others) the first question you should ask yourself is why would the US Army / Special Forces be asked to assault or defend a school instead of any surrounding buildings? Read some military news, look on wikipedia for famous or controversial engagements, go to the various gulf war vets' forums and read their experiences. If your idea has any basis in reality, you *will* find a reference. One thing history is full of is wars, and consequently war stories. I'm not saying you should ape whatever you find letter for letter, but if you have a clear idea in your head of why your teams are doing what they're doing in the mission, it will subtly alter your choices of everything from decor, to layout, to player spawns, to objectives to weapon loadouts. That difference will in turn mean the people who eventually play this mission will 'feel' it the way you do, and will immediately set your mission apart from the crowd.
The key question then is: What takes a nice idea for a design (let's stick with the school setting) and makes it into a nice idea for a mission?
Therein, in a nutshell, lays the crucial point. In AAME we are not editing maps, but missions. As far as I'm concerned, if I want a million and one custom 'maps' I can go to UT, Quake, Call of Duty, even Counterstrike. What should and will set AAO apart is the 'real world' aspect.
This leads us nicely into the design part of realism.
I have seen so many maps so far which describe themselves in similar terms to this: "Large open area, some small rooms for CQB, a basement, some vents yadda yadda yadda"
Now, having spent a fair bit of time designing in UnreadEd now, I can understand why, to the mapper, their map can become a 'collection of stuff' to them. Adding each new element presents its own challenge and when thinking about your map it can sometimes feel like (and look like in some) all you did was bring a few room brushes together with some window and door brushes to make a 'combat area' with a couple of objectives plonked in for good measure. This is not good AAO map design, however.
Let's talk about research again. Going back to the idea of setting a map in a school, now we have the 'concept' gained through historical precedents we looked into last paragraph, how should this effect our design?
Well, the first thing I would do is google for some floorplans and pictures of the buildings you want to create. For example, a quick google for 'school floorplan' gave me this picture.
Now in that picture you can see that there are a lot of rooms, and a good mix of large and enclosed spaces. You can either find one you like and copy it, or look at a few, and design your own from the core design elements you see again and again. Note we haven't even fired up and editor yet, and we *already* have a good idea in our mind of the look and layout of our map. This will be key later, as a strong idea of what you want to achieve will make it much easier for you to guide the editor instead of the editor guiding you.
At this point, I would recommend that you buy yourself some graph paper and pencils with the usual cheap math set (ruler, protractor, pair of compasses etc etc). Why? Because if you're anything like me - when you are at this stage - with the idea for the map firmly embedded in your mind and layout ideas coming thick and fast you're not always going to be at the computer when ideas strike, and even if you were, it's easier to sketch out ideas on paper than in Photoshop.
This is the point where you ask yourself, given my layout and my concept, what will make my environment different to an undamaged building, if anything? Let's say your concept is that a terrorist has been hiding out inside a school, and both teams are trying to locate and/or rescue him. It's just happening; maybe the kids all ran away when the first shots were fired, there was a crushing rush to make it to the fire escapes. In this instance you'd expect to see desks pushed over in the rush to escape, lockers fallen in corridors, maybe ovens left on, some blackboards abandoned, maybe showers left on, perhaps even a small fire where the staff left their cigarettes in the ashtrays as they scrambled out of the staffroom. In other words, this is not going to look like a school just after the cleaners left, it's going to look like what it is - post evacuation chaos.
Sketch your best ideas out, make scale drawings (thank you ruler and compass) of little areas of layout. Think about what you would see if you walked into a school just after an evacuation, and put all that in your sketchbook.
Now, if you've followed the ideas presented so far, you're already ahead of the curve compared to (a conservative estimate) 90% of the mappers trying their hand with AAME, and you've not even had to actually open any editor to do it.
The aspect of realism does not end here - as you move on to actually make your map, keep referring to your sketchbook. Don't feel you have to include every idea you've had, and don't be afraid to nix ideas that don't work as well in the editor as they did on paper. Every so often, maximise the perspective view and position yourself as if you're walking into a room at head height, look around. Does it 'feel' real to you? If not - what 'feels' out of place?
Some more bullet-pointed tips for 'realism'
* Walls are rarely just walls in real life. They have alcoves, windows, dado rails, light fittings, things to break up the monotonous straight lines. Show me a plain single-textured wall, and I'll tell you we're in a prison camp.
* Furniture is not placed solely to provide access to vents or ledges. It's generally placed in rooms to provide utility to the people who originally use the room. Sofas tend to face TVs or windows. Beds tend to have bedside cabinets. kitchens are arranged so the cook can move easily from prep area to ovens to serving area. etc
* Rooms are rarely placed at random. You will rarely find a toilet next to a kitchen, or a living room opening up to a garage. etc
* Windows tend to be larger on walls which will catch the sun (east-south-west).
Now we have a realistic concept, a basic realistic idea for a design/layout, how can we make sure our players don't just think they're running about in a beautiful oil painting?
This is the gameplay aspect. There are several facets to gameplay which we would normally consider together. However, given that Americas army does not rely on powerups, many weapons or other interactive objects like healthpacks, we'll concentrate on gameflow.
Gameflow is a term used to describe how you expect actual players, when they play your level to move about the locations and where you expect them to engage each other. (Note that as a designer you can not CONTROL how your map is played, only GUIDE.)
Generally a map can be dissected into two 'types' of areas. Let's call them 'Stadiums' and 'Thoroughfares'.
Stadiums are where you expect the 'firefights' to be at. Generally speaking you want them to be more 'open' than thoroughfares. By this I mean they should have multiple ways to enter and exit the area. Some examples of stadiums would be the Main Pump Room in Pipeline (6 entrances/exits: Prime Valve, West Corridor, Main Corridor Intersection, Bathroom Corridor, Vent, Main Control Room) and practically anywhere on maps like River Basin or HQ Raid. Stadiums will also generally contain your objectives, which guide the players to these areas best suited to tactical firefights.
Thoroughfares are the rooms/corridors/vents/outdoor routes by which your players reach the stadium areas. These generally have less entrances/exits, but there are obviously more of them. Remember that as they are 'closed' areas, it's wise to provide at least some cover in them, as there's nothing more annoying than being on your way to an objective or firefight, just to find yourself facing an enemy with no way to back away or get cover. Again using Pipeline as an example: Think East Maintenance Tunnel. Although this is an empty corridor, you have an 'elbow' to corner, steam in one half and vertical pipes to dodge behind approaching prime valve.
However, on the subject of cover, remember to keep it realistic. A school is just simply not going to have large oil drums hanging around in corridors in real life, so you have to be creative. Think lockers, portable bin-carts from the cleaner who left in a hurry, etc - you get the idea.
Finally, you should consider scale. If you assume most rounds take 6-8 minutes on a medium-sized map - how much can you realistically expect people to do in that time? It might fit in with your plot to have assault search all 50 rooms in your school for your terrorist, but in the times allowed, that's just not realistic, it would never happen.
So consider placing your objective near the centre - or opposite side of the map to assault. That way there's a clear direction assault want to travel and it's easier to choose appropriate stadium areas.
It also follows to be aware of providing *too many* optional thoroughfares. If this is the case, you run the risk of assault and defence running right past each other and the timer running out before a bullet is fired.
I'd say a good rule of thumb would be to have 3-4 'stadiums' (not including spawns) on your map and maybe 2-3 different ways of moving between them. Bear in mind the number of stadiums is probably more flexible than the number of thoroughfares for each one.
Also consider that people like to play a single map they like over and over on Americas Army. So include some details if you can which 'reward' replaying the map a lot. Maybe some random-looking debris that allow you to reach a vent system it wasn't obvious you could reach. Maybe an innocuous grate which you can actually lift and make your way into a little tunnel taking you somewhere it would otherwise be difficult to reach safely. This will give your map lasting appeal and ensure it stands out above the rank-and-file of other maps.
Penultimate word of warning on gameplay: In AAO there is *never* any reason to have a 'dead end' on your map. Trapping someone in a dead end is no fun for the trapped and unsportsmanlike for the trapper. Every single point on your map should have at least 2 entrances and exits.
Absolute final word of warning is 'dead zones' in the map. These are areas where there is no good reason to visit them for a player IE they don't lay between 2 objectives, they aren't on the way to objectives from spawn and they don't provide a decent line-of-sight to a stadium or thoroughfare. These can come about by sticking too close to the floorplan you saw when researching your map, or just objectives which you decide against in the final build and remove. Either way my advice for any zones like this would be block them off and forget about them. No point giving yourself heartache carefully modelling and mapping an area no one will ever see. It also means players who are new to the map are less likely to get 'lost' and frustrated.
3) Technical Achievability
Ok the final point to consider when at the design stage of your map is the technical achievability. Probably not a real term as I just made it up. What it means it: There are some things you will *never* be able to make into an AAO map. NO matter how great the concept, no matter how perfect the layout would be for great gameplay - it just wouldn't work within the constraints of the Unreal 2.5 Engine, moreover - it won't work because of the particular constraints of the AAME specifically.
Some examples - you will never be able to swim between objectives. Although the unreal engine supports water, it doesn't really work in AAO: there's no player animations for swimming, for a start.
Having a level in the middle of a laser show would be impossible as it would rely on too many dynamic lights. Although the engine 'theoretically' supports dynamic lighting; it's implementation in AAO is buggy at best.
Of course, another factor is that in AAME we are limited to using just the meshes, textures and sounds provided to us by the devs. This means I doubt there'll ever be a full-on cathedral map, or one set in a complex of igloos. Basically, if it can't be made with the materials you have provided to you - it probably can't be made.
Edit: It turns out you *can* fudge your own static meshes as demonstrated here: http://forum.americasarmy.com/...=247309 (credit: zait) but you still have to make them from BSP brushes and existing textures.
Other restrictions are the removal of the UnrealScript editing, 2d shape tool (used to make complex brushes) and a few more besides. You can find a more comprehensive collection on forums.americasarmy.com . Basically - if your map idea relies on being able to do something you cant do within AAME, it's time to think about how to adapt it to make it workable, or even scrap the idea altogether.
Finally, in a first person shooter game like Americas Army - FPS is king. And by FPS I mean Frames Per Second. The entire final testing stage for your map should be Frame Per Second run-throughs. As a rule of thumb, you should never ever let the FPS in your map (which you can see by pressing F6 by default) drop below that which you get when playing Steamroller. Obviously this will be different depending on what hardware you use, but if SR is considered to be playable by most people (which it is) then you can 'afford' to go as low as that map when making yours, whatever your specific FPS may be. I won't go into the technical details of how to use Portals, zone portals, antiportals and semi-solid brushes in this document, as that's not what I'm aiming to cover here, but there is plenty of help out there on how to improve FPS performance. Some broad tips are: Use static meshes instead of BSP brushes wherever possible, make complex BSP brushes like stairs, if you absolutely have to use them, semi-solid, and remember that the larger the area you can see on screen - the sparser it will have to be by necessity to maintain FPS.
I know this has been a long read, and I'm sure 50% of you already knew 90% of it, but I hope everyone who's read this has understood it and I hope it's provoked a new thought for each of you. I truly believe that if we all took time to consider the three aspects I've covered here when embarking on the life-draining exercise that is designing a map - they will all rise in quality and we stand a much better chance of getting them approved.
Bottom line is that I'm only trying to offer what tips I can. I hope it helps someone.
Hooah soldiers, and keep up the good work!