Form or function - what is your approach?

    So I am on a somewhat of a mapping roll right now, I find absolute joy putting down lines and small details, but what I find happens is I make the entire map and then I'm faced with the problem of populating the map with baddies. How do you make your map? Do you make a skeleton "room" that primarily serves to house some interesting encounter, and then flesh it out or is your process similar to mine?

Function. FPS maps do not obey the laws of the real world and truly realistic/detailed locations generally don't make for good combat arenas. Ideally your map should have some kind of unifying theme, but the primary focus should be on making areas that flow well and facilitate combat.

I almost always do gameplay first with raw level shapes in startan, the blockout approach as I believe it's called in the biz. Then I take what works and do the visuals.

The problem with doing visuals first IMO is that it has the killer combo of firstly taking a lot of time and also being great for procrastinating when you are stuck on the stuff that's actually important. On top of that, I believe almost everyone experiences some natural resistance to throwing out work they have put a lot of time into and that at least in some aspects pleases them. With blockouts, if it doesn't flow well, or if the combat setup I had in mind is in reality no fun at all -- which happens often -- I can throw it away and pursue the next idea with minimal psychological friction.

Always exceptions of course. For maps which are setpiece based or which start as a visual idea I might do aesthetics first in one area to set the mood and create a concrete example of the style so I don't forget it. But then... back to blockout.  You can do fun maps both ways for sure, but I think designing your architecture around gameplay is an easier route to success than looking at a collection of rooms and working to gameplay backwards from there. 

Functionality first or at least have all steps and placements planned if you go off the road... It's easier to adjust gameplay and layout when nothing much is done detailing wise. Sometimes it hurts to scrap detailed rooms, readjust them, because they can't offer interesting gameplay scenarios, talking from experience. To summarise this - gameplay first, detailing later. Place things together with your lines, it should give you insight how it will play and rooms won't feel empty when you finish with them. 

Functionality trumps form. If something detracts from the gameplay it should be altered or removed. Example: I don't like putting superfluous indents and borders on the floor because it messes with your view height during combat. Another one is putting details on the wall you can get snagged on. This is probably the wrong answer for this question but it's what I think of with this kind of thing.

So to answer the question, ideally you should work out the whole level before you make it pretty. I'm guilty of jumping the gun on detailing though. I think everyone has that problem at some point. 

5 hours ago, Super Mighty G said:

So to answer the question, ideally you should work out the whole level before you make it pretty. I'm guilty of jumping the gun on detailing though. I think everyone has that problem at some point. 

    I hear ye, using the old sketchbook dramatically sped up my mapping (before I'd make a room and then detail it, and then wonder 'what's next?'). Now I tend to draw at least 1/2 of the map on paper, and then draw the lines. I can even indicate where the monsters should go!

    I do generally agree that map should be fun to play, but from what I've seen not all share this function over form mindset, ex. Russian mappers really go crazy with lines and looks. I guess I am closer to them. DooM is an artform, and art is expression of the artist (did that sound pretentious?).

Function, why would someone want to play something that isn't fun or enjoyable? Good visuals can make a wad MORE enjoyable, but they can't, on their own, make a wad enjoyable. I want salt with my fries, not fries with my salt.

The fries come first, then I put the salt on them. Priorities.

This is a complex question with no 'right' answer so I'm going to speak from personal experience and I'm going to provide an alternative view point to everything already said :D 

I map form first. I have a visual presentation or concept or detail in mind that I want to bring to life and my map grows from that. I love taking real world architecture, other game designs or my own insane concepts into Doom and bringing them to life. The aesthetic concept has primacy. But, while I'm doing this there is always a subconscious thought of "what encounter/progression/gameplay loop will be fun here". So there is always an internal prodding thought of why are you doing this.

Now, to pull this off you need to have a degree of faith/arrogance that what you are making will result in good gameplay at the end of the day which is potentially risky and is dependent on mapping experience. You need to be secure in your style and know what concepts lend themselves to good gameplay. Office building based layout? Shit. Layout based on a 6th century church? Exquisite.

I do 80% of the map before placing any combat elements but I know what combat elements I generally want to place. This is exactly the process I followed for Bastion of Chaos. No enemies were placed until I had finished building the environment. And there were areas that I had no idea what kind of fight I wanted until I started placing enemies (the bfg fight and the fight before the end cathedral is a good example of that).

Now sometimes I don't have a cohesive visual concept in mind for the next area I want to make and in this case I usually think "what fight concept haven't I made yet" and go from there. Caco swarm? Archviles with pillars? Cyberdemon cluster in a crowded hall? That's the starting point. But it will then point me towards a visual concept that I will build the area off. Archvile fight with pillars for cover means lots of pillars, I'm gonna make a sexy pillar first which becomes 2 pillars which has an ajoining beam between which is supporting a vast vaulted ceiling which has glowing goo below to underlight the pillars which looks great and oh that's a pain sector and the fight now involves trying to use the pillars for cover while avoiding the environmental hazard of the pain sector and sometimes you might need to take a pain tic or 2 to avoid an archvile blast and what's your objective ..... and now we have an engaging fight in a visually arresting area.

Long story short, do what feels best to you as a mapper. There is no one best way. You need to find what motivates you to map and channel that into maps people want to play. You can map form first, you just need to know what forms will also function well at the end of the day (hint not city maps)

I make ideas, so this question doesn't feel like the right one, to me. Some of those ideas are structures or visuals, some of them are gameplay ideas or scenarios. A lot of them are answers to mapping challenges, themes or restrictions.

when mapping I usually use the same process as Bridgeburner56, since it is a matter of getting some level architecture realized, not a matter of actually making a doom level

other times its more of an ideer that something along the lines of this would be cool

and rarely its about the flow of the level

I place monsters and decorations etc after the level geometry is completed, I have done this 99% of the time where I have made levels.. both in the 90s and now

48 minutes ago, Phobus said:

I make ideas, so this question doesn't feel like the right one, to me. Some of those ideas are structures or visuals, some of them are gameplay ideas or scenarios. A lot of them are answers to mapping challenges, themes or restrictions.

Pretty much what Phobus said. I suppose it depends on the map and whether I have an idea of what I want or I'm just going in blind to see what happens. I sometimes build all the geometry before placing a single monster, or I'll build an area and populate it right away.

2 hours ago, Bridgeburner56 said:

*Stuff*

    That's kind of what I was going for, this kind of mapping procedure. I do think about "what I am going to put here" while mapping. For example, I made a series of spooky dark rocky tunnels in ground (think E1M5 of Dusk, "Mine control"). What do I put here? Spectre ambushes, enemies popping behind and around you unexpectedly, cave-ins revealing a Baron that is terrifying in close space!

    Generally, I do think of "what enemies are strong in this space" when building the map.

EDIT: City maps not working @Bridgeburner56? What about Hellbound?

I tend to fill rooms as they are created. Generally I will draw a handful of rooms in a "macro" way, with the main consideration being the player's movement through the area and the opportunities for fun combat. I focus on creating interesting angles and ways for the map to be revealed as you explore, while avoiding "arena battles" where you open a door to find a set-piece room full of perfectly placed monsters. Instead, I try to envision the entire level as the arena, with each encounter building or relieving the pressure. I want the player to have lots of options, such as being able to approach areas from more than one location, retreat to find more supplies or a better tactical position, or run past an encounter hoping for those supplies/position to be on the other side. Hopefully, if I've done a good job, all of those options will introduce at least a small risk of waking up even more baddies.

For me, introducing those considerations in the earliest geometry placement helps a lot. Like others have said, it's harder to modify something that is heavy with detail work. But if a big empty room needs more entrances, ledges, etc, it takes all of five seconds. Plus, I find that detailing is a bit easier for me later on when there's plenty of functional aspects of the structure. Each entrance, ledge, stairway, and so on gives you a nice spot for some border textures/sectors, and a reason to do some cool lighting. On the other hand, looking at an emtpy square and trying to make it look pretty can grind my enthusiasm to a halt. Anyway, I do tend to fill these rooms with actors and decoration pretty early on. As long as the room's connections to other places seem sufficient, I like to see it in action. So I'll often make the start of a map + a couple rooms, then switch to placing monsters and decorations for a bit, then add more macro areas once the initial cluster is working to my liking.

I find drawing out the entire map first to be almost impossible. I like to test each segment, so that I can feel out where the map "wants" to go. I think it helps me create a level that feels organic and natural. I try not to fret too much about perfect monster placement, which becomes increasingly difficult the more non-linear your map is. You kind of have to accept that sometimes the monsters will be in a bad position, which I think can actually make the map more fun (again, giving the player options, and a variety of encounters, some that overwhelm the player, some that make the player feel like they caught the demons unaware)

I admire how much skill and thought some fellow mappers put into the craft. I always make sure I end up with something that looks half decent and I’ll always play test until the difficultly level strikes a good balance between CBT and a casual romp, but ultimately diving into mapping theory in any meaningful way is just too highfalutin for me. I of course have a core list of “do’s” and “don’ts” in mapping that I always follow automatically and have for years now, but it’s not exactly high level stuff - anyone could remember these “mapping commandments” in a matter of minutes (and I actually recently made a short video on “what not to do if you’re new to Doom mapping”, and I fit it into just 6 minutes!)

Anyway, aside from all my prose and rambling, my process could be boiled down to: Slap some lines down, populate, playtest a few times, then release whatever comes out upon the world after getting feedback from a handful of players. Preferably try to add some music, textures and sprites that make it feel a bit more memorable than “wad with unchanged assets #486,368”.

I’m an impatient mapper, too.. It’s probably why I’ve never released any grand-slam-tier wads, but it’s also why I have 250something maps on my Doom resume, including a large handful of decent length episodes and even a couple megawads. In my dreams I have the dedication and motivation to make a new map every day of the damn week..! The reality is unfortunately pretty far from this.

I sympathise with the approach of further refining things until they’re marvellous. In this era of truly endless content, it only makes sense to make something that really does stand out from the rest. Personally though, I’m a simple Doomer. I play simple wads, and that’s the flavour that truly resonates with me, so that’s largely reflected in the stuff I put out.

I fear this answer was long but offered nothing to mappers seeking to improve their craft.. Hopefully it at least makes some sense. Not much in the way of “form”, but whatever comes out, I’ll make a damn solid effort to make it function as best as it can.

Function, always.  I map one room at a time, making enough architecture for the section to be largely complete, and I populate it with monsters and items and test how it plays before I make it final.

It's far better to discover problems with the gameplay when you're still building the level than it is to discover it after you've finished and detailed the level. I cannot imagine intricately building and detailing an entire level without putting in any weapons, armor, health, or monsters.

As @Bridgeburner56 says above, it is a complex question. I'm a coder by trade, so I'd like to say 'function'. But that is not entirely true...

I'm still kind of finding my feet when it comes to the 'right' approach for me. My levels so far have been triggered by an idea, possibly inspired by a comment somewhere. Certainly, function is important, but that is actually form that trumps I think. I always describe code I am happy with as 'elegant', and I think the same is true for mapping. I can screw around with ACS or line actions etc. to achieve a function - but that is just logic. It is of course possible to make a hugely functional map that is just ugly as sin - thus making it unplayable.

My maps so far have certainly not been planned in any way, I get an inspiration and just start, and do what feels right. Sure, I'll make notes and test stuff as I go, but the initial inspiration is the idea, rather than the technical solution to that idea.

Also, as I don't plan out the map, further ideas will get inspired as I go, and I'll figure out the technical solution as needed (very much like my actual coding I suppose).

So, form is key for me.

Prioritizing function over form doesn't mean you don't have an idea in mind, and it does not mean deliberately making the map look bad. You can typically make a level which plays well look better with a little additional work. If you take a map which looks fantastic, but plays like shit, making it play better is likely going to require much more work.

Ugly maps that function properly cannot be "unplayable" by definition.

7 hours ago, Pechudin said:

EDIT: City maps not working @Bridgeburner56? What about Hellbound?

LOL

The best city maps are ones that look like cities through textures, skybox design etc while having a layout nothing like a city. Valiant is a good example of fun city maps

8 hours ago, magicsofa said:

I tend to fill rooms as they are created. Generally I will draw a handful of rooms in a "macro" way, with the main consideration being the player's movement through the area and the opportunities for fun combat. I focus on creating interesting angles and ways for the map to be revealed as you explore, while avoiding "arena battles" where you open a door to find a set-piece room full of perfectly placed monsters. Instead, I try to envision the entire level as the arena, with each encounter building or relieving the pressure. I want the player to have lots of options, such as being able to approach areas from more than one location, retreat to find more supplies or a better tactical position, or run past an encounter hoping for those supplies/position to be on the other side. Hopefully, if I've done a good job, all of those options will introduce at least a small risk of waking up even more baddies.

For me, introducing those considerations in the earliest geometry placement helps a lot. Like others have said, it's harder to modify something that is heavy with detail work. But if a big empty room needs more entrances, ledges, etc, it takes all of five seconds. Plus, I find that detailing is a bit easier for me later on when there's plenty of functional aspects of the structure. Each entrance, ledge, stairway, and so on gives you a nice spot for some border textures/sectors, and a reason to do some cool lighting. On the other hand, looking at an emtpy square and trying to make it look pretty can grind my enthusiasm to a halt. Anyway, I do tend to fill these rooms with actors and decoration pretty early on. As long as the room's connections to other places seem sufficient, I like to see it in action. So I'll often make the start of a map + a couple rooms, then switch to placing monsters and decorations for a bit, then add more macro areas once the initial cluster is working to my liking.

I find drawing out the entire map first to be almost impossible. I like to test each segment, so that I can feel out where the map "wants" to go. I think it helps me create a level that feels organic and natural. I try not to fret too much about perfect monster placement, which becomes increasingly difficult the more non-linear your map is. You kind of have to accept that sometimes the monsters will be in a bad position, which I think can actually make the map more fun (again, giving the player options, and a variety of encounters, some that overwhelm the player, some that make the player feel like they caught the demons unaware)

I like that approach. Usually the geometry comes to me first and then I relish in added details (most of the time)... but maybe I could combine it with your way by doind an area, its detailing and then place monsters and decos and test it out to get a feel for the "flow" and where the level might "wish" to go

30 minutes ago, Bridgeburner56 said:

LOL

The best city maps are ones that look like cities through textures, skybox design etc while having a layout nothing like a city. Valiant is a good example of fun city maps

I don't know, I guess it will depend on the particular map. Personally I hate "pretend" city maps that aren't REALLY city maps.

But then again, I am also a kind of a purist in my own way... for example...

I prefer to do either TRULY vanilla maps (ie. ones that could work on an old dos machine... NOT BOOM, chokolate or any of the other weird vanilla-but-not-really-vanilla doom ports out there)

or truly modern maps... And for me, that is GZDoom with as many 3D floors, 3D monsters and various other advanced features as possible... This is also why I have started building a very large repo of 3D monsters for GZDoom... and after that it will be 3D props and anything else that GZDoom will allow me to make as a 3D object... and... POTATO rigs be damned... potato rigs are better off playing BOOM or vanilla maps anyway IMO

I am NO LONGER going to BUTCHER perfectly nice GZDoom maps, just so a few potato rigs can play the 'map' on 5 fps... The map will afterall be nothing but a shadow of what the map was intended to be afterwards anyway

I have taken part in BOOM projects and it has been fun and all but generally I do not care for those halfway modern and halfway potato engines like BOOM and others

Function. I'm not a visual guy at all in the first place and I've always been gameplay oriented and while I appreciate good looking maps, I just don't have the patience or visual creativity to make anything beyond passable.

1 hour ago, CBM said:

I like that approach. Usually the geometry comes to me first and then I relish in added details (most of the time)... but maybe I could combine it with your way by doind an area, its detailing and then place monsters and decos and test it out to get a feel for the "flow" and where the level might "wish" to go

Just to be clear, when I said 'decorations' I meant sector/line details as well as thing decorations. Basically anything that's there just for the appearance. 

Quote

I prefer to do either TRULY vanilla maps (ie. ones that could work on an old dos machine... NOT BOOM, chokolate or any of the other weird vanilla-but-not-really-vanilla doom ports out there)

The point of Chocolate Doom is to reproduce the original executable's behavior, bugs and limitations included. Pretty much anything that works in choco, should work in doom.exe. It's just more of a pain in the butt to run mods since you might have to build a modified exe and custom merged IWAD. I still do it just so I can say "tested with doom.exe" in the text file, but it's almost a formality at this point. If anyone's using doom.exe, it's probably because they purchased it bundled with dosbox anyway. Choco would be better described as not-vanilla-but-actually-it-is-vanilla

 It is precisely the level of intricacy/connectivity of the layout that dictates the overall pacing and flow of the map, what type of combat certain areas will have and what gives the main distinction from a map to another, keeping things fresh and interesting. Form IS function. Detailing is an afterthought once you have a 'functional form', usually to convey a theme and set the mood for the player. Watching videos like @HAK3180's can give a lot of insight on what makes DOOM combat fun and be aware of other perople's playstyles and there're tons of content about level design in general regarding layouts and how to keep them varied and interesting to explore and advance through.

 Since creative flow is something that varies from person to person, it is impossible to create a 'recipe' for map making that suits everyone. 3 things I find extremely usefull though is always have combat and progression in mind when designing layouts, run through the map without monsters a couple times to test the overall flow, and feedback, lots and lots of them.

5 hours ago, Iacobus said:

 It is precisely the level of intricacy/connectivity of the layout that dictates the overall pacing and flow of the map, what type of combat certain areas will have and what gives the main distinction from a map to another, keeping things fresh and interesting. Form IS function. Detailing is an afterthought once you have a 'functional form', usually to convey a theme and set the mood for the player. Watching videos like @HAK3180's can give a lot of insight on what makes DOOM combat fun and be aware of other perople's playstyles and there're tons of content about level design in general regarding layouts and how to keep them varied and interesting to explore and advance through.

 Since creative flow is something that varies from person to person, it is impossible to create a 'recipe' for map making that suits everyone. 3 things I find extremely usefull though is always have combat and progression in mind when designing layouts, run through the map without monsters a couple times to test the overall flow, and feedback, lots and lots of them.

I like your Cammy avatar. Are you in to Amigas by any chance?

Form or Function?

A balance. 

Bias towards one or the other can change from map to map, project to project due to time/creative constraints/liberty (as well as other factors). Sometimes I have a clear idea before entering the editor of what I want to accomplish visually or gameplay-wise for example:

- A scripted scene in UDMF format to have lights turn red and the doors to a hangar open with hordes of monsters behind them whilst the player waits for an elevator (cliché I know but please humour me for the sake of example).

- Or for gameplay I might have the pre-conceived notion that I want a cyberdemon fight with lowering/raising geometry shields.

Sometimes I have no clue what I want to accomplish and must make do on the fly.

I see the general consensus in the community is function triumphs form, and this has been stated in this exact thread - I disagree, for a map purely based on function with no form (or heavily biased) can be somewhat fun to play however, can be quite forgettable without substantial form to make it into a place, not just a level.

That said, sometimes people do want to make/play a level and not a place. Sometimes maps benefit from a bias towards form or function, a scenic cliffside looking towards vast castles clung like barnacles to the vertical stone may very well want the player to focus on the grand views rather than distracting them with combat (not to say you cannot have both), or a fast paced blow everything to smithereens type deal where you have little time to take in the sights therefore they aren't needed as much. It depends.

I appreciate it if you made it past my meandering and hope my view wasn't too convoluted.

I've always focused more on the looks of a map than the function aspect. I feel I couldn't come up with an actual fun combat map even if my life depended on it. I just found it fun to create an environment that I could run around in.

12 minutes ago, Chopkinsca said:

I've always focused more on the looks of a map than the function aspect. I feel I couldn't come up with an actual fun combat map even if my life depended on it. I just found it fun to create an environment that I could run around in.

    Hah, I sometimes feel, after creating a room, and asking myself: "Alright, what WAD has a fun fight in a similar looking (geometrically) room?", and then going from there.

I've never had to choose one over the other to be honest, my decorative progress just kind of goes hand in hand with my gameplay planning, since every bit of decoration i add makes me think "how does this affect gameplay? if it doesn't, is there any way to incorporate it into gameplay?" and similar thought processes, which i feel is a development process that helps the map just flow, both for the mapper and the player, though a few friends told me they can't quite follow that process themselves so i'm unsure how easy/hard that habit would be for someone else to build up in themselves while mapping.

Usually form then function, and then form again, with some more function LOL. I'll visualize an imaginary screen shot of the first room before I even start. I'll be bored at work or something or have a dream and get an idea. Then once I have that room made as close to my visualization as possible, i branch off and flow from that into a more detailed concept. I'll start drawing random shaped sectors anywhere from squares to octagons connected by doors or hallways. Once I have a basic skeleton of the map I can usually start making my ideas a reality. A great way give yourself a direction on where to go with the map is to fill those empty sectors with the keys and decide what connecting doors are going to be locked (at least that's what I do because I'm more of a linear mapper). When it comes to monsters and ammo, I will start by populating each room linearly from start to finish. The way I calculate monster to ammo ratio is by dividing the number of rounds each ammo item carries by the amount it takes to put each monster in the room down with a little extra ammo left over. So for example, 1 Imp and a Pinky equals like 8 shells when calculating for missing, RNG and surplus, leaving the player with most likely 3 rounds left for what ever comes next on top of what they collect for ammo before the next encounter.  (first time players of a map wont know what to expect though they will adapt.) . Throughout my time building the map, I stay VERY strict to Romero's rules for design. Look them up and they do wonders for the aesthetics of your maps. Mapping is a fine balance between fun and challenge. The best way is to mix both is to make exciting and logically surmountable battles while keeping the game play challenging and they player on their toes. After all is set and done and I have decided where I want to put things ect, I start looking for places to put details and such. Once it's done and I know the map is challenging but enjoyable at the same time while looks pleasing to the eye, it's time to share! In the end though I'd say function is the most important of the both because a pretty map doesn't mean a fun one.