Mapping Theory & Combat Design

[magicsofa]

Fair points y'all but keep in mind, the question is about design rather than process. So what is it about that process that leads to a better result? Or more accurately, why is the result better if you do process A vs B?

[Treehouseminis]

18 hours ago, magicsofa said:

Fair points y'all but keep in mind, the question is about design rather than process. So what is it about that process that leads to a better result? Or more accurately, why is the result better if you do process A vs B?

I can't for certain say a fully planned layout is always better than area to area spontaneous building  because I've never done it, and idk if my favorite wads or levels used which building style or a completely different one.  

 

But I feel like a fully planned, interconnected level can be better because everything is thought out more.  It also depends on the goal of the level.  I've made big levels before but I have yet to capture the liminal feeling of scaling a big singular structure of the likes of Sunder or Abandon.  I would like to do something like that one day and I feel a fully planned layout would help a lot, like no matter which part of the level you are in there's a big structure in the backround to which you are getting closer too etc. I think it would take a lot more architectural planning lol

[TheHambourgeois]

18 hours ago, magicsofa said:

Fair points y'all but keep in mind, the question is about design rather than process. So what is it about that process that leads to a better result? Or more accurately, why is the result better if you do process A vs B?

I think it really just comes down to if you plan you are more consciously thinking about the design. You are thinking about the layout in a holistic way, you are thinking about the encounter designs and how they relate to each other and not as isolated units, etc.

[zokum]

I don't think the process matters much. As long as there is a vague overall idea about what you want in the map, that should be good enough. My plans tend to morph and change based on how the map plays so far and how I can manage it to look.

What I really dislike are open maps that aren't really open at all, they force you to do things in a very specific order and unless you do it this way, the map is borderline impossible. It feels like you're being punished for exploring, and that you're basically running into traps you cannot predict. Map10 in Doom 2 is an interesting example, there are tons of ways you can choose where to go and which order to do things in. There are probably some rooms that require more ammo than you get when clearing them, but overall you are fairly free to do things in any order. Map15 uses a simple approach, give you a decent starting ammo and present you with tons of small combat scenarios. Only a very limited amount of them are mandatory to finish the map, but in a blind run you would most likely finish a large percentage of them before you reach the exit.

More linear maps are also fine, especially if you are well guided and that any backtracking is clearly guided. I dislike finding a fork, choosing one section, and after a log of time ending up in some sort of dead end where I should have chosen the other fork instead in order to progress in this one.

The best combat situations for me are the ones where you can adjust the risk to decrease the time spent. It allows you to utilize your level of skill to do it as fast as you can manage. Multiple possible approaches will allow for great max kill speed runs, but also make it doable for less skilled players. With enough such combats of varying difficulty, a player can adjust the difficulty of the approach in a quite fine tuned way as the skill increases. Hard to get right though.

Edited October 13, 2023 by zokum

[LadyMistDragon]

Generally, I just follow the process of creating a space intended for a certain amount of monsters I've made a mental note of, texture things so I don't get bored to death instantly, then later, add in more detailing and some monsters. Speaking of monster placement, I always think about how to make them count. Not like it's Sunlust or anything but I at least want the player to think for a second or two instead of just blindly firing at z group of monsters.

 

On detailing, it usually occurs to me to put extra detailing like Doomcute in other areas after making something approaching an indoor space because I don't really have a detailed design document, just a general idea of the combat flow I want and if I have a particular concept such as Doomguy waking up, then I might spend a lot of time on the house.

 

PS. That's obviously not meant to be authortative.

Edited October 14, 2023 by LadyMistDragon

[LoatharMDPhD]

 

Edited October 14, 2023 by LoatharMDPhD

[LoatharMDPhD]

On 10/11/2023 at 12:13 PM, magicsofa said:

Fair points y'all but keep in mind, the question is about design rather than process. So what is it about that process that leads to a better result? Or more accurately, why is the result better if you do process A vs B?

...But the two are linked...

[magicsofa]

2 hours ago, LoatharMDPhD said:

...But the two are linked...

 

Yes. Also, ontologies are deeply rooted in things that logically must be true :P

[RonnieJamesDiner]

After 55 some-odd maps, I still actually have no idea how to describe my design process. It's like a weird blur of getting one idea that morphs into another, and then another, and then at some point I'll get a sudden, random sense of the full layout, and eventually push myself across the finish line because I feel like I've come too far to give up. I bounce back and forth like a pinball between layout work, detailing, combat, aligning textures, picking skies, finding midis, and endlessly playtesting, probably based on nothing more than losing interest in one thing, so I switch to another. 

 

I really wish I had something more articulate and enlightening to say, especially considering how much I seem to wax poetically in my WAD reviews, but it's always been this strange enigma. I'm guessing that's why it's so easy to find pages of tutorials on the technical end of mapping, and much harder to find detailed explanations on the theoretical side. 

Edited October 15, 2023 by RonnieJamesDiner

[zokum]

I think many mappers have different design approaches depending on the type of map they are making. They might have one approach for a city map ala Downtown and another one for a boss map or e1 style map. I, and maybe the iwad maps and many others have a tendency to start the map from the bottom or left side, and have it flow upwards/left to the exit. I rarely put the start area on the right side and then the exit on the bottom.

If one is working on a very large vanilla map, some sort of box shape is often what you need to plan out ahead. Then it pays off to have a rough sense of the entire layout to avoid venetian blinds due to blockmap problems.

[baja blast rd.]

A lot of my biggest improvements have been process-related.

 

I've shifted from trying to work in a set, systematic order to something really chaotic. Like, these days, I'll routinely do stuff like design features that I don't even know exactly what they'll be used for until weeks later.  Or...half-design features that I half-know what, switch from one thing to another in the middle of it. I'll avoid context switches heavily in weird ways, like constructing a lift and then not feeling like wiring up the functioning part because I find switching constantly between static features and wiring up their functioning tedious and worth avoiding even if it takes 10 seconds to do, so I simply IDCLIP over that lift whenever I test the map. When I'm drawing a layout, I'll have a bunch of unclosed sectors and random line segments not connected to anything, for a while. 

As a counterpoint to some ideas being discussed, despite how chaotic that sounds, that isn't absence of process, though. To me it's very much process. I think the idea that you're not using a defined process unless it happens in the same external way every time, and can be clearly verbalized with a clear sequence of mapping concepts like layout, encounter design, detailing, lighting (or any other way you might split up aspects of maps, like spatially), makes process out to be more narrow than it truly is. If process is sort of a tacit decision tree, then something like "work on whatever inspires me to work on it then and there, and leave lots of open loops that inspire me later in unexpected ways, all while trying to make the mapping process fun and motivating rather than feeling like it's work" (or more specifically, the Vibes-hunting that can be described roughly that way) is still a process, even if it isn't defined with mapping-specific concepts like "layout, then combat, then texturing and architecture, then small-scale design," even if it ends up looking very chaotic when translated into mapping, and even if it's done by feel rather than followed consciously like a written-down flowchart might be. 

 

The more chaotic and less traditionally expressible my process has gotten, the better I've gotten at process. One of my biggest process improvements in the past two years was consciously shifting a lot more towards something RJD just described, which is: working on what interests me most at the time, whatever it is, even if it involves completely abandoning what I was working on before. And I very consciously did that. So in RJD's case he claims he can't verbalize it. But I can verbalize my chaos and I think it's very processy. :P  

 

It is useful to think of that as process instead of something ineffable because then you can get consciously better at it and share ideas.

[SkeletronMK666]

i like making a HUGE fight with demons on one area, and having to revisit the area only to have archviles come in and revive all the monsters you've killed prior. very good encounter trick. archviles are fun.

[zokum]

10 hours ago, SkeletronMK666 said:

i like making a HUGE fight with demons on one area, and having to revisit the area only to have archviles come in and revive all the monsters you've killed prior. very good encounter trick. archviles are fun.

It's an interesting trick, but also very hard to balance around. If you return quickly, you will not need that much ammunition. If you spend a bit more time, you will have to spend a lot more ammo. I suppose you could offset that by having a few shotgun guys in the mix, since they tend to generate 4 times the ammo you need to kill them. You could always design it around the idea that the viles will ressurect everything, 95% of the time, and use that to balance.

I do like this trick myself, and it is one of those changes where speed running makes an encounter significantly faster if being fast greatly reduces the amount of monsters ressurected.

Edited October 17, 2023 by zokum

[SiMpLeToNiUm]

On 10/15/2023 at 6:17 PM, RonnieJamesDiner said:

After 55 some-odd maps, I still actually have no idea how to describe my design process. It's like a weird blur of getting one idea that morphs into another, and then another, and then at some point I'll get a sudden, random sense of the full layout, and eventually push myself across the finish line because I feel like I've come too far to give up. I bounce back and forth like a pinball between layout work, detailing, combat, aligning textures, picking skies, finding midis, and endlessly playtesting, probably based on nothing more than losing interest in one thing, so I switch to another. 

 

I really wish I had something more articulate and enlightening to say, especially considering how much I seem to wax poetically in my WAD reviews, but it's always been this strange enigma. I'm guessing that's why it's so easy to find pages of tutorials on the technical end of mapping, and much harder to find detailed explanations on the theoretical side. 

Flow. State.

[zokum]

The best approach is probably the one that fits your mind and way of thinking. Good designers know about all the tool and how they can be used and selectively use the ones that can do a quicker or better job. When it comes to Doom mapping, it's mostly about doing things faster and to reduce user errors.

One interesting approach I have seen in a large Doom project was to have one person do the map, place out a few key monsters for specific situations, but otherwise let another designer do the monster placement. The monster placement guy might be placing monsters in a large amount of the maps, so that the difficulty curve can be fairly smooth. It's one thing to make the UV balanced, but you also want easier skills to be balanced and maybe you want nightmare to be possible as well. Monster placement can drastically change how viable a map is on nm, but not matter that much on uv max or uv speed.

Properly implementing all the skill levels is probably an aspect many mappers ignore. One common trick is to balance skill 3 about the same level of difficulty as doom 2 on skill 4. Skill 1,2 as slightly easier than doom 2 skill 3, but not quite as easy as the original skill 1 and 2. Then you can pile on the monsters you like and if people just want a Doom 2-like experience, they can play on skill 3. Coop is a whole different ball. Some balance the map around the classic up to 4 players, extra monsters and ammo style. Other balance around respawning players and respawning ammo. That can make the maps seem brutally unfair when skilled players play as they keep running out of ammo and have to backtrack, or in some cases suicide in order to get ammo from the earlier parts.

Anders J introduced me to the concept of adding teleports and other helpful items in the start area to avoid having to do the 2 minute weapon run when you respawn. As you reach key points where you pick up a weapon, you also make the same weapon available in the start area, but only in multiplayer. That along with keyed doors that stay open make maps a lot more fun in coop. You could also include the keys in the start area. A few simple tweaks make the maps a lot more enjoyable in coop, especially the first time when you are likely to die more if you have extra boss monsters etc.

I think people are more likely to enjoy the maps when you remove the boring parts :). I dislike when you can return an area that you have opened up earlier, but then have to backtrack because you need the key once more. Ideally keys should only be put on doors where you can reach all other doors of the same key color without keys. Avoid putting two keyed door of the same color after each other. It's perfectly fine if it is a shortcut, but if it is mandatory, it is a a design weakness. You can after all have a key open up all the keyed doors in the area with the same switch. Or you can do something ala doom 2 map17 where you need a key to open up a door to lower bars. You could remove all the bars to that area, but provide multiple locations to do it from. It doesn't matter if the bars lower to 8 units above lowest neigh. floor multiple times.