Matching a MIDI with a Map – Visuals or Gameplay?

This is something that’s been starting to bother me, as I really don’t know if there’s a “correct” approach to this, or if it’s entirely subjective. But, I’d love to hear what some of the thoughts are.

• First of all, I’m curious how mappers approach this, and certainly veteran mappers. Do you consider it more important to match a song with the visual design of your map, or the gameplay? Do you try to find a song that manages to strike a balance between the two?

• If you're a MIDI composer, when asked (or commissioned) to write a song for a specific map, do you tend to approach it from one direction over the other?

• As a player, do you find it more jarring if the song doesn’t fit the visual style/mood of the map, or if it doesn’t fit the gameplay?

Essentially: Do you prefer the MIDI to be in service of the maps gameplay tempo, or in service of the maps general ambiance?

visuals all the way for me!! I haven't actually really ever considered matching music to gameplay, though maybe thats just because most of my maps have pretty similar gameplay

11 minutes ago, Tango said:

visuals all the way for me!! I haven't actually really ever considered matching music to gameplay, though maybe thats just because most of my maps have pretty similar gameplay

That's interesting, for whatever reason I've always been the complete opposite. I go nuts trying to match the tempo of the song with the "ideal" or "intended" tempo of the player throughout the map, if that makes any sense. I don't know how successful I've ever really been at this, but I've received a lot of feedback over time that my MIDI choice sometimes clashes with the mood or atmosphere of my maps, and I find myself wondering- "oh, is that what I was supposed to be aiming for?"

When I think about it, it does make sense to match the song with the visuals, to keep things from feeling dissonant. I guess what my question boils down to is... is the song supposed to be in service of the maps gameplay tempo, or in service of the maps general ambiance?

I appreciate the feedback.

Both approaches are equally valid and for a map set I'd suggest using both. For example with a short episode:

1. MAP01: Introduction map, you have to present the setting, go with "mood" music
2. MAP02: Picking up the pace, now that the player is "in" things are getting serious, go with "action" music
3. MAP03: As the player gets deeper in enemy territory, the weight of the task at hand becomes evident, "mood" music
4. MAP04: With the player getting close to the target, enemy opposition intensifies drastically, "action" music
5. MAP05: Nearly there! A long and grueling trek through heavy opposition is all that separates the player from the enemy boss, "mood" music
6. MAP06: Boss fight! Definitely go with action music there.

Okay so I wrote it in a way that alternates perfectly. Of course you can have several "action" or "mood" tracks in a row. And also, what corresponds to "action" and "mood" depends on the maps and their gameplay. For example, E1M1: Hangar has a rather calm gameplay (it was, after all, meant to be an intro map to players who may never have tried an FPS before) but with a kicking tune. It actually qualifies as mood music, and that mood is badass action! Inversely, E2M2: Containment Area has a quiet and eerie tune that is fits well the exploratory gameplay of that sandbox level.

That's a great way of looking at it, especially that episodic breakdown.

From your examples, I suppose I'm getting the feeling that it's a highly contextual decision, where the mapper simply has to ask themselves what role they want the music to play. Which, honestly, is an extra bit of depth that I hadn't actually considered! Alternating the use of music between mood and gameplay, depending on how best its served in each given map in a set.

For a single map release, this does seem a bit tougher. You're only going to use the music one way, and so if it doesn't work/fit, there aren't a series of proceeding maps to cover that up. I guess you just need to be confident in what role you want the music to play, and stick with it.    

Gameplay for me because I always adapt the pace to the music

I like to match theme (broadly...) And try to keep tempo roughly suited to whether the game play is chilled out and fun, slow and tense or fast-paced and heavy-going. I tend to draw from vgmusic.com, but I'm branching out more.

i feel like most of the people who play a map aren't gonna constantly advance to match the music's tempo anyway, they'll backtrack for medkits or get suspicious about certain walls or just plain not attack an encounter like you planned. i always think midis need a certain amount of atmospheric content cos fast metal tunes sound ridiculous when you're looking at the automap or waiting for a barrier to fall <3

I go with the theme of the map while taking consideration of the gameplay. Yes....it's a mix of both. 

If I have to say a percentage, I'd put it 60% visuals, 40% gameplay. I know it sounds weird but that's how I do it...

Constant 1 note guitar harmonics for every map

Often a mismatched song goes a long way, not gonna lie I got pumped up and played longer when SQUAD417.WAD started playing bad Van Halen and White Zombie tunes.

Gameplay. love me some @TerminusEst13 midi playlist mod when playing slaughtermap :D

8 hours ago, RonnieJamesDiner said:

If you're a MIDI composer, when asked (or commissioned) to write a song for a specific map, do you tend to approach it from one direction over the other?

Inclined to take notes from the visuals over the gameplay, though in a lot of cases there isn't a choice, there are cases where I've written music for maps with only screenshots available to me (none of the Elementalism composers have access to the maps, for example) or the the map just straight up doesn't exist yet and you only get a description of what the map will be like. On the other hand, the music for Eviternity MAP30 was matched somewhat to the timings of the opening stages of the boss fight assuming the player immediately runs forward at the start of the map and never opens the menu: enter that string riff as the boss is revealed, blow up to full orchestra when the wall lowers to reveal the other monsters.

I find you can get away with wildly different things more when it comes to gameplay as opposed to visuals. If I were to be writing for some macro-slaughter map, my instinct would be to go down the symphonic power metal route a la Knightrider of Doom, but the finale of Sunlust's God Machine demonstrates that subdued ambient tracks also have their place in the same scenarios.

Edit: Also temp tracks are a thing. Sometimes the mapper will already have a particular mood or style in mind. The tracks I did for Obsidian's Maskim Xul and dobu's The Given were both heavily influenced by the temp tracks they provided.

As long as it's not D_RUNNIN

10 hours ago, RonnieJamesDiner said:

First of all, I’m curious how mappers approach this, and certainly veteran mappers. Do you consider it more important to match a song with the visual design of your map, or the gameplay? Do you try to find a song that manages to strike a balance between the two?

I usually prefer it when a MIDI is at least not "disruptive", and I like it even better when the MIDI is "fitting" for the map in question.

From the "I occasionally make maps perspective", there are two main approaches I have. One of which happened during mayhem19 where I made a collab-map with rdwpa. Once the map was put together, and we had a sense of how it plays and looks in a finished state, I started digging through my MIDI collection to find something that both rd and I felt was working well with the map we made.

For my UDINO contribution I took a different approach. I had a MIDI I always wanted to put in a map, so what happened there was that I was listening to the MIDI quite often while mapping, which contributed to how I felt various areas and setpieces should be arranged, spaced apart, how many calm and deliberately slow paced moments I wanted in the map and so on.

Also, I wouldn't necessarily pick a MIDI that feels "electronic" to me, and put it in a map that doesn't have any tech-aspects, for example, so there's that too...

So in a sense I like to either pick a MIDI that fits a map I already have, or I make something that I think fits a MIDI I already have.

Even though I sometimes don't listen to a WAD's soundtrack while playing, I am not a fan of MIDIs that don't feel to me like they belong with a map. Like a slow paced map with a "headbanger MIDI" is something that I'd argue is disruptive. But, as with all things art, everything is subjective in the end.

EDIT: I also think one aspect of choosing a MIDI is the length of a map. Even if it's a large, high octane slaughter map, would I want to listen to the same 3 minute headbanger MIDI for an hour? I think that's a question worth asking oneself when making something larger, or something so difficult that players might need a considerable amount of time to get past the map. Basically it's sometimes nice to consider how a MIDI comes across after long periods of time. Barring some speedmaps I made, I never picked a MIDI for any of my maps that I couldn't listen to for a least half an hour on repeat.

My method is quite similar to @Nine Inch Heels, though in my case, I overwhelmingly prefer mapping to the music as opposed to finding music that fits. The latter involves wading through incredible numbers of tracks to find something you not only like but which you believe kinda-sorta fits your map. This is especially the case when I'm going through Amiga tracker music. In general, I prefer slow to mid-tempo music, especially music that imparts an ominous, menacing, creepy feel. Dark Ambient is a great example of that. Alas, most Amiga tracker music that I like was made in the '90s and is skull-pounding techno. This is one reason I decided to make Amiga Demo Party maps short -- 5 to 15-minute blasts of pure carnage -- because that was the most I could stand with that kind of music. ;D

Worth noting is that tracker music tends to loop correctly in GZDoom and Risen3D, and further plays nice alongside midi, mp3 and ogg tracks in terms of volume. This is not the case in PrBoom, where continuers will face a jarring shift in volume going from a midi map to a tracker map, and because PrBoom has unusual looping behavior, tracker music can simply stop at the end of the song or loop infinitely through one part of it. As a result, unless you can edit the tracks in a program like OpenMPT, you'll need to find someone who can.

By contrast to skull-pounding techno, slower-paced, more moody music helps inspire me to create atmospheric visuals and dark places that feel dangerous. So when a player is in a darkened room looking at a switch they know will set off a trap, I'm hoping the music makes them sweat a little as they wonder, "How bad can it be?"

Slower-paced music also helps me make a more "pacey" map. In reality, the gameplay in my maps tends to be all-out-war from beginning to end, regardless of the music, but there are subtle differences in how the setpieces are spaced apart, or how severe the traps are, depending on the music choice.

To answer another of your questions, which does the music impact more, visuals or gameplay? The answer is twofold -- slow, ominous tracks influence the visuals more, while faster, more exhilarating tracks influence the gameplay more.

At the risk of pimping my own shit, I'll offer 2 examples, both in TNT:Revilution. This is one of the rare cases where I was able to pick out music to map by, and IMO I chose wisely. Map04 was made to reflect Sea Bats by Viscra Maelstrom. It's a fast-paced, bouncy tune, so I made it heavy on action which is sometimes almost goofy in its relentless pace. It didn't impact the visuals much, because I was consciously seeking to emulate some BTSX E1 techniques for creating complex architecture while avoiding HOMs. By contrast, Map14 was created under the influence of a super-creepy tune by Jeremy Doyle. This inspired me to make darker areas and attempt to create a sense of mystery, and the big mystery is the huge secret area which a lot of people never find. :D Just the same, I went for atmosphere in room design, while the gameplay is just as nasty as in Map04, but in subtly different ways.

Thank you for all the feedback, this has given me a lot to consider (more than I realized).

9 hours ago, yakfak said:

i feel like most of the people who play a map aren't gonna constantly advance to match the music's tempo anyway, they'll backtrack for medkits or get suspicious about certain walls or just plain not attack an encounter like you planned. i always think midis need a certain amount of atmospheric content cos fast metal tunes sound ridiculous when you're looking at the automap or waiting for a barrier to fall <3

This is some truth, right here. And the biggest reason why I've pretty well completely avoided ever using crazy fast metal songs, haha. I could never reconcile sitting there staring at the automap lost, while insane blast beats and machinegun drumming are pounding in my ear. I think I could only use a song like this if the map was quite literally just one huge, single arena, and nothing more.

6 hours ago, Eris Falling said:

Inclined to take notes from the visuals over the gameplay, though in a lot of cases there isn't a choice, there are cases where I've written music for maps with only screenshots available to me (none of the Elementalism composers have access to the maps, for example) or the the map just straight up doesn't exist yet and you only get a description of what the map will be like. On the other hand, the music for Eviternity MAP30 was matched somewhat to the timings of the opening stages of the boss fight assuming the player immediately runs forward at the start of the map and never opens the menu: enter that string riff as the boss is revealed, blow up to full orchestra when the wall lowers to reveal the other monsters.

I never, ever considered the possibility of a song being written purely from visuals (with no option to even play the map). I think that alone is testament to the power of a MIDI matched to the visual ambiance of a map, over its gameplay tempo. That said, the way your titular song matched MAP30 and its boss fight in Eviternity was so effective and memorable, and has stayed with me. It was a good reminder of why I personally love to match the song with the gameplay, because when it clicks... it's really powerful. And I often find that to be the more compelling end-result. I'm extremely interested in hearing the Elementalism stuff.

6 hours ago, Nine Inch Heels said:

I also think one aspect of choosing a MIDI is the length of a map. Even if it's a large, high octane slaughter map, would I want to listen to the same 3 minute headbanger MIDI for an hour? I think that's a question worth asking oneself when making something larger, or something so difficult that players might need a considerable amount of time to get past the map. Basically it's sometimes nice to consider how a MIDI comes across after long periods of time. Barring some speedmaps I made, I never picked a MIDI for any of my maps that I couldn't listen to for a least half an hour on repeat.

This is a really great point, and tends to affect my choices as well. Hell, sometimes I've ended up switching songs at the tail-end of the process simply because I've grown sick and tired of hearing the thing (through testing, etc.), which maybe isn't the best reason in retrospect.

I definitely share the feeling that, for example, a very "electronic" sounding MIDI feels just... dissonant and "off" in a map with zero techno-themed elements. So, on some basic level I do understand or empathize with the idea of matching a song visually to a map, even if I don't apply it much in practice. There are always going to be those extreme examples or cases where mismatched MIDIs are just so overwhelmingly apparent -- both from the visuals and gameplay. I appreciate your thoughts on the different approaches, and I actually think I want to start mapping while listening to a MIDI more often, to let it help shape all aspects of design rather than just being loosely paired with the gameplay tempo, or the atmosphere, or whatever. I think there's a strong argument to be made for the level of cohesiveness that can be gained from this.

4 hours ago, Steve D said:

Slower-paced music also helps me make a more "pacey" map. In reality, the gameplay in my maps tends to be all-out-war from beginning to end, regardless of the music, but there are subtle differences in how the setpieces are spaced apart, or how severe the traps are, depending on the music choice.

This is something I needed to hear! My maps are generally the same thing... all-out-war, and I never thought about letting the MIDI help me pace the combat effectively. Love this. 

I think you raised a great point as well, on the idea that slower, menacing/moody tracks can have a greater impact on the visuals, while faster/heavier tracks will invariably have a greater impact on the players' tempo. This sort of loops my thoughts back around to what Gez was discussing, and the importance of deciding what role you want the music to play in the map.

And to think, there is CRIMINAL SCUM out there who play with the music muted after all this effort that goes into the selection/composition process!!!

Lots of great answers here so I'll just say that, for me, I like the MIDI to fit the overall mood of the map, and to me the difficulty/monster count plays into the mood of a map just as much as the visuals do - the two are not separate things in a vacuum.

A huge dark church map with stained glass windows and empty halls may want a slow eerie track. The exact same geometry filled with ten thousand imps and barons might warrant something a little more action packed, that still doesn't "butt heads" with the visual theme by being too different. This MIDI wouldn't exactly fit the described map, no matter how many or how few enemies there were.

One thing that also irks me for wads with 2 or more maps is when the music is totally inconsistent. When I go from one map to the next and the midi changes from something Doomy to something decidedly not-Doomy, that really puts me off the experience. Bonus points if the volume is way quiet on one track then way loud on the next. The music doesn't have to be the exact same from map to map or something, but there needs to be some degree of thematic consistency.

2 hours ago, Doomkid said:

And to think, there is CRIMINAL SCUM out there who play with the music muted after all this effort that goes into the selection/composition process!!!

Haha... yeah, yeah... that crosses my mind. It's close to the same feeling of getting all sweaty trying to make the health and ammo absolutely perfect, knowing full and well that someone is bound to IDKFA or IDDQD their way through -- there's nothing I can do to stop it, and to each their own (Doom's supposed to be fun after all, so whatever that means to you). But! It's slightly disheartening all the same :P

I can appreciate the approach of treating difficulty/monster count and visuals as equal parts of the mood, and finding the right song accordingly. I think there'd be a greater sense of cohesion there, at the end of the day, if you can really nail down that balance. Definitely agree with your point on thematic consistency in a mapset, which again I think I've struggled with. I'm starting to realize that perhaps music is a bigger part of pre-planning than I may have thought.   

i just whatever song i like :)

I strongly feel both are a factor. If the map is fast-paced, a slow-paced track won't work; but it also won't work if the map is medieval themed, but the track is very techno. So when I look for an appropriate music track I try to find something that fits the overall personality of the map.