rethinking melodies and supporting lines

Please remember the terms of your membership agreement.

Moderators: valis, garyb

Post Reply
User avatar
kensuguro
Posts: 4432
Joined: Sun Jul 08, 2001 4:00 pm
Location: BPM 60 to somewhere around 150
Contact:

rethinking melodies and supporting lines

Post by kensuguro » Wed Aug 08, 2018 1:12 pm

Listening to old game music classics, I think it was a good opportunity to get back in touch with the basics of melody writing, or line writing in general. With limited voices, older game music were extremely clever in how they utilized each voice to depict aspects of the composition. Many times, the melody line would not only communicate the general mood and the melody of the piece, but also set up the general groove structure, while creating forward momentum to carry the piece from section to section. It's a lot of responsibilities to place on 1 line, but I feel with enough skill and thought, every line's efficiency can be maximized. Nothing worse than finding lines that 2 years down the line makes you go "eek, that wasn't necessary". With near unlimited polyphony and number of tracks, it's progressively become much too easy to layer in inconsequential fluff that only add aural stimulation and no material benefit to the composition. I think a bit of self regulation is a good exercise so the composition only retains the most effective, essential parts of the highest potency.

With hugging harmony ensembles, it's easy to make it work harmonically. The notes simply belong there or doesn't, and it's easy to tell. The default would be to hug along the main melody and move in complete unison, but at times the supporting harmony may each cut their own rhythm either so the part individually makes more sense, or becomes more exciting. That's where the mood of the main melody or the section needs to be maintained, or better yet, sold even more. Usually I think the parts should always work with the main mood, unless an ambiguous mood is the theme, but that's rarely the case. (or even if it is, expressing it with internal conflicts in the arrangement shouldn't be the central strategy) With timbre, I think the main melody should come out on top without the supporting harmonies obscuring it. This can get harder when there are multiple supporting lines, but can be fixed by panning or careful volume consideration, and limiting attention grabbing movement in the supporting lines when the main melody is doing something that shouldn't be missed.

With supporting lines that aren't tightly glued to the main line, the negotiation is much more complex. Timing wise, I think the base groove of the section needs to be maintained and complimented.. fortified, even. The supporting lines should not flip the weak/strong beat relationships within the groove, and definitely should not break the groove by not being consistent through the section. Masking avoidance still applies here, so the supporting lines should stand back or stay still while the main line does interesting things, and then fill in when there are gaps. This sets up a natural call & response structure, which I don't think needs to be created consciously, but as main melodies tend to oscillate between activity and less activity (from just good phrasing), it seems to play out that way. If the main melody is so busy and prolific that there are no natural gaps for the supporting lines to hop in, it probably means the main melody is too machine gun and noodling around. (and if played on a wind instrument, the performer will turn blue and keel over) Also, when the supporting lines have their own thing going, then you tend to need at least 2 parts to sell the contrast. The supporting lines need to sell the contrast enough against the main melody's natural dominance, or else the whole point of having the supporting lines do something different falls apart, and you end up with the supporting lines becoming obsolete, not accomplishing much for the composition.

The there is the entire spectrum between these two extremes. How much the supporting lines conform to the melody can shift throughout the entire piece, which I believe should be chosen and used consciously. There should be some logic to the developments, like not together, gradually coming together to a climax in unison, or maybe the other way around. But I think "how together" the lines are should be a cleanly communicated point, so it doesn't drift around or jump around which gets confusing and really, just messy. The contrast between the unison parts and disparate parts should be great enough to maintain the impact of when you switch modes. I personally like when everything lines up and starts moving in unison for the climax, but that impact would be lessened if the same thing was happening in a bunch of places in the piece. The choice of "how together" the parts are and more importantly, its sequence, needs to make sense for what you want to accomplish.

After the bass, melody, and supporting lines are in order, then other comping parts can be dealt with. You may already have comping parts to help with tracking the lines, but once all the lines are in, you may need to pull back a bit on the comping. Depending on how full the melody and harmonies sound, you may not even need any comping. Even if you do, hopefully there's enough substance in the melody's harmony that you won't need to blast too much to sell the chord progression. If you find yourself adding tons of comping parts to fill things out, it may be more effective to beef up the melody+support lines' harmony. Sometimes, it may also help just to swap out the lead sound with something more assertive, rather than adding a bunch of space filler parts. If done right, the melody, supporting harmony, and comping, and bass should all come together to produce a single image... Especially the supporting harmony should just blend in with the comping, making it difficult to tell them apart since they essentially should be operating as one mega unit.

With this level of reduction, I would say bass (1 voice), lead (1 voice), support lines (2-3 voices), comping (5-6 voices), and drums (polyphonic, but we'll just say it's 1 voice, or "track") at a total of 10-12 tracks + a few extra tracks for voice changes should be enough to do the job. For extravagant projects that need extra "sell", there may be a few more tracks for incidental SFX and one shots, but those aren't really a composition consideration, but are truly just for aural dazzle to draw attention.

User avatar
at0m
Posts: 4586
Joined: Sat Jun 30, 2001 4:00 pm
Location: Bubble Metropolis
Contact:

Re: rethinking melodies and supporting lines

Post by at0m » Sat Aug 11, 2018 2:47 am

For those who are younger than us, or simply as a reminder of these old days: The Perfect Amiga Game Music Compilation - Over 3 hours! (youtube)

edit: oh youtube for music, why oh why? from youtube-dl:
[download] Destination: The Perfect Amiga Game Music Compilation - Over 3 hours!-PvSfqBJi0ss.f137.mp4
[download] 100% of 456.31MiB
[download] Destination: The Perfect Amiga Game Music Compilation - Over 3 hours!-PvSfqBJi0ss.f251.webm
[download] 100% of 188.02MiB
more has been done with less

User avatar
kensuguro
Posts: 4432
Joined: Sun Jul 08, 2001 4:00 pm
Location: BPM 60 to somewhere around 150
Contact:

Re: rethinking melodies and supporting lines

Post by kensuguro » Mon Aug 13, 2018 9:40 am

ha, nice. Been a while since I've heard the characteristic tracker music portamento.

User avatar
at0m
Posts: 4586
Joined: Sat Jun 30, 2001 4:00 pm
Location: Bubble Metropolis
Contact:

Re: rethinking melodies and supporting lines

Post by at0m » Wed Sep 26, 2018 11:29 am

in other news, seems https://modarchive.org/ is back

*_*
more has been done with less

Post Reply