I record everything when I am working through ideas. I was in the studio last night starting Part IV of my epic song. I have Parts 1-3 done. The goal is to create a continuous song that nears an hour in length. Here is the process that I used to start building riffs and ideas.
1. I picked a topic/concept. Knowing that the concept is dark, I chose to work in a set of different tunings than I normally would use. I set my guitars up for C# Standard, Drop B, and Open B (open G in C# Standard). Then, I chose a key to work in - C#m.
2. Next, considering the dark concept, I chose a genre to work with. Being a rock guy, I decide on the doom/stoner genre. Blues riffs sound killer in C#.
3. Next I chose a tempo. This had to be a tempo that "fit" genre parameters, and allowed me the freedom to move up/down in feel based on adding or subtracting to the subdivisions in the beat. I chose 105BPM.
4. From there, I (sorry to be mildly insulting - this step is obvious, lol) create my Pro Tools project. I already have a template setup that gives me all my drums on their own tracks, the subgroups, aux buses, and mono audio tracks assigned to the proper interface inputs. This saves a TON of time. I highly recommend this step if you wanna be able to create on the fly.
5. At this point, I start looking for a groove to inspire me in EZDrummer. For this piece of music I decided to experiment with swing time.
6. Once I have a groove that "feels" cool, I start riffing. I always hit record at this point. I am looking for a riff that jumps out as something that could hook people in.
7. Once I have a riff that meets that requirement, I activate a second track and play variations on the riff to see if I can "improve" on it. Sometimes this is adding embellishments, other times it is only tightening up the playing. Once the riff is "improved" I record the second track.
8. At this point I go back and record over the first rack with a tighter take of the riff. I keep those riffs doubled.
9. Here I go back into EZDrummer and add another 4-8 bars into the song. Basically, this is a repeat of step 6.
10. At this point the task is to find a riff or progression that can build off the opening riff(s) without changing key or being too boring (i.e., I-IV-V) and predictable. Those "predictable" progressions work great for a chorus or hook. Save them! Unless the music calls for it...
11. From here I repeat steps 6-10 until the song structure is completed. At this point my project looks messy because every section is on different tracks, so I consolidate the riffs into one continuous track. I might have 12-15 individual riffs doubled - so 24 o 30 tracks that get bounced to 2. Much easier to look at and manage!!
12. Next step is laying down the bass guitars. To get going on this, I loop sections and play. I try NOT to always follow the guitar riff - but sometimes that's what is easiest. If I DO follow the guitar, then I will go back and try to work a riff variation on guitar to go on top of what's already there.
13. Here are some "extras" that I do in certain cases. If I want to modulate keys within the song, I go back to my key signature and look at available scale degrees/chord options. I come off the riff/progression and try different chords/degrees until I find something I like. Most of the time it is a standard half-step or whole-step move.
However, with the piece I am currently working on my movement between Parts is based on modal movements. I am looking at the various modes and the tonalities they impart of the music I am creating for the concept. I know each will work since they are scale variations off C# minor.
And to give an example of what I am talking about, here is Parts 1-3 of my epic. Beware - the sub sonics (sub bass) on this mix are awful. If you run your monitor system with a lot of bass or have a sub, PLEASE turn it down. I have a sub on order to help me get it under control when I do the final mix. This is a test run, lol....