Finding a musical workflow in SC that suits you is a tough task, it took me years before I made the first track that I liked. For those interested in my workflow:
I do all mixing in SC, and for me the mixing process is just repeatedly typing in dB values in SynthDefs until it sounds right. I mix as I go while designing sounds. When finishing a track up I will try to dedicate a session or two just for setting levels. Mixing entirely with text is probably insane, but I’ve gotten used to it.
I don’t use any GUIs, meters, or VSTs. I have many quibbles with the DSP in the built-in UGens, but as a musician I have learned not to care that much. I prefer to just make the best of what I have and not fuss too much over what DSP I’m missing out on.
I generally avoid compression or limiting as corrective tools for mixing. Certainly not on the master bus. As all my sounds are synthesized, dynamics issues are easy enough to fix with levels.
I record the master out of SC; I don’t export stems. I didn’t do any premastering for my first album, just did a quick inspection in Audacity to ensure it wasn’t clipping.
This is a stubbornly ascetic workflow, sort of like vegan cooking, and obviously it’s not for everyone. But, hopefully my music is sufficient proof that SC out of the box is a complete production environment.
SC users (including myself) often spend way too much time fixating on their toolchains and not enough on how the music sounds. My advice is to not fret about the “efficiency” of your workflow and focus on producing and iterating on sounds. My workflow is awkward, inefficient, and idiosyncratic, but I make music with it and that’s all I care about.
I’d push back on that slightly and say it depends on your goals and workflow. I think you’re right that doing mixing and mastering of a professional nature in SC might not be as conducive to quick, effective work as in a DAW. But I think it’s also possible that doing your mixing (and mastering? but I know less about this) in SC might lead you to make certain creative decisions that you wouldn’t have made otherwise, and that’s cool, as long as it doesn’t cost you a well-paying gig…
All that said, I haven’t ever actually done this in my own work, so my argument is theoretical rather than empirical and could be taken with a few grains of salt.
“Mastering” of course originally referred to the creation of a vinyl master by a transfer engineer. The limitations of lathes were specific challenges that recording engineers didn’t need to know or understand. And then there were the loudness wars where people were trying to get a competitive advantage by getting as much loudness out of physical media.In the absence of physical media mastering is largely a mystification. If we are transferring old master tapes, or making compilations of material from disparate sources, or adapting material from one reference listening level (film or TV) to another, we may need to do compression and eq (mastering) to get consistency sure but…
I produced an entire album in SC. I sent it to a friend (who is a mastering enginner) and he said it needed no mastering. However I did do some small cut in the high end and a tiny bit of limiting in Logic.
So, in my case, yes. But the music is not something that would get played in a nightclub or on commercial radio.
I think given the level of control you have over sound in SC, it’s easier to get to a sound that needs little to no processing than if you were using a DAW with plugins. In fact, I don’t really like even thinking in those terms anymore, of piano rolls and plugins, it just feels much more constrained. Obviously, that’s quite a personal thing.
One person whose sound design I really like is Eli Fieldsteel. Some of the sounds he makes on his tutorials sound really enticing to me.
I hope you enjoy SC a lot. I’m by no means its greatest practitioner, but I have found it a deeply rewarding experience to work with.
I guess I see a distinction between audio production and audio engineering. These overlap considerably, but I think they’re not quite the same: if production is primarily about art, then audio engineering is primarily about science, and as such, I think when it comes to audio engineering, “creative decisions that you wouldn’t have made otherwise” have a higher likelihood of just being incorrect (or, more accurately, sub-optimal – but who wants their mix to sound sub-optimal?). I’d be skeptical of making creative decisions in the production phase so that later, erm, “unusual” engineering decisions would be justified. (E.g., I just heard Suzanne Ciani in concert, and then heard complaints online about the engineering work by the sound crew. None of that is mainstream music, but poor engineering did have an adverse effect on some listeners’ experience.)
In any case, I’m just reporting what works for me – I can’t, and don’t want to, force anyone to do the same. There’s a difference between disagreement and coercion. Please do things your own way, and don’t do anything just because of an opinion I stated in a forum post
Thankfully the loudness war is over. 25 years ago we made albums absurdly loud with a dynamic range of 10, 8 or even somethings 6 db. Now all streaming platforms perform loudness compensations (with a few non important exceptions) which is a blessing. I use https://www.loudnesspenalty.com/ to check my masters and aim to be a little bit louder than Spotify, so that if anything, they turn my master down 0.x db instead of turning it up into a limiter which I have no control over. For a lot of electronic stuff it is not hard make it loud enough, ie. around -16 - -14 LUFS, without compression.
The most important thing about choosing to mix in SC or go external is automation in my opinion. If you can control the internal level of sounds on a note-by-note or event-by-event basis in the creation process then you can stay in SC. If, on the other hand you are doing sound experimentation stuff, crazy on-the-fly tweaks, live input of some sorts etc. then you really need automation of levels and sometimes EQs, which is so much easier handled in a DAW. The stuff I do fall in this category. I also love analog both for the sound and the hands on performance element with hard-to-do-recalls, so I often run SC stuff through my Studer mixer and an analog mastering chain with a lot of goodies. But that is just me and because I am fortunate enough to have the gear.
If I was to mix SC in the box I would probably use some VST plugins via the VSTPlugin extension, but as the work of @nathan and others show, you don’t necessarily need it.
I understand compression isn’t need for controlling unwieldy audio in sc, I never use compression for that. I’m mostly in the analog world and a good compressor with a fat transformer can do amazing things to a drum part or a whole track, it can make a track into something else. Transform a drum machine into something else and give it a groove it could never have . There is a story about the Raspberries, they couldn’t get Go all the way to feel like a hit, until someone run the whole track thru an api 2500 pumping at the same tempo of the track, made it a whole new song, the Beatles used compression on everything , excessively, and thank god for it. I’m sure this will be an unpopular post. But I like to look at other forms of music, not just navel gaze into the electronic enclave.
Autechre said in an interview that supercollider sounded finished, that it didn’t need mastering.
I agree however I have a project where I am obligated to use sc for music production and specifically for recreating a music piece. However since I am a newbie in algorithmic music I really can’t understand how to analyze a song so that the frequency and all the other details work to recreate a song in an 8bit audio style. Do you have any suggestions?
Even better, overdrive a r2r 1/4 tape. Or use a plug-in that does that, it might sound more pleasant. I dont use plugins do I don’t know what’s out there. Neve makes a tape saturation 500 rack
Oh , and I loaded sheepshaver on my Mac, I have Sc 1 , and all of Curtis road’s patches for sc1.
Sc1 sounds like a totally different program,
It can be tricky getting some patches to load as os9 has funny file associations.
I’m skeptical of claims that “supercollider sounds like xxx” although I’ve been susceptible also… last summer I listened to some old music I made with PD and thought it was low fi in a way I’ve never intuitively gotten to with SC… and then trying to replicate it, I found when you don’t interpolate buffers and use aliased oscillators SC sounds basically identical. The differences may be in what processes feel most intuitive, and of course digital approximations of analog processes are always approximate.
Think about the gaming consoles and home computers which produced that style of music because of their limitations. Apply the same limitations to your production.
For example, to sound like a Gameboy, use max two pulse wave uGens, a gritty noise gen (is there anything comparable to the gameboys noise in SC?) and max 2 samples at the same time running at 4 bit and around 11kHz.
Another example, sound like a c64. Use 3 oscillators with either triangle, pulse or square wave and run the sum through a hpf, bpf or lpf.
If I had gotten your assignment, I would find out what system the original was played on, try to closely enough recreate the sound generator and then focus on rhythm and melody using that.
While rarely ever finishing anything, I found using Jack and some Ardour quark from the quark list really cool for the combination of supercollider and DAW workflows. I broke Jack on my Mac though and never got Jack Bridge to build again in newer OS versions. Might try again some day.
I am using a recreation of a blinding lights that I made in garage band and I tried implementing the MIDIClient in order to either play the melody on a synthesizer (which doesn’t work cause it’s an old MIDI device and win 11 doesn’t recognize it) or use the audio. However when transforming the wav audio file to MIDI I am not sure how I can make it sound close to the original song.
