Sign in to follow this  

Overcoming 'Sterility'

This topic is 4243 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

I'm having a problem with some of my tracks that's proving difficult to overcome: because I use a lot of synths, and soft-synths, the overall sound is very "sterile" - by which I mean, it sounds very computer generated (which it is). All of the instruments sound too pristine, the recording doesn't exist within any sort of real-world framework. Does anybody have any clever ideas to solve this? I've tried a few things which work okay, but still haven't really solved my problem. (I think the real solution is to get some more natural sounding instruments and integrate them into my songs) - Adding recorded room ambience at low volume levels in certain parts of the track - Applying subtle distortion to select instruments - adding atmospheric synthesizer effects with reverb and delay at very low volume levels to certain parts of the track Any other ideas?

Share this post


Link to post
Share on other sites
Typically reverb is used with sampled instruments, to make a sample recorded with very close microphones in an anechoic room sound like what a person would hear in a normal room at a fair distance. Careful equalization and linear distortion and stereo placement of imaginary sources are also important to create the illusion of listening to real instruments.
But this illusion applies to sampled instrument and patchworks of actual recordings, or to synthesizers used to simulate real instruments: if you are using synthesizers with more freedom, your music will sound artificial whatever you do because, as you say, it "doesn't exist within any sort of real-world framework".
Maybe what you don't like is not the artificial timbre of your synthesizers but more generic problems like flat dynamics, muddy overlaps and hollow gaps in the frequency spectrum, inhuman quantization of note times and lengths, clipping, humming and other types of involuntary noise, etc.

Share this post


Link to post
Share on other sites
Quote:
Original post by LorenzoGatti
Typically reverb is used with sampled instruments, to make a sample recorded with very close microphones in an anechoic room sound like what a person would hear in a normal room at a fair distance. Careful equalization and linear distortion and stereo placement of imaginary sources are also important to create the illusion of listening to real instruments.
But this illusion applies to sampled instrument and patchworks of actual recordings, or to synthesizers used to simulate real instruments: if you are using synthesizers with more freedom, your music will sound artificial whatever you do because, as you say, it "doesn't exist within any sort of real-world framework".
Maybe what you don't like is not the artificial timbre of your synthesizers but more generic problems like flat dynamics, muddy overlaps and hollow gaps in the frequency spectrum, inhuman quantization of note times and lengths, clipping, humming and other types of involuntary noise, etc.


Hey; Thanks for the feedback!

I'm not using a lot of sampled instruments, the only "real world" instruments in my song in which this is particularily noticable is a looped acoustic drum-kit. Obviously, in creating electronic music the music will sound artificial.

However, when I compare the overall sound of my recording to say, material on Aphex Twin's "Window Licker", I can hear a great difference. Obviously, I'm no Aphex Twin, but in comparing it to the averaged frequency spectrum of my song, there seem to be no noticable gaps in the spectrum; yet other material (Aphex Twin, Orbital, etc) sounds substantially more organic.

I suppose this problem could originate from using software-synths instead of actual analog hardware, and the fact that AT and Orbital probably get their stuff mastered by people that know what they're doing.

As far as stereo placement goes, I'll try adjusting that. There's already reverb on each instrument, although generally it's mixed so low as to be close to inaudible; but I will take a stab and improving it a bit.

Is there a consensus on whether or not every instrument should be subject to the same reverb. Is it considered reasonable / unreasonable to apply in addition to track reverb, a small amount of reverb on the entire track?

Lastly, you seem pretty knowledgable; any books / magazines you'd reccomend?

Share this post


Link to post
Share on other sites
All those things Lorenzo pointed out are great. Solid ideas to tinker with!

Your best bet? Well every composition is different and every mix is different, so you may get more help if you had a specific example. Like, post a clip of the song so we can deduce what exactly THAT song needs. Because, what if you are going for a "sterile" sound? Maybe it's not instrument choice but your choice of multiband compression. Who knows, take every song one at a time. There are certain fundamentals you can always count on, but mix to what feels good for that song.

Tony

Share this post


Link to post
Share on other sites
Quote:
Original post by LorenzoGatti
Maybe what you don't like is not the artificial timbre of your synthesizers but more generic problems like flat dynamics, muddy overlaps and hollow gaps in the frequency spectrum, inhuman quantization of note times and lengths, clipping, humming and other types of involuntary noise, etc.
This is the best summary of what I find to be wrong with the tracks I author. What I find is after I put together a track, I have to run through it note by note messing with the levels and timing to make the track feel more alive. Very time consuming.

YMMV but I find that if I steer clear of mastering tools until I've finished the track it will come out sounding better; if the levels already sound good, mastering just adds some punch to it.

Share this post


Link to post
Share on other sites
Quote:
Original post by skjinedmjeet
I suppose this problem could originate from using software-synths instead of actual analog hardware

Maybe you are using the wrong softsynths: there are a number of more or less faithful emulations of old analog hardware, which try to reproduce the imperfections of actual electronic devices, but most softsynth designers prefer to invest their complexity budget in flexibility and advanced features and their CPU budget in massive calculations, oversampling, etc. They are not worse synths because they sound "sterile", it's simply a different path leading away from analog systems and towards complete user control.
Can you explain what software synths are you using and what analog hardware would sound good in your music?



Share this post


Link to post
Share on other sites
Here's a link to a song that I felt had a particular problem with it. Unfortunately, my server will not allow me to upload MP3s, so the download is a ZIP file.

I've taken a stab at improving it; any comments are thoroughly welcome. Two things that I'm already aware of is a volume spike in the bass near the very beginning of the song - also, the beginning of the song gets a bit long in the teeth for my tastes.

As far as soft-synths go; I'm on the cheap, so I've been using mostly default synthesizers from Jeskola Buzz, the default Rebrith settings (i.e., no mod) and aside from that the only audio sources in this particular song are some old drum loops that I have sliced up and "recomposed" into new patterns. I have an older Sound Canvas unit that I used some xylophone noises from, but they're not used excessively, and mixed very quietly.

Share this post


Link to post
Share on other sites
One technique i like to use incorporating synths with ambient-style tracks is to set up 2 mics (for stereo effect) at the opposite side of the room to the computer, then record the synth tracks playing from the computer speakers through the mics. Using this mixed in fairly quietly will add some natural subtle reverb and ambience to the sound of the synths. Variations on this can also be used creatively for more unusual sounds; I once attached a speaker to length of drainpipe and placed a mic at the other end and the resulting sound was almost like a slapback delay with many reflections but little interference between them. People often undestimate or overlook the creative power of a speaker, a mic and household objects as effects processors.

Share this post


Link to post
Share on other sites
Hey! You're basically describing what's known as "Re-Amping". That is, usually when you take a DI signal (like from a split guitar that's already been tracked) and then send it out to an amp in another room, mic the amp and record the mic and any room ambience you may want.

Again, people usually do this with guitars to try it in different cabs and mic placement configurations, but it works great for warming up synths. Don't overlook some great amp-sim plug-ins, but the best audio you can record is the kind that actually displaces/pushes air. Mic'ing up amps is where it's at. The sooner you're synth sounds turn into malleable audio, the more windows open up to get the mix you want.

Tony

Share this post


Link to post
Share on other sites
Maybe you just need more 'pro' software to synthesize and mix. Like cubase for sequencing and Tassman Aoplied Acoustic systems, or Reaktor. Or a bunch of fancy hardware synthesizers. It's very hard to get a pro sound out of a freeware or middle-of-the-road soft synth. You could end up twaeking/mixing your sounds to eternity, when it's just quality of the generated sounds that's the problem.
A 'trick' i did years ago (don't need it anymore..) was to record a sterile sound on a chrome cassette tape,and again record that sound on my sound card. By the way- your sound card is a decent one? :D

Share this post


Link to post
Share on other sites
I don't think you need more gear; you can apply what you have differently.

I'd suggest trying a groove quantize or similar, non-regular quantization format.

I'd also suggest adding subtle percussion tracks that hit in back-beat and on the "up" beats (although for soft, dreamy stuff, percussion doesn't really work).

Also try changing your lead instrument to something with more character (typically, sharper) -- perhaps Eq it with a broad passband, cutting the lows and highs (for that radio/megaphone sound).

Listen to tracks that you like that are similar to what you do, and really de-compose what's going on in those tracks -- how many instruments? How are they played, panned and mixed? What ambience is there? What expression, in rythm variation (off-beat) or strike?

Share this post


Link to post
Share on other sites
Sign in to follow this