Can you recommend some pieces of music in the ambient-esque genre that achieve good spacial placement of sounds without the use of artificial reverberation?
Obviously, concrete source material has the spatialization of mic placement and recording environment, but combining that with synthetic sounds, or creating a sense of depth in all synthetic material is becoming increasingly interesting to me.
Do you guys have any techniques (panning, EQ curves) to achieve this sort of thing in your own work?
hello, new to the forum. Find this very interesting topic.
From what I know, apart from convolution reverbs many of the reverb plugins nowadays are doing physical modeling to mimic the spacial FX. And conceptually if you achieve the same FX with a bunch of EQ or pannning, to me it’s the same “artificial” approach
Anyway, i believe 1 simple delay(FB < 100%) + 1 automated LPF(because hi freq vanish really fast in natural reverb ) + 1 automated stereo expander(optional) will do the trick
also some of the producers today are using empty room and microphone as their Send/Return track for getting natural reverb, think that might be another solution for your problem
Wow, these are some well thought out ideas and product recommendations that could definitely help enhance depth/distance in a mix while not sounding so much like an added effect.
The ideas I had been working with are:
-The inverse square law as a guide for amplitude relationships between sound objects
-Tilt or shelf EQ to simulate the varying high frequency attenuation of different environments
-Compression to reduce the dynamic range of a sound happening at a greater distance.
-Granular spatial scattering to bring the presence of a sound closer/further
I believe much of what is being suggested here could end up sounding more interesting than using reverb to place sounds objects. I’ll read up the recommendations posted earlier and try some stuff out, thanks!
also +1 on the use of an actual room and mic for reverb, that technique has been around forever and lends a distinctly personal feel to a recording in my experience. definitely something to try with every room in your place.
He covers historical approaches to reverb as well as some discussion of more modern approaches – it’s a great and super nerdy listen with tons of fun history on the subject and might be some good inspiration for coming up with alternative approaches to spatialization.
This talk is really informative! The slide show really helps too. Living in the area, I wish I’d known about it and had gone to hear it. It was nice to Barry Blesser and his work referenced a couple times. His book fits right into this discussion. The talk also brought back memories of the primitive digital reverbs we had in the Cornish College electronic music studio.
The CCRMA books are full of useful historical tidbits that sound surprisingly good today. For instance, on the old reverb algorithms, the spatialization is done as sort of a last-stage DSP block.
If you already have a stereo signal, and want to create a quadraphonic signal, the mixing matrix recipe is great: https://ccrma.stanford.edu/~jos/pasp/Schroeder_Reverberators.html
(Even if you don’t want quadraphonic, the channel combinations can provide interesting effects)
Essentially, just use extremely small delay lines on each channel that you want to upmix to (.05 seconds or so… check the diagram for their recommendations).
Finally, there are some great recipes on MusicDSP for enhancing stereo width: http://musicdsp.org/archive.php?classid=4#256
I’ve linked my favorite one, which is to translate the L/R stereo signal into Mid/Side. Even though it’s provided as code, the basics of the algorithm are easy to use in many environments.
Awesome, glad I could help! I thought M/S encoding and decoding was some sort of DSP dark art until I looked up the algorithm a month or two back for my Reaktor series. The musicDSP archive and mailing list are excellent resources for that sort of thing.
I’ve grown to really enjoy M/S since then, and have been implementing it into our products whenever it seems interesting. It’s computationally inexpensive (especially compared to a lot of algorithmic reverbs), yet it can have a very dramatic effect on the final signal.
