#22

Oh yeah one modal logical note about Jliat’s All Possible CDs is that there is a crucial difference between any impossible sound and all possible CDs.

With digital synthesis offering the familiar and well defined mappings between numbers and sounds (if not before) I think this question reduces to if impossible numbers exist. It seems to me that any number at all can exist by construction. But I hesitate to make mathematical and/or logical conclusions, as @Clarte’s insightful, æsthetic argument stands and challenges the link between music/sounds and mathematics.

1 Like
#23

This reminds me of the story of a record that cannot be played on all record players

that in turn reminds me of another Jliat project :slight_smile: tho imho he overplays the ‘danger’ of DC-offset silence: I believe many/most consumer means of playing a CD would filter it anyway

1 Like
#24

Good one:) If the sound is only what we can “record” or “playback” in some limited form (human ear, cd) and exclude any form of interpretation (pretending this can be done and/or makes sense) then all sounds are limited by bandwidth of the medium, if we include interpretations then oh boy fractal “sounds”, visual “sounds”, the sound that generate itself from itself, big bang, ommmmm:)

Edit: by limited by medium if we assume the world is quantized at the end and you can’t zoom to infinity:)

1 Like
#25

I had some different thoughts when I saw the topic:
I was thinking about how our brain processes sound in relation what we have learned. So sounds can be impossible for one person (to hear) because he/she hasn’t learned to hear it. An example would be language: you think you pronounce something like a native speaker. It sounds exactly like you‘ve heard it but it doesn’t because you‘re not able to hear the nuances (or even big differences).
If I remember it correctly that‘s the reason why Japanese pronounce r like l. It just sounds the same for them.

3 Likes
#26

As a relatively naïve and uneducated/untrained listener as well as sound producer this is highly relatable. “A snare hit is a snare hit, what’s with all the fuss?” :stuck_out_tongue_winking_eye:

2 Likes
#27

Reading the title, my mind immediately jumped to impossible colors, which relates to visual perception.

In the audio domain, a few “auditory illusions” or “audio paradoxes”, come to mind, such as the famous Risset/Shepards Tone which makes things sound like they are endlessly rising, or the rhythmic equivalent known as the risset rhythm which sounds like something is constantly speeding up, or in autechre’s case, constantly slowing down.

Not exactly “impossible”, but Alex Chechile’s work on auditory distortion product synthesis, is really fascinating. Basically, particular sounds get produced by a speaker, which causes your inner ear to make more resonances and sounds. Really trippy stuff.

10 Likes
#28

Yeah illusions! Illusion seems to be somehow the opposite of “impossible” :wink: Nice brain twister, thanks. Anyhow I just finished reading Moore’s “Dreams of Computer Music: Then and Now” (Computer Music Journal, 1996) linked by @dailybells which talks about illusions and Risset

Some interesting sounds have been discovered, however, most notably an entire category of “acoustical illusions,” named for their ability to “fool the ear” in ways analogous to the manner in which optical illusions “fool the eye.” (p. 30)

and

Based on recently understood principles of psychoacoustics, he created a long tone whose pitch gradually descends forever without getting any lower (like the barber pole stripes that appear to go down forever without getting anywhere) to accompany the pilot’s dream of endless falling. (p. 32)

Referring to this

3 Likes
#29

My attitude is: if it can be recorded and played back then it can be synthesized. This means that unless it can’t be represented in a recording then it can be synthesized. This doesn’t mean it’s practical or even viable, but fundamentally if a sample could embody it, it is synthesizeable.

1 Like
#30

Interesting rule of thumb. So, what does it mean for something to be “synthesized”?

Like, theoretically, I could load a recording of Beethoven 9 into a convolution reverb (or really any general convolution unit), hit it with a single impulse, and get that whole recording back as the output. Does that mean I synthesized an orchestra playing Beethoven?

2 Likes
#31

My attitude would be: algorithmically or systematically (mechanically, electrically, electronically etc) generated in a manner that is parameterised (or otherwise controllable) allowing for the coverage of a large sonic space where the space is many times larger than any media included. This is possibly a little broad, but it’s the best I can do off the top of my head.

1 Like
#32

Perhaps an impossible sound is one whose vibrations would be perfectly tuned to destroy its medium via extreme sympathetic resonance or high amplitude, or a combination of those.

10 Likes
#33

i think this comparison falters a bit. the word “tree” is only related to trees by convention and mutual agreement. further, it cannot reproduce a real tree in physical form (words are not spells, at least not in that sense) , and can mean any one individual tree or just the idea of a tree. a recorded sound can be played back however, generating a (imperfect) copy made out of physical waves in the air. these copies are not exactly clones, of course, since the recording and playback process will alter the sound in some ways. perhaps a photograph of a tree is a better analogy, or a 3d scan, mold or sculpture of a a tree?

i think there are no impossible sounds in theory, unless you find a way organize moving particles of air in a way that defies the laws of nature. that would make it impossible. in practice… that’s another question : )

however, i think it’s a good practice to think of impossible sounds that are yet possible, since it can put your mind in new places: like the sound of one hand clapping. or the word-sound at the beginning of the gospel of john: “in the beginning was the word, and the word was with god, and the word was god.” … i imagine the universe and our lives to be the echoes and reverb of that word.

1 Like
#34

It would be difficult to re-create the sound of thunder in a convincing way. We can sample it and play it back at high volume but it wouldn’t capture the real experience. Still when listening music a sample of thunder is good enough no need for it to be that convincing.

How about the sound of instruments that haven’t been invented yet combined with ones that have ? Is that impossible ?

1 Like
#36

Jonathan Green also talks, as beautifully as ever, about the Kauaʻi ʻōʻō in this episode of The Anthropocene Reviewed.

Starts at 12:40:

2 Likes
#37

sound is houdini, escaping in all manner of possible ways.

as someone said above, a great thought experiment.

i love it when sound bleeds into its sibling topics, or its related partners.

to me sound speaks to the body, but it’s also incredibly seismic. fractures in the particles of space time physics.

to me, impossible sound is those experienced by other bodies, other dimensions. in beings you are not, worlds you will never be in, universes yet to be born

the old tree in the woods - for me interpreted as, sound as phenomena, the breaking and snapping of mediums. but ‘true’ sound is within a living body, interpreted, via the senses - all invariably tactile

4 Likes
#38

Along the hmm let’s call them anthropological questions, embodied (in good feminist tradition), one observation from myself is the following:

By very much enjoying producing sounds very low in the sub-bass frequency range mmmmmm as well as all the way up there with barely audible miiiiip, I am myself making sounds which will be impossible in the future for myself and for others who will due to their age-induced disabilities¹ will have the frequency range of their hearing capped.

What i am saying is that as bodied creatures we too, like sonic vibrations, exist, are modulated and transform over time².

¹ and we can question whether they ought to be considered “disabilities”.
² assuming one agrees that sound is a temporal phenomenon.

2 Likes
#39

The human voice is still impossible to synthesize convincingly. Think about how easy it is to distinguish between people by listening to their voice, even when they’re making the same sounds. And this goes for any “untrained” person as well.

I agree with those who say we still have a loooong way to go in synthesis. I hope we will soon see some advancements in AI assisted synthesis programming with far more parameters than a human can handle. That could make impossible sounds common maybe. Not just talking emulating the human voice, but about a vast new range of expressive sounds which you could liken to it in complexity.

1 Like
#40

How about physical limitations? Since sound have a speed, is it possible to think of a frequency f.ex. that is impossible to make because it moves faster than the air or any other material can move?

#41

this is maybe the number one thought i have when thinking of sound design and “music” i want to make. i feel like maybe a combination of field recordings and synthesis is the best way to achieve unique original sounds. how to arrange them in a “musical” context with the gear available today is another story. thats one of my big problems. trying to fit my weirdo sound design into a structured format that isnt just amorphous ambient blob haze.

i guess the conclusion that i have arrived at is that it’s not really about making completely new sounds, but making the most pleasant unique sounds possible and then putting them together in layers to create pleasant new textures and environments as something to experience audibly.

elsa justel, alva noto, beatriz ferreyra, xenakis come to mind when thinking of unheard sounds. and those artists did most of their most interesting work around the 60s.

I do love the idea of coming up with new forms of synthesis or processing with modern hardware though. but its always a back and forth between that and recording samples. the human voice is a pretty good tool i find, when i cant get a texture im imagining elsewhere, i’ll try to make it with my mouth and record, then manipulate that, maybe layer it with synthesis

5 Likes
#42

I think you hit the nail on the head!

Sounds louder than 194dB in air and ~270dB in water are impossible because they do not propagate as vibrations. At that point you are shaking the medium hard enough to “break” it instead and make shockwaves.

Things traveling faster than the speed of sound in a medium also produce shockwaves, as do trombones! These non-linearities make them impossible(ish) to synthesize using physical modeling alone.

TL;DR: trombones :trumpet:doot doot

17 Likes