You are here

Physics anyone?

For fans of synths, pianos or keyboard instruments of any sort.

Re: Physics anyone?

Postby Arpangel » Tue Jun 02, 2020 10:40 am

blinddrew wrote:
Eddy Deegan wrote:Any possible sound can be reconstructed by layering pure sine waves of varying frequencies on top of each other
Pedantry alert: my understanding was that this is only true of tonal sounds and that some sounds are based on noise rather than tones?

Are you sure? I’m not sure either, but....
Isn’t "noise" white noise as an example, just sine waves at equal levels across the whole frequency spectrum?
User avatar
Arpangel
Jedi Poster
Posts: 4284
Joined: Sat Jul 12, 2003 12:00 am

Re: Physics anyone?

Postby Hugh Robjohns » Tue Jun 02, 2020 11:44 am

blinddrew wrote:Pedantry alert: my understanding was that this is only true of tonal sounds and that some sounds are based on noise rather than tones?

Yes... and transient conditions obviously can't be modelled with sine wave constructions either.

But any steady-state repeating waveform shape can be constructed using sine waves of different pitches.

And for Tony, while white noise does contain all frequencies at equal levels, it is not a steady-state signal. The presence or absence of different frequencies at any moment in time is random.
User avatar
Hugh Robjohns
Moderator
Posts: 27618
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Physics anyone?

Postby N i g e l » Tue Jun 02, 2020 10:29 pm

@ OP

An interesting article from 2011 but the basic technology hasnt changed !

https://www.soundonsound.com/sound-advi ... hear-pitch


innerchord wrote:
Arpangel wrote:I always wanted to route Midi controller data to individual operators in an FM system, is this at all possible now?

You mean like NI's FM7 offered almost 20 years ago? Where have you been? :)
I'm sure dozens of other FM synths can do the same now.

Reface DX works well with Behringer BCR2000, theres even a [2nd source] custom Reface controller.

Image
N i g e l
Frequent Poster
Posts: 841
Joined: Sun Aug 12, 2018 2:40 pm
Location: UK

Re: Physics anyone?

Postby Arpangel » Wed Jun 03, 2020 7:17 am

innerchord wrote:
Arpangel wrote:I always wanted to route Midi controller data to individual operators in an FM system, is this at all possible now?

You mean like NI's FM7 offered almost 20 years ago? Where have you been? :)
I'm sure dozens of other FM synths can do the same now.

I’ve got FM8, I only use it to play my old DX7 sounds, I don’t program now.
User avatar
Arpangel
Jedi Poster
Posts: 4284
Joined: Sat Jul 12, 2003 12:00 am

Re: Physics anyone?

Postby dxvxd » Sun Jun 07, 2020 12:56 am

I've been reading those synth secrets series. I believe it's the most comprehensive instructional on sound synthesis i've yet to see. So thank you Eddie Deegan for the suggestion.

Here's a new question:
Going back to the little experiment I proposed in the initial post to this thread. You sequence a series of repeated notes and then speed up the tempo until the repeated notes become audible as an oscillator. My question is, why is this frequency modulation and not amplitude modulation? I'm not changing the pitch, i'm just playing the same note over and over again so it seems to me the amplitude is modulating from zero to something not zero repeatedly.

This is how it would make sense to me:
If you held a note with an LFO acting on the pitch, and you increased the LFO speed eventually you would get frequency modulation.

If you did the same thing but with the LFO acting on the amplitude of the note, eventually you would get amplitude modulation.

I have no certainty as to whether that is right. I welcome correction!

Cheers,
Dave
dxvxd
New here
Posts: 8
Joined: Wed May 13, 2020 10:33 pm

Re: Physics anyone?

Postby Eddy Deegan » Sun Jun 07, 2020 1:28 am

dxvxd wrote:Going back to the little experiment I proposed in the initial post to this thread. You sequence a series of repeated notes and then speed up the tempo until the repeated notes become audible as an oscillator. My question is, why is this frequency modulation and not amplitude modulation? I'm not changing the pitch, i'm just playing the same note over and over again so it seems to me the amplitude is modulating from zero to something not zero repeatedly.

It's frequency modulation because you're varying the tempo, which determines the time between each successive note played and thus the frequency at which they are sounding. Think of it as 'notes-per-second', if that helps.

If you have 440 of them per second then your 'virtual oscillator' is generating an A4 tone of some kind. The manner in which this interacts with the actual frequencies of each individual note being played is complex, and I'm not going to pretend I understand the details, but as with many things FM-related it's worth experimenting!

dxvxd wrote:If you held a note with an LFO acting on the pitch, and you increased the LFO speed eventually you would get frequency modulation.

Well, technically you've got frequency modulation at all points there, because as soon as you start modulating the pitch (which is the frequency) you're, well, modulating the frequency.

Where it becomes really interesting is when the frequency at which the pitch is being modulated reaches the audio range itself. In this case the note being modulated is the carrier, and the LFO is the modulator. Depending on whether the frequencies of the carrier and the modulator are harmonically related (ie: integer multiples or divisions of one another) or not, you'll get a harmonic or inharmonic tone as a result.

dxvxd wrote:If you did the same thing but with the LFO acting on the amplitude of the note, eventually you would get amplitude modulation.

By definition, yes. As with frequency modulation you actually have amplitude modulation as soon as you start modulating the amplitude. Initially it's just an obvious variation of the loudness of the tone but once your modulation is operating in the audible range you'll hear new resulting tones emerge based on that modulation frequency.
User avatar
Eddy Deegan
Moderator
Posts: 4354
Joined: Wed Sep 01, 2004 12:00 am
Location: Brighton & Hove, UK
Some of my works.
Please consider supporting the SOS Forum Album project.
 

Re: Physics anyone?

Postby Hugh Robjohns » Sun Jun 07, 2020 5:06 pm

dxvxd wrote:Going back to the little experiment I proposed in the initial post to this thread. You sequence a series of repeated notes and then speed up the tempo until the repeated notes become audible as an oscillator. My question is, why is this frequency modulation and not amplitude modulation?

Because you are not changing (modulating) the amplitude of the sequence, but you are changing (modulating) its repetition rate (frequency). Simples!
User avatar
Hugh Robjohns
Moderator
Posts: 27618
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Physics anyone?

Postby dxvxd » Sun Jun 07, 2020 6:34 pm

Hugh Robjohns wrote:
dxvxd wrote:Going back to the little experiment I proposed in the initial post to this thread. You sequence a series of repeated notes and then speed up the tempo until the repeated notes become audible as an oscillator. My question is, why is this frequency modulation and not amplitude modulation?

Because you are not changing (modulating) the amplitude of the sequence, but you are changing (modulating) its repetition rate (frequency). Simples!


OK, but what if we're talking about one pitch? My note is repeating at a rate of 440 times per second. No variation in rate, no variation in frequency. No frequency modulation
dxvxd
New here
Posts: 8
Joined: Wed May 13, 2020 10:33 pm

Re: Physics anyone?

Postby Eddy Deegan » Sun Jun 07, 2020 6:46 pm

dxvxd wrote:
Hugh Robjohns wrote:
dxvxd wrote:Going back to the little experiment I proposed in the initial post to this thread. You sequence a series of repeated notes and then speed up the tempo until the repeated notes become audible as an oscillator. My question is, why is this frequency modulation and not amplitude modulation?

Because you are not changing (modulating) the amplitude of the sequence, but you are changing (modulating) its repetition rate (frequency). Simples!

OK, but what if we're talking about one pitch? My note is repeating at a rate of 440 times per second. No variation in rate, no variation in frequency. No frequency modulation.

I think I already answered this in my previous post, but in case you missed it:

Eddy Deegan wrote:It's frequency modulation becase you're varying the tempo, which determines the time between each successive note played, and thus the frequency at which they are sounding. Think of it as 'notes-per-second', if that helps.

If you have 440 of them per second then your 'virtual oscillator' is generating an A4 tone of some kind. The manner in which this interacts with the actual frequencies of each individual note being played is complex, and I'm not going to pretend I understand the details, but as with many things FM-related it's worth experimenting!

The point is that it doesn't matter if all the notes in the sequence are the same or not. It's the rate at which they are being played back at, in terms of 'notes-per-second' that is important.
User avatar
Eddy Deegan
Moderator
Posts: 4354
Joined: Wed Sep 01, 2004 12:00 am
Location: Brighton & Hove, UK
Some of my works.
Please consider supporting the SOS Forum Album project.
 

Re: Physics anyone?

Postby dxvxd » Sun Jun 07, 2020 8:04 pm

Eddy Deegan wrote:
dxvxd wrote:
Hugh Robjohns wrote:
dxvxd wrote:Going back to the little experiment I proposed in the initial post to this thread. You sequence a series of repeated notes and then speed up the tempo until the repeated notes become audible as an oscillator. My question is, why is this frequency modulation and not amplitude modulation?

Because you are not changing (modulating) the amplitude of the sequence, but you are changing (modulating) its repetition rate (frequency). Simples!

OK, but what if we're talking about one pitch? My note is repeating at a rate of 440 times per second. No variation in rate, no variation in frequency. No frequency modulation.

I think I already answered this in my previous post, but in case you missed it:

Eddy Deegan wrote:It's frequency modulation becase you're varying the tempo, which determines the time between each successive note played, and thus the frequency at which they are sounding. Think of it as 'notes-per-second', if that helps.

If you have 440 of them per second then your 'virtual oscillator' is generating an A4 tone of some kind. The manner in which this interacts with the actual frequencies of each individual note being played is complex, and I'm not going to pretend I understand the details, but as with many things FM-related it's worth experimenting!

The point is that it doesn't matter if all the notes in the sequence are the same or not. It's the rate at which they are being played back at, in terms of 'notes-per-second' that is important.

But I am NOT varying the tempo. It's 440 notes per second the whole time.
As Mr. Gordon Reid says, FM is simply a very fast vibrato. There is no vibrato in my example. Just the same note repeated over and over again very rapidly at the same speed.
dxvxd
New here
Posts: 8
Joined: Wed May 13, 2020 10:33 pm

Re: Physics anyone?

Postby Eddy Deegan » Sun Jun 07, 2020 8:27 pm

dxvxd wrote:
Eddy Deegan wrote:The point is that it doesn't matter if all the notes in the sequence are the same or not. It's the rate at which they are being played back at, in terms of 'notes-per-second' that is important.

But I am NOT varying the tempo. It's 440 notes per second the whole time.
As Mr. Gordon Reid says, FM is simply a very fast vibrato. There is no vibrato in my example. Just the same note repeated over and over again very rapidly at the same speed.

It's good you're digging into it at this level and I'd be interested in the specifics myself but I'm not going to claim to know the intricacies of the process here. What I will say is that you have an oscillator resonating at the frequency of the notes-per-second and you have the frequencies of the notes being played.

One of these will be acting as the carrier and one the modulator. Modulating anything at high frequencies will result in interesting outcomes. Maybe after reading the related synth secrets articles you'll be able to explain what's going on better than we can, but I'm primarily a musican and it's application rather than theory that makes noises so I tend to go with the ear rather than the brain when it comes to this sort of thing!
User avatar
Eddy Deegan
Moderator
Posts: 4354
Joined: Wed Sep 01, 2004 12:00 am
Location: Brighton & Hove, UK
Some of my works.
Please consider supporting the SOS Forum Album project.
 

Re: Physics anyone?

Postby Hugh Robjohns » Sun Jun 07, 2020 8:38 pm

dxvxd wrote:OK, but what if we're talking about one pitch? My note is repeating at a rate of 440 times per second. No variation in rate, no variation in frequency. No frequency modulation

Correct.

You have a steady note at 440Hz.

To have modulation, you need to modulate something with something else in some way.

There are plenty of pages on the web that explain the theory and maths of modulation in all it's forms.
User avatar
Hugh Robjohns
Moderator
Posts: 27618
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Physics anyone?

Postby Folderol » Sun Jun 07, 2020 9:04 pm

I'll probably regret this but...
Take it in slow motion first.
You have your 440Hz (carrier) and your secondary oscillator at some low frequency. If you observe this on an oscilloscope you'll see your carrier groups amplitude going up and down. Seems straightforward.

Now speed up your secondary oscillator, and stretch out the trace and you'll see some of the cycles with different amplitudes to the others - gets a bit more interesting.

Speed up your secondary oscillator till it's quite close to 440Hz, and you'll see that the amplitude is changing within each cycle. This is now adding harmonic component that wasn't there before and is therefore modulation.
User avatar
Folderol
Jedi Poster
Posts: 10875
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Physics anyone?

Postby dxvxd » Sun Jun 07, 2020 9:28 pm

Sorry guys, I don't mean to be a pain in the ass!
As eddie Deegan says, what matters in the end is the sound not the theory, and at some point this just becomes pedantic. And yes, much of the physics is over my head. So I'll drop the matter.

Best,
Dave
dxvxd
New here
Posts: 8
Joined: Wed May 13, 2020 10:33 pm

Re: Physics anyone?

Postby merlyn » Mon Jun 08, 2020 2:47 pm

If we put some numbers on it :

Sixteenth notes at 60 bpm gives a frequency of the attack points of 4Hz -- 60 bpm is one beat per second, and sixteenth notes give four attack points per beat. That will be heard as a stream of individual notes.

Now at a tempo of 600bpm the frequency of the attack points is 40Hz and that will be heard as a bass note. To get 440Hz you would need a tempo of 6600 bpm. Is that what you're using?

As noted above Dr. Chowning discovered FM synthesis when experimenting with fast vibrato. There is no vibrato here, so it's not FM synthesis.

You've made an oscillator out of a repeating note. It would also work with a drum sound or a click and that is how the human vocal chords work -- they produce a stream of clicks.

Using a note (or a fixed frequency) is how a sync oscillator works and sync oscillators are covered in Synth Secrets.
merlyn
Regular
Posts: 86
Joined: Thu Nov 07, 2019 3:15 am

Previous