Loopy Pro: Create music, your way.

What is Loopy Pro?Loopy Pro is a powerful, flexible, and intuitive live looper, sampler, clip launcher and DAW for iPhone and iPad. At its core, it allows you to record and layer sounds in real-time to create complex musical arrangements. But it doesn’t stop there—Loopy Pro offers advanced tools to customize your workflow, build dynamic performance setups, and create a seamless connection between instruments, effects, and external gear.

Use it for live looping, sequencing, arranging, mixing, and much more. Whether you're a live performer, a producer, or just experimenting with sound, Loopy Pro helps you take control of your creative process.

Download on the App Store

Loopy Pro is your all-in-one musical toolkit. Try it for free today.

The Dangers of Audio Software Development

I had this very involved dream last night that I was writing a reverb plugin. I was using this particular song as a test during the development. When I woke up, the song was lodged firmly in my head and keeps popping in. I don't think I've actually heard this song in years. I've gotta say though, pretty appropriate for a dream about reverb.

IMAGE ALT TEXT

https://youtube.com/watch?v=p98PjtSfNWo

Comments

  • That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

  • @NeonSilicon said:
    I had this very involved dream last night that I was writing a reverb plugin. I was using this particular song as a test during the development. When I woke up, the song was lodged firmly in my head and keeps popping in. I don't think I've actually heard this song in years. I've gotta say though, pretty appropriate for a dream about reverb.

    IMAGE ALT TEXT

    https://youtube.com/watch?v=p98PjtSfNWo

    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

  • @ErrkaPetti said:

    @NeonSilicon said:
    I had this very involved dream last night that I was writing a reverb plugin. I was using this particular song as a test during the development. When I woke up, the song was lodged firmly in my head and keeps popping in. I don't think I've actually heard this song in years. I've gotta say though, pretty appropriate for a dream about reverb.

    IMAGE ALT TEXT

    https://youtube.com/watch?v=p98PjtSfNWo

    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    Oh yes, even the old Brand X stuff was amazing already.

    @NeonSilicon said:
    I had this very involved dream last night that I was writing a reverb plugin. I was using this particular song as a test during the development. When I woke up, the song was lodged firmly in my head and keeps popping in. I don't think I've actually heard this song in years. I've gotta say though, pretty appropriate for a dream about reverb.

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

  • @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

  • edited August 2021

    The psychology of app development begins long before the creation of an app by https://mlsdev.com/services/mobile-app-development. The idea behind the successful development process is simple: to succeed, your app developer must first understand his/her customer. Many app developers focus on the technical details of making an app, ignoring the basic goals to which the app will be directed. Most successful app developers share a common goal: to provide an effective means of communicating with their customers. To achieve this end, app developers must understand their customer's needs and desires, which lead them to understanding the market and developing an app that solves their customer's problems.

  • edited August 2021

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

  • @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

  • @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

    One idea: Since Apple's CPUs can do FFTs extremely fast, one fun experiment could be a convolutor which can also do live convolution of two input signals. There will be a short delay of course but offering FFT lengths of e.g. 128/256/512/1024/2048 and a few spectral shift/mangling options will certainly make an interesting effect.

  • @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

    One idea: Since Apple's CPUs can do FFTs extremely fast, one fun experiment could be a convolutor which can also do live convolution of two input signals. There will be a short delay of course but offering FFT lengths of e.g. 128/256/512/1024/2048 and a few spectral shift/mangling options will certainly make an interesting effect.

    Sort of a ring mod but with convolution? I'll have to think about it and do some playing around. Could be interesting.

  • @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

    One idea: Since Apple's CPUs can do FFTs extremely fast, one fun experiment could be a convolutor which can also do live convolution of two input signals. There will be a short delay of course but offering FFT lengths of e.g. 128/256/512/1024/2048 and a few spectral shift/mangling options will certainly make an interesting effect.

    Sort of a ring mod but with convolution? I'll have to think about it and do some playing around. Could be interesting.

    That's the idea but going beyond that. As you say, playing around with options and finding what's musically useful (unlike ring mod which is highly sensitive to pitch vs shift ratio and often sounds odd with tonal sources).

  • edited August 2021

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

    One idea: Since Apple's CPUs can do FFTs extremely fast, one fun experiment could be a convolutor which can also do live convolution of two input signals. There will be a short delay of course but offering FFT lengths of e.g. 128/256/512/1024/2048 and a few spectral shift/mangling options will certainly make an interesting effect.

    Sort of a ring mod but with convolution? I'll have to think about it and do some playing around. Could be interesting.

    That's the idea but going beyond that. As you say, playing around with options and finding what's musically useful (unlike ring mod which is highly sensitive to pitch vs shift ratio and often sounds odd with tonal sources).

    The concepts I'm researching are currently filter centric.

    Such as.. What would it sound like if you were able to start with the basic familiar operation of a LP filter, but have an ability to substitute that typical analog type feedback resonance with something else that you can imagine.

    My initial idea was about having control to create an unnatural sequence of harmonics to replace the "natural" harmonic series of harmonics.... i.e.. Use a different sequence of multipliers, enable pitch control of harmonics in programable ways, enable amplitude and pitch modulations of the artificial sudo-harmonics.

    Today, inspired by an article I read about old filter designs, I read about using diodes to "clip" feedback as a method for taming filter res runaway. I then played with two filters in series, LP with Res to produce harmonics, next through a HP to remove the OSC "carrier" signal, and produce just harmonics. Next I clipped the V-peaks off the harmonics and produced little pulse waves instead of sign waves. I mixed the pulse wave harmonics back into the OSC "carrier", and the result was really interesting, and very musically useful.. IMO..

    So I'm thinking, What else can be done to build "artificial resonant filter like effects"? How about saw wave harmonics.. Or PW modulated square wave harmonics? Or distorted harmonics? Of course these things can't exist in the analog world. But with computers they probably can.

    It might sound cool, or it might sound like nothing, I just think it's fun to be able to find ways to experiment with musical sound by looking or the parts that make up sounds, and seeing what it sound like if you make changes.

  • @horsetrainer said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

    One idea: Since Apple's CPUs can do FFTs extremely fast, one fun experiment could be a convolutor which can also do live convolution of two input signals. There will be a short delay of course but offering FFT lengths of e.g. 128/256/512/1024/2048 and a few spectral shift/mangling options will certainly make an interesting effect.

    Sort of a ring mod but with convolution? I'll have to think about it and do some playing around. Could be interesting.

    That's the idea but going beyond that. As you say, playing around with options and finding what's musically useful (unlike ring mod which is highly sensitive to pitch vs shift ratio and often sounds odd with tonal sources).

    The concepts I'm researching are currently filter centric.

    Such as.. What would it sound like if you were able to start with the basic familiar operation of a LP filter, but have an ability to substitute that typical analog type feedback resonance with something else that you can imagine.

    My initial idea was about having control to create an unnatural sequence of harmonics to replace the "natural" harmonic series of harmonics.... i.e.. Use a different sequence of multipliers, enable pitch control of harmonics in programable ways, enable amplitude and pitch modulations of the artificial sudo-harmonics.

    Today, inspired by an article I read about old filter designs, I read about using diodes to "clip" feedback as a method for taming filter res runaway. I then played with two filters in series, LP with Res to produce harmonics, next through a HP to remove the OSC "carrier" signal, and produce just harmonics. Next I clipped the V-peaks off the harmonics and produced little pulse waves instead of sign waves. I mixed the pulse wave harmonics back into the OSC "carrier", and the result was really interesting, and very musically useful.. IMO..

    So I'm thinking, What else can be done to build "artificial resonant filter like effects"? How about saw wave harmonics.. Or PW modulated square wave harmonics? Or distorted harmonics? Of course these things can't exist in the analog world. But with computers they probably can.

    It might sound cool, or it might sound like nothing, I just think it's fun to be able to find ways to experiment with musical sound by looking or the parts that make up sounds, and seeing what it sound like if you make changes.

    Once you get into doing things inside the filters, there's lots of fun to be had. The nonlinear/driven filters in Spirangle and GyroVibe are multi-stage state variable filters with waveshaping inside of each stage. The wavershapers are based on a set of constrained Bezier curves I made up. The drive parameter for these filters really alters something like 18 parameters in the waveshaper in each stage. For a four stage filter that's a lot of parameters to mess with. If you expose and play with all the parameters there's a huge range of sounds, and noise, you can get out of the idea. This is all very similar to using diode waveshapers inside of filters in the analog domain. In digital there's a lot more weird stuff you can do. Like maybe try putting a granular stage inside the filters.

    I'll add two warnings into these thoughts. The first is safety based. Protect your ears! Put limiters on the end of any signal chain where you are doing these kinds of experiments. I do this and then I have a physical mute and volume knob before my monitors and headphone amp. I mute and zero the output with the analog controls when I try something new and look at the output volume on my interface before I unmute and slowly bring up the volume. To me, playing with feedback and distortion is like working on a tube amp with high voltage except that I'm much more likely to poke around inside my amp without a ground strap than I am going to not follow the procedure I described above.

    The second thing is that you are probably going to need to really oversample if you start doing nonlinear things to your signal in a feedback loop. The more general non-constrained Bezier waveshaper I use easily ends up needing 8x oversampling to deal with the aliasing that happens.

  • @NeonSilicon said:

    @horsetrainer said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

    One idea: Since Apple's CPUs can do FFTs extremely fast, one fun experiment could be a convolutor which can also do live convolution of two input signals. There will be a short delay of course but offering FFT lengths of e.g. 128/256/512/1024/2048 and a few spectral shift/mangling options will certainly make an interesting effect.

    Sort of a ring mod but with convolution? I'll have to think about it and do some playing around. Could be interesting.

    That's the idea but going beyond that. As you say, playing around with options and finding what's musically useful (unlike ring mod which is highly sensitive to pitch vs shift ratio and often sounds odd with tonal sources).

    The concepts I'm researching are currently filter centric.

    Such as.. What would it sound like if you were able to start with the basic familiar operation of a LP filter, but have an ability to substitute that typical analog type feedback resonance with something else that you can imagine.

    My initial idea was about having control to create an unnatural sequence of harmonics to replace the "natural" harmonic series of harmonics.... i.e.. Use a different sequence of multipliers, enable pitch control of harmonics in programable ways, enable amplitude and pitch modulations of the artificial sudo-harmonics.

    Today, inspired by an article I read about old filter designs, I read about using diodes to "clip" feedback as a method for taming filter res runaway. I then played with two filters in series, LP with Res to produce harmonics, next through a HP to remove the OSC "carrier" signal, and produce just harmonics. Next I clipped the V-peaks off the harmonics and produced little pulse waves instead of sign waves. I mixed the pulse wave harmonics back into the OSC "carrier", and the result was really interesting, and very musically useful.. IMO..

    So I'm thinking, What else can be done to build "artificial resonant filter like effects"? How about saw wave harmonics.. Or PW modulated square wave harmonics? Or distorted harmonics? Of course these things can't exist in the analog world. But with computers they probably can.

    It might sound cool, or it might sound like nothing, I just think it's fun to be able to find ways to experiment with musical sound by looking or the parts that make up sounds, and seeing what it sound like if you make changes.

    Once you get into doing things inside the filters, there's lots of fun to be had. The nonlinear/driven filters in Spirangle and GyroVibe are multi-stage state variable filters with waveshaping inside of each stage. The wavershapers are based on a set of constrained Bezier curves I made up. The drive parameter for these filters really alters something like 18 parameters in the waveshaper in each stage. For a four stage filter that's a lot of parameters to mess with. If you expose and play with all the parameters there's a huge range of sounds, and noise, you can get out of the idea. This is all very similar to using diode waveshapers inside of filters in the analog domain. In digital there's a lot more weird stuff you can do. Like maybe try putting a granular stage inside the filters.

    I'll add two warnings into these thoughts. The first is safety based. Protect your ears! Put limiters on the end of any signal chain where you are doing these kinds of experiments. I do this and then I have a physical mute and volume knob before my monitors and headphone amp. I mute and zero the output with the analog controls when I try something new and look at the output volume on my interface before I unmute and slowly bring up the volume. To me, playing with feedback and distortion is like working on a tube amp with high voltage except that I'm much more likely to poke around inside my amp without a ground strap than I am going to not follow the procedure I described above.

    The second thing is that you are probably going to need to really oversample if you start doing nonlinear things to your signal in a feedback loop. The more general non-constrained Bezier waveshaper I use easily ends up needing 8x oversampling to deal with the aliasing that happens.

    I'll experiment with Spirangle and GyroVibe later today to see what those custom Bezier curves sound like.

    The limiter is a good idea. I do keep the volume lowered, but it's probably easier on the hardware to not hit it repeatedly with runaway transients.

    Thanks.

  • @horsetrainer said:

    @NeonSilicon said:

    @horsetrainer said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @rs2000 said:

    @NeonSilicon said:

    @horsetrainer said:
    That song could be used in a case study of the multitude of ways reverb can be used.

    Speaking of dreams. Once in a while I have dreams of original complete songs with vocals. But all memory of the song quickly disappears the moment I wake up. Then I saying to myself "dam... that was a good song. I wish I could remember it).

    Yeah, it's pretty much a masterclass in reverb.

    That feeling as a dream slips away and you know something useful was there is so frustrating.

    Korg Gadget to the rescue! Whenever I want to be quick laying down my ideas, Gadget is it.

    @ErrkaPetti said:
    [...]
    Frida made two albums together with Phil Collins in the late 80’s - very good!
    Phil is a marvelous songwriter and producer/musician!

    I didn't know or remember that Collins was involved with the song. But, when I woke up and had to go listen to it, I instantly thought "that has to be Phil Collins." So I looked up who was involved, that's an impressive list of people.

    Even the stuff he did in the 70s (look up "Brand X") is outstanding.

    @rs2000 said:
    [...]

    You must be an involved developer!
    That just reminded me of some AMS impulse responses... 😁

    Nah, I just have some fun dreams sometimes. I've never looked into anything with convolution reverbs or IR's. That's really kind of weird because I've used the math in several non-audio projects and the idea is really interesting. I should go do some playing around.

    The good thing with convolution is that it's so simple to do - Compute the fft of both signals and just multiply their spectra, then iFFT. An excellent audio playground, and simplifying the shifting of harmonics like @horsetrainer mentioned.
    Python and a few dedicated libs are so easy to learn, Inwould say anyone interested can do it with a bit of patience.

    Edit:

    Thanks for the link. I have heard it before, but it's been a while. I like pretty much everything Collins has been involved with.

    The really silly part of me not having used convolution in a plugin is that I have all the C++ code I need for it sitting in my library of old code I've written. I used convolution a bunch in some research I was involved with in the early 90's. I just need to drag the old code out and put it in my new framework -- should be pretty easy. As soon as I figure out a solution to my current global chip shortage induced issues, I really should go and try to make a simple IR based AU.

    One idea: Since Apple's CPUs can do FFTs extremely fast, one fun experiment could be a convolutor which can also do live convolution of two input signals. There will be a short delay of course but offering FFT lengths of e.g. 128/256/512/1024/2048 and a few spectral shift/mangling options will certainly make an interesting effect.

    Sort of a ring mod but with convolution? I'll have to think about it and do some playing around. Could be interesting.

    That's the idea but going beyond that. As you say, playing around with options and finding what's musically useful (unlike ring mod which is highly sensitive to pitch vs shift ratio and often sounds odd with tonal sources).

    The concepts I'm researching are currently filter centric.

    Such as.. What would it sound like if you were able to start with the basic familiar operation of a LP filter, but have an ability to substitute that typical analog type feedback resonance with something else that you can imagine.

    My initial idea was about having control to create an unnatural sequence of harmonics to replace the "natural" harmonic series of harmonics.... i.e.. Use a different sequence of multipliers, enable pitch control of harmonics in programable ways, enable amplitude and pitch modulations of the artificial sudo-harmonics.

    Today, inspired by an article I read about old filter designs, I read about using diodes to "clip" feedback as a method for taming filter res runaway. I then played with two filters in series, LP with Res to produce harmonics, next through a HP to remove the OSC "carrier" signal, and produce just harmonics. Next I clipped the V-peaks off the harmonics and produced little pulse waves instead of sign waves. I mixed the pulse wave harmonics back into the OSC "carrier", and the result was really interesting, and very musically useful.. IMO..

    So I'm thinking, What else can be done to build "artificial resonant filter like effects"? How about saw wave harmonics.. Or PW modulated square wave harmonics? Or distorted harmonics? Of course these things can't exist in the analog world. But with computers they probably can.

    It might sound cool, or it might sound like nothing, I just think it's fun to be able to find ways to experiment with musical sound by looking or the parts that make up sounds, and seeing what it sound like if you make changes.

    Once you get into doing things inside the filters, there's lots of fun to be had. The nonlinear/driven filters in Spirangle and GyroVibe are multi-stage state variable filters with waveshaping inside of each stage. The wavershapers are based on a set of constrained Bezier curves I made up. The drive parameter for these filters really alters something like 18 parameters in the waveshaper in each stage. For a four stage filter that's a lot of parameters to mess with. If you expose and play with all the parameters there's a huge range of sounds, and noise, you can get out of the idea. This is all very similar to using diode waveshapers inside of filters in the analog domain. In digital there's a lot more weird stuff you can do. Like maybe try putting a granular stage inside the filters.

    I'll add two warnings into these thoughts. The first is safety based. Protect your ears! Put limiters on the end of any signal chain where you are doing these kinds of experiments. I do this and then I have a physical mute and volume knob before my monitors and headphone amp. I mute and zero the output with the analog controls when I try something new and look at the output volume on my interface before I unmute and slowly bring up the volume. To me, playing with feedback and distortion is like working on a tube amp with high voltage except that I'm much more likely to poke around inside my amp without a ground strap than I am going to not follow the procedure I described above.

    The second thing is that you are probably going to need to really oversample if you start doing nonlinear things to your signal in a feedback loop. The more general non-constrained Bezier waveshaper I use easily ends up needing 8x oversampling to deal with the aliasing that happens.

    I'll experiment with Spirangle and GyroVibe later today to see what those custom Bezier curves sound like.

    The limiter is a good idea. I do keep the volume lowered, but it's probably easier on the hardware to not hit it repeatedly with runaway transients.

    Thanks.

    Don't listen specifically to how the Bezier curves sound in the app's. You may or may not like them. But, when thinking about how to use non-linear responses or clipping in filters it might be useful for you to hear what happens when you stack multiple filter stages with processing internal to the filter stages.

    As far as using Bezier curves for distortion, what I do in those two plugins is a very limited subset of what you can do with Bezier waveshaping. I used to make a distortion unit that cascaded two more general Bezier waveshapers and wrapped these with what is basically a version of LRC5 on either side. That could do some really extreme warping. The only constraint I had on those waveshapers was that the slopes of each Bezier curve section matched where the sections joined.

Sign In or Register to comment.