ilovefuzz.com

multi_s wrote:you should also generate a frequency domain graph of the sine response no? very easy to see what harmonics its generating instead of staring at the time domain version, which while interesting, dosn't as easily give insight into what sort of overtones are happening.

edit:

it may also be interesting to run the sine test at a few different amplitudes. Ive looked at a few different fuzzes on scopes that give some extra shape characteristics when you feed it larger signals. (the bazz fuss comes to mind).

like this?

response with different amplitude is crucial.

smallsnd/bigsnd wrote:response with different amplitude is crucial.

Agreed. I have an overdrive design that although is quite well received, I am unhappy with because its decay has a bit of non-harmonic distortion. I had to play guitar through the prototype with my scope at different points in the circuit to find it.

I think on Monday I'm going to measure just my sound card and my fuzz pedals in by-pass mode. I also have two more pedals to contribute to this project.

AC128 wrote:

multi_s wrote:you should also generate a frequency domain graph of the sine response no? very easy to see what harmonics its generating instead of staring at the time domain version, which while interesting, dosn't as easily give insight into what sort of overtones are happening.

edit:

it may also be interesting to run the sine test at a few different amplitudes. Ive looked at a few different fuzzes on scopes that give some extra shape characteristics when you feed it larger signals. (the bazz fuss comes to mind).

like this?

well no not exactly what i meant, although any additional graphs are good .

what i meant was, with your sine input, the one with a fixed frequency, you have the time series output. there is a theory that goes something like "any time series signal can be represented as a sum of harmonically related sinusoids", so if you view the frequency domain of that same sine output, you will see which sinusoids make up the output given a pure tone input. you dont have to redo anything just take the time series and apply a Fourier transform to it, or pmtm, or whatever your flavor is.

multi_s wrote: "any time series signal can be represented as a sum of harmonically related sinusoids"

like additive synthesis?

Well im not really sure what the definition of additive synthesis is. Check out this applet maybe you will see better what im talking about. You can draw some arbitrary wave and it shows you the harmonic composition. I wouldn't worry too much about that he lists sines and then cosines separately it is just a way of inferring phase (a sine + a cosine = some sinusoid at some particular phase).

http://www.falstad.com/fourier/

multi_s wrote:Well im not really sure what the definition of additive synthesis is. Check out this applet maybe you will see better what im talking about. You can draw some arbitrary wave and it shows you the harmonic composition. I wouldn't worry too much about that he lists sines and then cosines separately it is just a way of inferring phase (a sine + a cosine = some sinusoid at some particular phase).

http://www.falstad.com/fourier/

do you mean like a frequency by harmony graph? I don't get it... or amplitude by harmony?

multi_s wrote:Well im not really sure what the definition of additive synthesis is.

WIKI: Additive synthesis is a technique of sound synthesis that creates musical timbre by explicitly adding sinusoidal overtones together.

multi_s wrote:Check out this applet maybe you will see better what im talking about. You can draw some arbitrary wave and it shows you the harmonic composition. I wouldn't worry too much about that he lists sines and then cosines separately it is just a way of inferring phase (a sine + a cosine = some sinusoid at some particular phase).

http://www.falstad.com/fourier/

will do, thanks.

*edit* woa, this is very cool!

do you know of a proggie that does what you're asking? maybe a vst plugin?

I want a fuzz that turns a sine wave into this

AC128 wrote:I want a fuzz that turns a sine wave into this

I have something that actually looks pretty close to that...

Schlatte wrote:

AC128 wrote:I want a fuzz that turns a sine wave into this

I have something that actually looks pretty close to that...

please tell me more!

AC128 wrote:
please tell me more!

It was mod of my old waveform-generator... you basically have two separated sawtooth-generators... the first one is sawtooth down, the other one sawtooth up... you control the amplitude of the second sawtooth (that has about three to four times the frequency of the first one) with the first sawtooth. you end up with pretty much your shown signal... at least it's pretty close
I actually thought about doing it in a pedal or something.. but.. you need dual voltage supply and too much space for stuff.. and gäh... i just was sick of it after a while... I think it's a good thing that the waveform-generator broke irreparably a short while after..

Schlatte wrote:
do you mean like a frequency by harmony graph? I don't get it... or amplitude by harmony?

im not sure what is frequency by harmony or what you mean by it. The graph would have frequncy on the x axis and amplitude on the y axis. aka taking the fourier transform of a time series. or use PMTM or some other spectrral estimation technique. Fourier transform will easily be the most popular because it can be done using a technique known as FFT (Fast Fourier Transform) which is a particular implementation that is faster to compute using ... computers.

It is to me, 'the frequency domain representation' of a signal.

AC128 wrote:I want a fuzz that turns a sine wave into this

You could do this no problem with dsp and a look up table. Basically just map the transfer function. Imagine you sample and get a value x. somewhere there is a chart that says: value x maps to value y, the output. It can be completely bizarre and non linear, who cares, you just make a map and thats it. I will try to demonstrate when i have time. Of course doing things analog is always sexier, but this could be used as a brute force technique, or hjsut for experimenting, ie "what would that sound like"

also check out Falstads other applets if your into physics. the guy is a total genius, some great ways to visualize theory and also a great online circuit simulator. http://www.falstad.com/mathphysics.html

Also yes there are lots of programs that can do FFT. there are probably ones online that do it for you but if you want send me the data and i will try to import it into matlab or some other software i have. I have a cheap usb oscillocope that actually does FFT in 'real time' to the signal going in. I will try to make a screen capture with a teh crowther hotcake to see whats what . Its not perfect but again, just to get an idea.

multi_s wrote:You could do this no problem with dsp and a look up table. Basically just map the transfer function. Imagine you sample and get a value x. somewhere there is a chart that says: value x maps to value y, the output. It can be completely bizarre and non linear, who cares, you just make a map and thats it. I will try to demonstrate when i have time. Of course doing things analog is always sexier, but this could be used as a brute force technique, or hjsut for experimenting, ie "what would that sound like"

also check out Falstads other applets if your into physics. the guy is a total genius, some great ways to visualize theory and also a great online circuit simulator. http://www.falstad.com/mathphysics.html

Also yes there are lots of programs that can do FFT. there are probably ones online that do it for you but if you want send me the data and i will try to import it into matlab or some other software i have. I have a cheap usb oscillocope that actually does FFT in 'real time' to the signal going in. I will try to make a screen capture with a teh crowther hotcake to see whats what . Its not perfect but again, just to get an idea.

great info, thanks!

I uploaded the files and sent you a pm with the details.