Not a fan of one of the genres?





(You can't hide all the genres)
 

New VST similar to Venm / Logic Clip Distortion

All about Software and Hardware (Shared HS/HC)
  

Postby Harha » 27 Mar 2017, 14:34

Soundphase; Good analysis as said, but what about the 'tone' filter, which is a highpass filter?

https://documentation.apple.com/en/logicstudio/effects/index.html#chapter=3%26section=2%26tasks=true

Wouldn't it break your results in that graph, meaning if the highpass filter comes after symmetry, the resulting waveform would be centered again?

After being amplified by the Drive value, the signal passes through a highpass filter.


I'm just relying on this information provided by the Clip Distortion documentation though.

iCore 2 Duo @ 3.4Ghz | 4Gb 800Mhz DDR2 RAM | M-Audio Delta Audiophile 2496 | Windows 7 x64 & Mac OSX 10.6.8 Dualboot | KRK Rokit RP5 G2 | Access Virus TI Desktop | Access Virus C Desktop | Korg K49 MIDI Keyboard
User avatar
Harha (24)
State Native
 
Posts: 861
Location: Finland (fi)



Postby Netris » 29 Mar 2017, 01:49

Soundphase:if you make a new update, remove Team AiR tag, and make a new professional skin with more standard colors and without people in it lol, then it would be a nice contender to bad bus mojo


I'm 100% with you on this one ^^
Image
Netris (28)
State Citizen
 
Posts: 127
Location: Belgium (be) Liège

Postby Soundphase » 29 Mar 2017, 10:29

Harha:Soundphase; Good analysis as said, but what about the 'tone' filter, which is a highpass filter?

https://documentation.apple.com/en/logicstudio/effects/index.html#chapter=3%26section=2%26tasks=true

Wouldn't it break your results in that graph, meaning if the highpass filter comes after symmetry, the resulting waveform would be centered again?

After being amplified by the Drive value, the signal passes through a highpass filter.


I'm just relying on this information provided by the Clip Distortion documentation though.

the highpass filter really just modifies the shape of the slope after the symmetry is altered, it doesn't center the waveform. I tried experimenting with that at the time and I could make the same waveform shape as the logic clip dist waveform I posted, but the sound was completely different (in a negative way) and this was the main problem
Soundphase (27)
State Celebrity
 
Posts: 1012
Location: Netherlands (nl)

Postby Harha » 29 Mar 2017, 12:35

Soundphase:
Harha:Soundphase; Good analysis as said, but what about the 'tone' filter, which is a highpass filter?

https://documentation.apple.com/en/logicstudio/effects/index.html#chapter=3%26section=2%26tasks=true

Wouldn't it break your results in that graph, meaning if the highpass filter comes after symmetry, the resulting waveform would be centered again?

After being amplified by the Drive value, the signal passes through a highpass filter.


I'm just relying on this information provided by the Clip Distortion documentation though.

the highpass filter really just modifies the shape of the slope after the symmetry is altered, it doesn't center the waveform. I tried experimenting with that at the time and I could make the same waveform shape as the logic clip dist waveform I posted, but the sound was completely different (in a negative way) and this was the main problem


Hmm. Interesting, I've also been programming my own clip distortion for a few days now because I want to see how close I can get to the original thing and currently both the waveform and the sound is quite similar, only if I apply symmetry in either + or - direction does the sound get way different. My signal flow is what apple describes on the documentation about clip distortion.

Also I don't quite understand how clip distortion's drive algorithm works. It can't be just signal overdrive by certain dB amount, it seems like it does square the waveform of a 909 kick for example, but the square shape decays into crunched sinewaves very quickly instead. I've been trying to model it in matlab with little to no luck at the moment...

Code: Select all
float input = sample;
float output = sample;
float a = std::sin(((drive * 0.5f) / 100.0f) * (PI / 2.0f));
float k = 2.0f * a / (1.0f - a);
output = ((1.0f + k) * (input)) / (1.0f + k * std::abs(input));
output = output * uV(drive);


This gets me slightly closer but it's still not quite right. That is a modified guitar distortion overdrive algorithm before multiplying the input signal by the actual overdrive amount in uV. What it does is that it just simply transforms an input signal from it's original state into near-square-wave shape based on the input drive amount.

Or then the overdrive is just very simple but instead the actual distortion algorithm differs in so that the waveform quickly decays from a square one to a more sinusoidal one. I don't know, it's trial and error I guess...

iCore 2 Duo @ 3.4Ghz | 4Gb 800Mhz DDR2 RAM | M-Audio Delta Audiophile 2496 | Windows 7 x64 & Mac OSX 10.6.8 Dualboot | KRK Rokit RP5 G2 | Access Virus TI Desktop | Access Virus C Desktop | Korg K49 MIDI Keyboard
User avatar
Harha (24)
State Native
 
Posts: 861
Location: Finland (fi)

Postby Phazon » 29 Mar 2017, 15:14

Harha:Hmm. Interesting, I've also been programming my own clip distortion for a few days now because I want to see how close I can get to the original thing and currently both the waveform and the sound is quite similar, only if I apply symmetry in either + or - direction does the sound get way different. My signal flow is what apple describes on the documentation about clip distortion.

Also I don't quite understand how clip distortion's drive algorithm works. It can't be just signal overdrive by certain dB amount, it seems like it does square the waveform of a 909 kick for example, but the square shape decays into crunched sinewaves very quickly instead. I've been trying to model it in matlab with little to no luck at the moment...

Code: Select all
float input = sample;
float output = sample;
float a = std::sin(((drive * 0.5f) / 100.0f) * (PI / 2.0f));
float k = 2.0f * a / (1.0f - a);
output = ((1.0f + k) * (input)) / (1.0f + k * std::abs(input));
output = output * uV(drive);


This gets me slightly closer but it's still not quite right. That is a modified guitar distortion overdrive algorithm before multiplying the input signal by the actual overdrive amount in uV. What it does is that it just simply transforms an input signal from it's original state into near-square-wave shape based on the input drive amount.

Or then the overdrive is just very simple but instead the actual distortion algorithm differs in so that the waveform quickly decays from a square one to a more sinusoidal one. I don't know, it's trial and error I guess...


Turning from a square wave to a sine wave based on the amplitude decay sounds like a hyperbolic function, or something like arctan to me.

Try this:

Code: Select all
float input = sample;
float output = sample;
float asymmetry = asymmetry_slider_value; // 0 (negative signal only) to 1 (positive signal only)

input = input * drive; // Drive is just gain pre-distortion

// Apply positive distortion
if (input > 0)
{
    output = atan(input) * asymmetry;
}

// Apply negative distortion
else if (input < 0)
{
    output = atan(input) * (1-asymmetry);
}
Phazon
State Newcomer
 
Posts: 33
Location: United Kingdom (uk)

Postby Harha » 29 Mar 2017, 23:37

Phazon:
Harha:Hmm. Interesting, I've also been programming my own clip distortion for a few days now because I want to see how close I can get to the original thing and currently both the waveform and the sound is quite similar, only if I apply symmetry in either + or - direction does the sound get way different. My signal flow is what apple describes on the documentation about clip distortion.

Also I don't quite understand how clip distortion's drive algorithm works. It can't be just signal overdrive by certain dB amount, it seems like it does square the waveform of a 909 kick for example, but the square shape decays into crunched sinewaves very quickly instead. I've been trying to model it in matlab with little to no luck at the moment...

Code: Select all
float input = sample;
float output = sample;
float a = std::sin(((drive * 0.5f) / 100.0f) * (PI / 2.0f));
float k = 2.0f * a / (1.0f - a);
output = ((1.0f + k) * (input)) / (1.0f + k * std::abs(input));
output = output * uV(drive);


This gets me slightly closer but it's still not quite right. That is a modified guitar distortion overdrive algorithm before multiplying the input signal by the actual overdrive amount in uV. What it does is that it just simply transforms an input signal from it's original state into near-square-wave shape based on the input drive amount.

Or then the overdrive is just very simple but instead the actual distortion algorithm differs in so that the waveform quickly decays from a square one to a more sinusoidal one. I don't know, it's trial and error I guess...


Turning from a square wave to a sine wave based on the amplitude decay sounds like a hyperbolic function, or something like arctan to me.

Try this:

Code: Select all
float input = sample;
float output = sample;
float asymmetry = asymmetry_slider_value; // 0 (negative signal only) to 1 (positive signal only)

input = input * drive; // Drive is just gain pre-distortion

// Apply positive distortion
if (input > 0)
{
    output = atan(input) * asymmetry;
}

// Apply negative distortion
else if (input < 0)
{
    output = atan(input) * (1-asymmetry);
}


Thanks for the suggestion. I did experiment with hyperbolic functions before without much luck. Anyways, I gave atan a try again now when the plugin is more complete than back then and it is quite close indeed. Tanh seems to be even closer, the clip distortion algorithm might be some kind of approximation of tanh, not quite sure yet. D: Up to this point I have been using a 3-stage distortion algorithm that has a threshold setting controlling how much of the signal is driven through static (threshold <-> -threshold), quadratic and 1.0 <-> -1.0 functions.

WARNING, loud volume!


There's the same plain 909 Kick sample driven through first clip distortion and then my distortion with exact same settings and lastly the difference between those two samples. This might even be it, nearly, the difference might come mainly from my IIR filters being 12dB/Octave and Clip Distortion's filters being 6dB/Octave, so I just have to adjust my filter coefficients a bit better. I shall experiment more next weekend if I have some time.

iCore 2 Duo @ 3.4Ghz | 4Gb 800Mhz DDR2 RAM | M-Audio Delta Audiophile 2496 | Windows 7 x64 & Mac OSX 10.6.8 Dualboot | KRK Rokit RP5 G2 | Access Virus TI Desktop | Access Virus C Desktop | Korg K49 MIDI Keyboard
User avatar
Harha (24)
State Native
 
Posts: 861
Location: Finland (fi)

Postby D-Charged » 14 Apr 2017, 15:24

Soundphase:if you make a new update, remove Team AiR tag, and make a new professional skin with more standard colors and without people in it lol, then it would be a nice contender to bad bus mojo


thats too much for me to care about haha ;P i wont continue on it, i have it the way i like now :)
http://www.facebook.com/dcharged
http://www.soundcloud/dcharged
User avatar
D-Charged (20)
State Outsider
 
Posts: 9
Location: Sweden (se) Helsingborg

Postby D-Charged » 14 Apr 2017, 15:34

Soundphase:if you make a new update, remove Team AiR tag, and make a new professional skin with more standard colors and without people in it lol, then it would be a nice contender to bad bus mojo


but yea maybe in the future, but atm i wont do anything on it!
http://www.facebook.com/dcharged
http://www.soundcloud/dcharged
User avatar
D-Charged (20)
State Outsider
 
Posts: 9
Location: Sweden (se) Helsingborg

Postby The RaveN » 14 Apr 2017, 17:20

But are you gonna update it sound-wise? :? :D
User avatar
The RaveN (18)
State Tourist
 
Posts: 68
Location: United States (us)

Postby Soundphase » 14 Apr 2017, 17:37

D-Charged:
Soundphase:if you make a new update, remove Team AiR tag, and make a new professional skin with more standard colors and without people in it lol, then it would be a nice contender to bad bus mojo


but yea maybe in the future, but atm i wont do anything on it!

that's alright, it's still a decent plugin :)
Soundphase (27)
State Celebrity
 
Posts: 1012
Location: Netherlands (nl)

Postby Traxorz » 20 Apr 2017, 20:09

Is it possible that i could get the source for this?
Traxorz (22)
State Newcomer
 
Posts: 30
Location: Netherlands (nl) Rotterdam



Quick Reply

Register or login to share your passion for the harder styles
   
 


  • Related topics
    Replies
    Views
    Last post

Who is online

Users browsing this forum: No registered users and 13 guests