Need help building an exponential attack

Or does anyone know where to find a macro with that feature? I basically need a normal ADSR which has an exponential attack.
Would be grateful if anyone can help!

 

Put an multipier at the output and multiply by it self.

 

Really? Can you post a macro?

Edit.. I tried but it doesn't do anything?

 

http://www.nireaktor.com/reaktor-tutorials/building-an-envelope-in-reaktor-core-part-i/

exponential envelopes are covered somewhere in this series.

 

Thanks salamanderanagram! You think you can post a macro? That would be sweet!

 

this is square not exponential!

 

true

 

If anyone can post a working solution here would be great!

 

you can use the macro Multi-BP env

you will find it here:
Macro/Building Blocks/Envelopes

ciao herw

 

er... square IS exponential where the exponent is 2.

@OP
Sometimes this is absolutely fine for an exponential control curve. Just use a control with range of 0..1, square it and then scale the result to your needs, and you have a very cheap exponential curve...
It's no use if you want the user to be able to adjust the curve though. in that case, there's a core macro for approximation of power function x^y. Again, stick to an input range between 0 and 1. That way, the end result is always between 0 and 1. After the exponentiation has been applied, scale the final result to your application.

 

squared envelopes sound pretty good in my ears and is VERY simple to create. Take the output of an envelope and connect it to both inputs of a multiplyer.

 

Thanks herw! That did the trick!

 

? ? ? sorry but you are absolutely wrong:

square is a power function y=x^2 with constant exponent (f.i. 2) and variable basis. Other examples are x^3, x^(-4) etc.
Exponential functions like y=2^x (or 1.5^x, 3^(-x) etc) have a constant basis and variable exponent.

ciao herw

 

Squaring is not a form of "exponential function". But it is a process of exponentiation, and therefor is exponential. This is the first time in the thread that anyone has mentioned exponential functions.
It is an important distinction though, and I suppose we should have first asked what the op meant by exponential attack.
In my experience with envelopes, I've never heard one described as having a "power attack"...
I think we're missing a trick there in marketing terms
Col

 

The main point is that an attack curve is an exponential function because in analog hardware synths the envelope is produced by charging and discharging of a capacitor.
If you are interested how to realize such curves then you can follow here:
Hüllkurven (envelopes)

ciao herw

 

I think the marketing term we're looking for is "snappy". I always wondered how NI got away with those linear envelopes. The breakpoint envelope is nice but very complicated to use.

 

So if the aim is to build an accurate model of an envelope from an analogue synth, then using an exponential function is important.

 

After trying the breakpoint envelope a bit I find it a little bit hard to work with.. especially with the release time. Someone experienced should try and build an ADSR with switchable functions log/lin/exp

 

true - i am working since two years. Calculations are clear for me but i must invest time to build.
The universal formula is

Have a look to the movie: http://www.reaktor-forum.de/download/file.php?id=2562
Start and stop several times the movie and have a look to the green formula especially for b=0.999!

The advantage of exponential growing is that from step to next step (SampleRateTicks) you mainly need only a fix multiplication (for several envelope phases) independently with which value the phase (attack, decay, release) starts.

ciao herw

 

I've got an idea to manipulate the input times while in that phase. I'm going to work on this next coming week... I'll report here.