Math functions in modules

Hello, my friends

I wonder if there is any manual, book, site...whatever, that explains in mathematical terms, the function behind, for ex, the Exp (Lvl-to-A) Module. I'm asking this because it would be great if I (and averybody) could establish "instantaneously" the DSP concepts and definitions with Reaktor modules.

Thank you

 

Dream on.

Seriously, understanding the DSP concepts and definitions in all Reaktor modules is years of study. Definitely worth the effort though.

If you already know the math and DSP, and just want to know the internals of particular modules, work it out.

ex. exp(lvl to A): clearly it's an exponential of some sort... so EITHER
*look inside the core version of the same function "dB2AF" (you could work this out if you already understand math and DSP).
OR
*google conversion of decibels to linear amplitude... you should find an explanation easy enough.

 

There are those who look at things the way they are, and ask why... I dream of things that never were, and ask why not?
J.F.Kennedy

Right! But...why not explain where the 1.12202 comes from? It's just as easy as add to the explanation the following: Two signals that differ by one decibel have an amplitude ratio of 1.12202.

That's exactly what I wanted to avoid. I spent two hours searching for a good explanation instead of enjoying working with Reaktor. Obviously, this is part of the work needed to work properly with complex softwares but it could be easier if the documentation was a little bit more specific.

Just a thought.

 

A JFK quote doesn't change the fact that you cannot absorb years worth of knowledge 'instantaneously'

To use the software, you need to know what the modules do, not how they do it. The manual explains clearly what they do, it is very specific in that respect.

If you look at the core dB2AF version, and you have the math knowledge you don't need to google to find out what's happening. It seems like what you need is DSP and/or math training. You shouldn't expect NI to provide that as part of Reaktor's documentation. There is a huge amount of study material available on the web on DSP and math, this thread is a great place to start.

 

Well, not instantaneously obviously. Hence the " ".

I'm not sure but I have the idea that, to program in Reaktor, one must know how the modules work. Otherwise, it would be more or less easy.

Math is not a problem to me, but DSP is something I must read, for sure. I'm not expecting NI to provide math/DSP lessons in the documentation but if some pointers on those directions were included in the manual, on each module (or as a chapter with just the basics), would help a lot those who don't have the knowledge.

Anyway, this is just my oppinion. I see I cannot expect nothing towards that direction so, let it be and move on... reading DSP books and searching the web.

Thanks

 

You need to know how they work in the sense that you need to know what their response will be to a particular input. You don't need to know how they work in the sense that you don't need to know how they process the input to achieve the output.

e.g. to divide one float by another, you need to know that the divide module divides the upper input by the lower input, and that the answer will be produced at the output port. You do not need to know how the module internally calculates the answer.
Just as well, because if you did 'need' to know, and took that logic to it's ultimate conclusion, you would have to move beyond Reaktor to the underlying system, then the cpu, ultimately, you'd need to brush up your quantum mechanics somewhat before getting to the answer. Thankfully, you can just assume that if you plug your two numbers in the inputs, magic happens and you get the correct answer at the output.
It's the same with the dB2AF, all you need to know is that if you have a signal level value in dB scale, and you want to convert it to a linear scale, you send it through the dB2AF module, and it gives you the correct answer.