Optimized Lambert W function

Why is it not available an optimized version of the Lambert W function (or Wright Omega function)?

I suppose that an efficient and accurate approximation has been proposed in one of the last DAFx conference, it would be nice to see it available in the dsp module

Hello! (That’s how we start a message in a new topic, even more when it is about a new feature request :grinning: )

Could you tell me what DAFX article you are talking about? And have you tried to implement yourself what is suggested here?

By the way, can I know why you need this function? I don’t know a lot of applications for this, which is one of the reasons you can’t find it in JUCE yet.


Hi @IvanC and nice to meet you! :smiley:
Sorry for my brutal start, really bad behavior. My fault!

Ok so, the article from DAFx is here:

It is very useful in the Virtual Analog context, to solve transcendental equation (such as BJT or Diode characteristics) :slight_smile:
Here you can find an example of mine:

But i suppose that in the article there is link to a better code!
Let me know what you think and thank you very much! :slight_smile:

1 Like

Hello again!

So I don’t understand your statements, here is the code linked by the DAFX 19 article:


Looks like your code is based on the DAFX19 article implementation already, isn’t it?

1 Like


Yes, it is! I think I implemented only two approximations of the four provided by the author