Using LUT on SNEX

orange

@Dan-Korneff Thanks.

Faust also has great interpolators.

interpolators - Faust Libraries

(faustlibraries.grame.fr)

griffinboy

@orange

If snex has access to juce, there are many excellent interpolators and there is a lookuptable class even I believe. I used it before with a good result.
I haven't tried accessing it in Snex though.

JulesV

@Dan-Korneff I tried to assign the code that you shared above to snex, but I get tons of errors.

Sorry for the inconvenience, do you have a chance to share the snex code directly?

template <int NV> struct LUTSNEX
{
	SNEX_NODE(LUTSNEX);
	
	
	// Constants for the Lookup Table --------------------------------------------------------------
     int tableSize = 200;       // The size of the lookup table array
     float xMin = -1.0f;        // Minimum input value (start of the input range)
     float xMax = 1.0f;         // Maximum input value (end of the input range)
     float xRange = xMax - xMin;  // Total range of input values
    
	
	 // Lookup Table Data ---------------------------------------------------------------------------
	 float lookupTable[tableSize] = {
	 -0.356490f, -0.357427f, -0.358393f, -0.359388f, -0.360412f, -0.361464f, -0.362544f, -0.363653f, -0.364790f, -0.365954f, 
	 -0.367146f, -0.368366f, -0.369613f, -0.370887f, -0.372188f, -0.373515f, -0.374870f, -0.376251f, -0.377658f, -0.379091f, 
	 -0.380551f, -0.382031f, -0.383487f, -0.384854f, -0.386069f, -0.387067f, -0.387784f, -0.388167f, -0.388236f, -0.388037f, 
	 -0.387613f, -0.387010f, -0.386272f, -0.385446f, -0.384574f, -0.383703f, -0.382877f, -0.382142f, -0.381541f, -0.381120f, 
	 -0.380925f, -0.380999f, -0.381387f, -0.382135f, -0.383288f, -0.384889f, -0.386944f, -0.389391f, -0.392161f, -0.395186f, 
	 -0.398397f, -0.401724f, -0.405101f, -0.408457f, -0.411725f, -0.414835f, -0.417719f, -0.420308f, -0.422534f, -0.424328f, 
	 -0.425621f, -0.426345f, -0.426464f, -0.425989f, -0.424939f, -0.423329f, -0.421179f, -0.418503f, -0.415321f, -0.411648f, 
	 -0.407502f, -0.402900f, -0.397859f, -0.392396f, -0.386529f, -0.380274f, -0.373649f, -0.366634f, -0.359082f, -0.350815f, 
	 -0.341658f, -0.331433f, -0.319963f, -0.307071f, -0.292583f, -0.276464f, -0.258907f, -0.240127f, -0.220341f, -0.199765f, 
	 -0.178613f, -0.157102f, -0.135447f, -0.113863f, -0.092567f, -0.071774f, -0.051659f, -0.032214f, -0.013375f, 0.004921f, 
	 0.022736f, 0.040134f, 0.057177f, 0.073928f, 0.090451f, 0.106808f, 0.123062f, 0.139276f, 0.155513f, 0.171836f, 
	 0.188308f, 0.204992f, 0.221950f, 0.239246f, 0.256943f, 0.275078f, 0.293520f, 0.312062f, 0.330498f, 0.348620f, 
	 0.366224f, 0.383101f, 0.399047f, 0.413853f, 0.427315f, 0.439226f, 0.449378f, 0.457566f, 0.463622f, 0.467645f, 
	 0.469852f, 0.470458f, 0.469678f, 0.467729f, 0.464826f, 0.461186f, 0.457023f, 0.452555f, 0.447995f, 0.443562f, 
	 0.439469f, 0.435933f, 0.433164f, 0.431234f, 0.430080f, 0.429634f, 0.429825f, 0.430586f, 0.431847f, 0.433540f, 
	 0.435594f, 0.437942f, 0.440514f, 0.443241f, 0.446055f, 0.448886f, 0.451665f, 0.454324f, 0.456793f, 0.459003f, 
	 0.460886f, 0.462372f, 0.463417f, 0.464063f, 0.464372f, 0.464405f, 0.464224f, 0.463891f, 0.463468f, 0.463016f, 
	 0.462596f, 0.462272f, 0.462104f, 0.462155f, 0.462484f, 0.463117f, 0.464024f, 0.465173f, 0.466533f, 0.468071f, 
	 0.469755f, 0.471553f, 0.473432f, 0.475361f, 0.477308f, 0.479239f, 0.481124f, 0.482929f, 0.484623f, 0.486174f, 
	 0.487549f, 0.488716f, 0.489644f, 0.490299f, 0.490650f, 0.490665f, 0.490311f, 0.489556f, 0.488369f, 0.486716f
	    };
	

	
	    // Define the amount and types of external data slots you want to use
	    int NumTables = 0;
	    int NumSliderPacks = 0;
	    int NumAudioFiles = 0;
	    int NumFilters = 0;
	    int NumDisplayBuffers = 0;
	
	
	
	// Initialise the processing specs here
	void prepare(PrepareSpecs ps)
	{
		
	}
	
	// Reset the processing pipeline here
	void reset()
	{
		
	}
	
	// Process the signal here
	template <typename ProcessDataType> void process(ProcessDataType& data)
	{
		for (int i = 0; i < data.size(); ++i)
		    {
		        float input = data[i];
		// If the input sample is below a small threshold, treat it as silence to save a little processing
		        if (Math.abs(input) < 1e-6)  // threshold value for silence
		        {
		            data[i] = 0.0f;
		            continue;
		        }
		
		// Normalize the input sample from its range [min, max] to [0, 1] so you can index the table
		        float normalizedX = (input - (-1.0f)) / (1.0f - (-1.0f));  
		
		        // Scale the normalized input to the lookup table index range
		        float scaledIndex = normalizedX * (tableSize - 1);
		
		        // Get the integer part of the index
		        int index = static_cast<int>(scaledIndex);
		
		        // Get fractional part for interpolation
		        float frac = scaledIndex - index;
		
		        // Interpolate between two points in the table
		        float y0 = lookupTable[index];
		        float y1 = lookupTable[index + 1];
		        float output = y0 + frac * (y1 - y0);  // Linear interpolation
		
		        //If the output is very small set it to zero to prevent denormal numbers
		        if (Math.abs(output) < 1e-10) output = 0.0f;
		
		        data[i] = output;  // Process with lookup table
		    }
	}
	
	// Process the signal as frame here
	template <int C> void processFrame(span<float, C>& data)
	{
	}
	
	// Process the MIDI events here
	void handleHiseEvent(HiseEvent& e)
	{
		
	}
	
	// Use this function to setup the external data
	void setExternalData(const ExternalData& d, int index)
	{
		
	}
	
	// Set the parameters here
	template <int P> void setParameter(double v)
	{
		
	}
};

Dan Korneff

@JulesV post the errors.

JulesV

@Dan-Korneff said in Using LUT on SNEX:

@JulesV post the errors.

I posted the code that I tried above

Dan Korneff

@JulesV but the code doesn't show me the errors you're getting.

ustk

@JulesV @Dan-Korneff The thing with SNEX is that the errors in the console show up in a lazy fashion... Everything gets on one line, and often line numbers aren't showing anything but the last line number of the scripting window.

I tend to compile at every new line I write so I know when and where I introduced a mistake

JulesV

@Dan-Korneff said in Using LUT on SNEX:

@JulesV but the code doesn't show me the errors you're getting.

Have you copy pasted my code above?

First of all, for xRange I get can't assign static constant to a dynamic expression error.

This time when I move this variable to the void process function, I get the Expected initializer for non templated membererror for the lookupTable array.

@ustk I am new to snex, have you tried the code above?

ustk

@JulesV said in Using LUT on SNEX:

First of all, for xRange I get can't assign static constant to a dynamic expression error.

Probably here:

float xRange = xMax - xMin;

This shouldn't be the case as the dependent variables are not static const
I get this all the time too, forcing to initialise using literals... @Christoph-Hart ?

Fortune

@Christoph-Hart Can we ask you for a proper working example snippet for the use of lut in snex, which is asked in this topic?

Dan Korneff

Sorry for the late reply. I just realized that I'm using a 3rd party node for my LUT test.
Here's the full project you guys can take a look:
https://hub.korneffaudio.com/index.php/s/xPcEwQicPH63Fkd

Using LUT on SNEX

interpolators - Faust Libraries

15

1.7k

11.8k

103.0k