I've attached the source code to a C program I found online. The program is from http://www.moshier.net/ellfdoc.html and they include key information for the design of elliptic filters.
Sounds good.
Just wondering, do you have an idea of when work on a Linux version could start?
Have a look at the ellip functions in MATLAB. Butterworth was straightforward, Chebyshev can also be done by cascading 2nd order sections and specifying cutoff frequency and Q for each section, but elliptic filters are not so much. I'm leaving the computations up to the computer. I'm not exactly sure how the coefficients for an elliptic filter are derived. In MATLAB to get the a and b coefficients you go [b, a] = ellip(order, 0.1, atten, 2 * fc / fs). If you're not willing to pay for MATLAB, GNU...
Have a look at the ellip functions in MATLAB. Butterworth was straightforward, Chebyshev can also be done by cascading 2nd order sections and specifying cutoff frequency and Q for each section, but elliptic filters are not so much. I'm leaving the computations up to the computer. I'm not exactly sure how the coefficients for an elliptic filter are derived. In MATLAB to get the a and b coefficients you go [b, a] = ellip(order, 0.1, atten, 2 * fc / fs). If you're not willing to pay for MATLAB, GNU...
Have a look at the ellip functions in MATLAB. Butterworth was straightforward, Chebyshev can also be done by cascading 2nd order sections and specifying cutoff frequency and Q for each section, but elliptic filters are not so much. I'm leaving the computations up to the computer.
I'd also like to see FIR (linear phase, so no phase shift, and can be configured as brick wall) high-pass and low-pass filters based on the modified Blackman function (with a0 = 0.42, a1 = 0.5, a2 = 0.08). Don't use the exact values, use those truncated values, because (as many DSP guides will tell you) the truncated values give a faster roll-off. As well as elliptic (Cauer) high-pass and low-pass filters with adjustable stopband attenuation (dB attenuation from 20 dB to -200 dB is in Gain, filter...
Why I prefer Peace: Logarithmic frequency scale (PulseEffects has a spectrum analyse but the frequency scale is linear, not logarithmic) Ability to configure Parametric EQ as Graphic (PulseEffects is graphic only, and I prefer the higher flexibility of PEQ) Better filters A lot more options Ability to switch between configurations is a big plus ...and many others.