Noise Bass

Cabbage recipes
1

After a long pause I managed to finish a new instrument. It is based on the concept of ‘noise bass’ presented in quite a lot of YouTube videos. I think it does not need so much explanations, and a save myself the documentation. The first line of controls determines the tonal sound of the bass. The second line controls the noise. In particular, ‘Level’ is the fraction of the noise in the output signal, ‘Concentration’ determines how much the amplitude is (anti-)synchronised with the tonal signal, 0 resulting in an uncorrelated flat noise and 10 in short bursts of noise at the zero crossings of the tonal part. The last row adds a low pass (‘O’: order), stereo width, a parallel fifth and/or octave, a sine subbass one octave below. The preset system is as always a bit fragile.

Noise_Bass.zip (7.2 KB)

GitLab: Noise_Bass · d44b13e434a2f43bbee88a4e1518b6a870af0705 · Reiner Zorn / RZ Cabbage Recipes · GitLab

One deficiency of Csound was really annoying me: Odd order filters are really difficult to construct. So I ended up combing clfilt (for the even orders) with zdf_1pole to generate the orders 1, 3, 5… Is there any simpler way to do this?

2 Likes
2

Some nice tones here :clap: I’ve no real experience with building odd order filters, but can’t you simple cascade as many one pole filter as you like to make up the order you want?

3

Thanks for the positive response.

That’s the idea of the tonex opcode. The problem I have with this is that the coefficients are calculated as described for tone. The formulae set the half-point frequency mathematically exact on spot and the 6dB/octave is as good as possible. But the consequence is that the filter never fully opens. So if you set hpf=22050Hz for sr=41000, there is still significant reduction around 10kHz. And pushing hpf further doesn’t make sense because of the cosine in the first formula on the tone manual page.

I like it more the way it is in Robin Schmidt’s Engineering Filters plugin (and many other filter plugins): The curve is forced to zero at sr/2 and when the cutoff frequency is set to sr/2 the filter becomes ‘transparent’.