Darker sounding filters?

[the_nihilist]

Anything in the freeware realm that runs at 48db/octave? I like 24db filters, but I could do with a bit darker sound.

[The Chase]

link 2 24db filters to the same control?

[CinningBao]

with the filters in series - in a line..

[aciddose]

most software 24db are just two 12db anyway, since the phase error is exponentiated for every stage, its almost impossible to make a working 48db digital iir integrator chain filter. you can get the same effect using fft or something though, you'll have to deal with it.

[smart]

Not freeware, but Fabfilter Volcano has 48db filters, and Quad Frohmage goes down to 36db which is pretty good too.

[TomBlock]

Not freeware - but almost as functional:

http://www.kvraudio.com/get/1762.html

I disabled automation and saving..

And a word to the ... well, whatever.
A chain of two 12 dB/oct filters is *not* a 24 dB/oct filter. A simple experiment: chain them and watch the frequency response. You will realise that the cut-frequency (the so called "-3dB point") moves towards dc. Just a bit but - we just chained two rather simplish filters. To achieve 48 dB/oct you have to chain four of them so it moves more.
Additionally it can be observed that the frequency slope around the cut frequency loses sharpness, gets softer.

Just compare (100 Hz cut, 44100 Hz sample rate):
http://www.blocc.biz/kvr/48db.png
http://www.blocc.biz/kvr/4_times12db.png

See in the second, chained filter the -3dB point moved almost to 50 Hz.

Greets,

Tom

[The Chase]

i learn something every day

[fritzman]

Doesn't the Obertone filterbank have a 48dB filter? I'm not sure because it's no more in my VSTplugins folder ...


Best wishes, FRitz

[aciddose]

tomblock, i should have replied much earlier, but you do realize what causes the knee to become sharper, dont you?

compare the impulse responce of the two filters you used as example and you'll see.

i fail to see ths difference between:

a += n * (i - a)
b += n * (a - b)
c += n * (b - c)
d += n * (c - d)

and:

a += n * (i - a)
b += n * (a - b)

a2 += n * (b - a2)
b2 += n * (a2 - b2)


can you see a difference?

[Debutante]

I found a free 48db filter with a syncronisable lfo. Very simple, very effective. You can find it at "thespoon.de" Its not pretty but it gets the job done quite nicely.

[fritzman]

Tom

Hi Tom,

sorry, I just have an OT question. On your website it says:

Dipl.-Ing. Thomas Block

Welcome to our website!

Why "Thomas Block" if it's "our" website? Who is "We develop"? Thanx for answering this OT question. Will check out your plugins, have just downloaded them.


Best wishes, FRitz

[TomBlock]

Good evening guys,

at least in germany

Ok, FRitz, as you know the german law ... i have to have my name as central part of the company public appearance to be a "Freiberuflicher Diplom Ingenieur", but i may have helping hands, right

Aciddose,

> but you do realize what causes the knee to become
> sharper, dont you

Yes, i do.

> compare the impulse responce of the two filters you
> used as example and you'll see.
> i fail to see ths difference between:

I don't get your point here.

To put it simple,
The cascade of two linear filters results in a filter that is made up by the convolution of the two single filters.

A convolution in the time domain results in a multiplication in the spectral domain.

Now your cut-point has a gain of -6dB which is 0.5 as linear measure (well, roughly). If you multiply 0.5 times 0.5 (remember, we cascaded two filters with the same cut) the result is (tata!) 0.25. So you cut off point moved.
The same effect is responsible for the more gentle slope. Just mutiply the two spectra pointwise ....

Have a nice sunday,

Tom

[aciddose]

the cutoff point is equal to the point at which phase first crosses 180 degrees from the origin.