I have some questions about use of filters.

I'm basically doing a simple subtractive-synthesizer. I have an
oscillator putting out a spectrum I'm happy with. Then I put it
through a resonant lowpass and put an envelope on the amplitude of the
signal and an envelope on the lowpass.

My problem is this. First note that I didn't do a spectrum analysis on
the signal. What I did was create an instrument that loads a very
brief soundfile, looks at one cycle and just loops that. The soundfile
contains a very short bit of sound produced by a cello sampler. This
was easier than doing a spectrum analysis. Eventually I'll do it with
many samples of different instruments and pitches. I don't know--it
just seems easier than doing a spectrum analysis, particularly because
I can automate it.

But, the lowpass can only subtract, so it takes luster off the
spectrum. What I would like to do is boost the high frequencies in the
source waveform so that after putting it through a lowpass, I can vary
it from (1) brighter than the original source, all the way to (2)
darker than the original source.

The way it is right now, it can only get darker than the source.

What would be some good ways to brighten the source sound? Put it
though atone maybe?

Dennis


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

I'm not 100% sure what you're trying to get sonically. My off-the-cuff
comment would be, if you want the tone to be brighter, why are you running
it through a lowpass filter? But perhaps I can be more constructive.

The reson opcode is a bandpass filter. You might try running two filters in
parallel, each with its own envelope and other controls -- one lowpass, and
the other a reson. Then you could blend the outputs of the two filters as
needed. If the lowpass is mostly open while the reson is set to a high
center frequency, you'll end up with a tone that's brighter than the
original, because reson will pick out the upper overtones. With this type of
configuration, you'll always have the lows in the tone, because they'll pass
through the lowpass filter, but you can use reson to adjust the spectrum as
a whole to be brighter or darker.

The one concern might be that adding the signals from two filters could
cause some phase artifacts in the composite tone, but you'll need to listen
to the sound to determine whether that's a real issue.

--Jim Aikin


--
View this message in context: http://csound.1045644.n5.nabble.com/filtering-tp4690948p4691335.html
Sent from the Csound - General mailing list archive at Nabble.com.


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Hello,

Maybe add some sort of waveshaping to get harmonic distortion?

P




On 11 August 2011 21:29, Dennis Raddle  wrote:
> I have some questions about use of filters.
>
> I'm basically doing a simple subtractive-synthesizer. I have an
> oscillator putting out a spectrum I'm happy with. Then I put it
> through a resonant lowpass and put an envelope on the amplitude of the
> signal and an envelope on the lowpass.
>
> My problem is this. First note that I didn't do a spectrum analysis on
> the signal. What I did was create an instrument that loads a very
> brief soundfile, looks at one cycle and just loops that. The soundfile
> contains a very short bit of sound produced by a cello sampler. This
> was easier than doing a spectrum analysis. Eventually I'll do it with
> many samples of different instruments and pitches. I don't know--it
> just seems easier than doing a spectrum analysis, particularly because
> I can automate it.
>
> But, the lowpass can only subtract, so it takes luster off the
> spectrum. What I would like to do is boost the high frequencies in the
> source waveform so that after putting it through a lowpass, I can vary
> it from (1) brighter than the original source, all the way to (2)
> darker than the original source.
>
> The way it is right now, it can only get darker than the source.
>
> What would be some good ways to brighten the source sound? Put it
> though atone maybe?
>
> Dennis
>
>
> Send bugs reports to the Sourceforge bug tracker
>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Maybe you can do something like this (in sudo code)

if p5 > 0
   use highpass filter
   amp = amp * (p5)
if p5 < 0
   use lowpass filter

Then you could replace p5 with your midi mod wheel value or something.

Jeremiah

On 08/11/2011 03:29 PM, Dennis Raddle wrote:
> I have some questions about use of filters.
>
> I'm basically doing a simple subtractive-synthesizer. I have an
> oscillator putting out a spectrum I'm happy with. Then I put it
> through a resonant lowpass and put an envelope on the amplitude of the
> signal and an envelope on the lowpass.
>
> My problem is this. First note that I didn't do a spectrum analysis on
> the signal. What I did was create an instrument that loads a very
> brief soundfile, looks at one cycle and just loops that. The soundfile
> contains a very short bit of sound produced by a cello sampler. This
> was easier than doing a spectrum analysis. Eventually I'll do it with
> many samples of different instruments and pitches. I don't know--it
> just seems easier than doing a spectrum analysis, particularly because
> I can automate it.
>
> But, the lowpass can only subtract, so it takes luster off the
> spectrum. What I would like to do is boost the high frequencies in the
> source waveform so that after putting it through a lowpass, I can vary
> it from (1) brighter than the original source, all the way to (2)
> darker than the original source.
>
> The way it is right now, it can only get darker than the source.
>
> What would be some good ways to brighten the source sound? Put it
> though atone maybe?
>
> Dennis
>
>
> Send bugs reports to the Sourceforge bug tracker
>              https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>



Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

Are you familiar with analog subtractive synths like the original
Moog? Starting with a bright signal, then putting it through a
resonant low pass with an envelope on the cutoff frequency is a simple
way to get very musical results. It lets one vary the brightness over
time to simulate instrumental attacks of various sorts. Wendy Carlos
did this kind of thing on Switch-On Bach, which I still think is one
of the most impressive bits of synthesized *music* ever.

I'm not sure if you are implying a "resonant filter" is by definition
a bandpass, but that's not true. There are resonant low pass filters
as well. Using those, and controlling Q, you can get very musical
results. At a high Q, a resonant low pass starts to function more like
a band pass.

Yes, I was concerned about the phase artifacts of putting filters in
parallel, so I was hoping someone could put this on a firm technical
basis. I'm more inclined to put filters in series, like maybe putting
the source signal through atone first (with a high cutoff, so it
basically tips up the entire spectrum).

Dennis

On Thu, Aug 11, 2011 at 3:36 PM, Jim Aikin  wrote:
> I'm not 100% sure what you're trying to get sonically. My off-the-cuff
> comment would be, if you want the tone to be brighter, why are you running
> it through a lowpass filter? But perhaps I can be more constructive.
>
> The reson opcode is a bandpass filter. You might try running two filters in
> parallel, each with its own envelope and other controls -- one lowpass, and
> the other a reson. Then you could blend the outputs of the two filters as
> needed. If the lowpass is mostly open while the reson is set to a high
> center frequency, you'll end up with a tone that's brighter than the
> original, because reson will pick out the upper overtones. With this type of
> configuration, you'll always have the lows in the tone, because they'll pass
> through the lowpass filter, but you can use reson to adjust the spectrum as
> a whole to be brighter or darker.
>
> The one concern might be that adding the signals from two filters could
> cause some phase artifacts in the composite tone, but you'll need to listen
> to the sound to determine whether that's a real issue.
>
> --Jim Aikin
>
>
> --
> View this message in context: http://csound.1045644.n5.nabble.com/filtering-tp4690948p4691335.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
> Send bugs reports to the Sourceforge bug tracker
>            https://sourceforge.net/tracker/?group_id=81968&atid=564599
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

> Are you familiar with analog subtractive synths like the original
> Moog? Starting with a bright signal, then putting it through a
> resonant low pass with an envelope on the cutoff frequency is a simple
> way to get very musical results. It lets one vary the brightness over
> time to simulate instrumental attacks of various sorts. Wendy Carlos
> did this kind of thing on Switch-On Bach, which I still think is one
> of the most impressive bits of synthesized *music* ever.

Yeah, my first synth was a Serge Modular, and not to name-drop or anything,
but I've had a few conversations with Wendy over the years.

> I'm not sure if you are implying a "resonant filter" is by definition
> a bandpass, but that's not true. 

No, I wasn't trying to imply that.

--JA

--
View this message in context: http://csound.1045644.n5.nabble.com/filtering-tp4690948p4691633.html
Sent from the Csound - General mailing list archive at Nabble.com.


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"

On Thu, Aug 11, 2011 at 5:45 PM, Jim Aikin  wrote:
>> Are you familiar with analog subtractive synths like the original
>> Moog? >
> Yeah, my first synth was a Serge Modular, and not to name-drop or anything,
> but I've had a few conversations with Wendy over the years.
>

Okay, well that's why I wanted to use a resonant low pass with a
cutoff-freq envelope, because it's simple and in my opinion incredibly
musical. Maybe something else would work too, but at the moment I want
to focus on the musical aspects of my project rather than the Csound
programming, so I want to go with something that works right out of
the gate.

Denis


Send bugs reports to the Sourceforge bug tracker
            https://sourceforge.net/tracker/?group_id=81968&atid=564599
Discussions of bugs and features can be posted here
To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"