linuxaudio.org

--_b246aed5-9568-4dfe-a4f2-411ded507a6b_

Content-Type: text/plain; charset="iso-8859-1"

Content-Transfer-Encoding: quoted-printable
> > If the user sends a 20khz sine wave into an application's

quoted text
> > "volume" port that's either their mistake=2C or its exactly

> > what they wanted to do.

>=20

> If that is what they want to do they should use the right

> tool=2C wich would be ring modulator in a synth. I'd expect

> synths to use audio rate controls at least for some units=2C

> even if in almost all cases 1/16 would be enough.

>=20

> If a 'normal' audio app with a volume control (e.g. a mixer)

> would actually accept a 20kHz sine wave as a control signal

> then I'd consider it to be of extremely low quality and bad

> design. The same app would happily add any noise present on

> its controls to its signals and I wouldn't trust it for any

> serious work. It's sort the software equivalent of using

> cheap low-quality analog faders.

Why is this a bad design or even low quality? Long before FM=20

synthesis was born Moog allowed this to be done and a whole

industry grew out of the result - he could modulate his oscillators

frequency from his oscillators frequencies - who cares if this is freq

or gain. It is very short sighted to suggest that because you think=20

something should not be done then it is thus a bad thing.
Look at the ARP 2600=2C it allowed just this to be experimented with.

Look at the Korg MS-20=2C look at a load of synths. Then take your

clappers off and consider what might come out of a mixer that=20

allowed you to do the same.
The 1/16 as suitable for synths is also ludicrous. There are a few

heads out there arguing that with the correct algorithms there is

no real reason that analogue should sound so much better than

digital. OK=2C it does now=2C probably because some people keep on=20

arguing that quantisation of parameters 'would be enough' as they

cannot be hacked to have their code to the necessary legwork.
BTW=2C why is a mixer a 'normal' audio app? If it is functioning on a

computer then isn't it mixing other audio apps=2C even  if they are

just capture processes? Are they then not 'normal' or is that just

the way you see them?
Fons=2C does your convolution algorithm cut corners? I doubt it as

it is recognised generally as an amazing peice of work. Why do=20

you then think that cutting corners elsewhere is acceptable?
Regards=2C nick
 		 	   		 =20

_________________________________________________________________

Your E-mail and More On-the-Go. Get Windows Live Hotmail Free.

https://signup.live.com/signup.aspx?id=3D60969=
--_b246aed5-9568-4dfe-a4f2-411ded507a6b_

Content-Type: text/html; charset="iso-8859-1"

Content-Transfer-Encoding: quoted-printable
&gt=3B &gt=3B If the user sends a 20khz sine wave into an application's=

&gt=3B &gt=3B "volume" port that's either their mistake=2C or its exactly&gt=3B &gt=3B what they wanted to do.&gt=3B &gt=3B If that is wha=

t they want to do they should use the right&gt=3B tool=2C wich would be=

 ring modulator in a synth. I'd expect&gt=3B synths to use audio rate c=

ontrols at least for some units=2C&gt=3B even if in almost all cases 1/=

16 would be enough.&gt=3B &gt=3B If a 'normal' audio app with a vol=

ume control (e.g. a mixer)&gt=3B would actually accept a 20kHz sine wav=

e as a control signal&gt=3B then I'd consider it to be of extremely low=

 quality and bad&gt=3B design. The same app would happily add any noise=

 present on&gt=3B its controls to its signals and I wouldn't trust it f=

or any&gt=3B serious work. It's sort the software equivalent of using&gt=3B cheap low-quality analog faders.Why is this a bad design o=

r even low quality? Long before FM synthesis was born Moog allowed this=

 to be done and a wholeindustry grew out of the result - he could modul=

ate his oscillatorsfrequency from his oscillators frequencies - who car=

es if this is freqor gain. It is very short sighted to suggest that bec=

ause you think something should not be done then it is thus a bad thing=

.Look at the ARP 2600=2C it allowed just this to be experimented wi=

th.Look at the Korg MS-20=2C look at a load of synths. Then take yourclappers off and consider what might come out of a mixer that allowed=

 you to do the same.The 1/16 as suitable for synths is also ludicro=

us. There are a fewheads out there arguing that with the correct algori=

thms there isno real reason that analogue should sound so much better t=

handigital. OK=2C it does now=2C probably because some people keep on <=

br>arguing that quantisation of parameters 'would be enough' as theycan=

not be hacked to have their code to the necessary legwork.BTW=2C wh=

y is a mixer a 'normal' audio app? If it is functioning on acomputer th=

en isn't it mixing other audio apps=2C even&nbsp=3B if they arejust cap=

ture processes? Are they then not 'normal' or is that justthe way you s=

ee them?Fons=2C does your convolution algorithm cut corners? I doub=

t it asit is recognised generally as an amazing peice of work. Why do <=

br>you then think that cutting corners elsewhere is acceptable?Rega=

rds=2C nick 		 	   		  Your E-mail and More On-the-Go. =

Get Windows Live Hotmail Free. Sign up now.

=
--_b246aed5-9568-4dfe-a4f2-411ded507a6b_--