[Csnd] AM question
| Date | 2014-05-26 19:44 |
| From | Askwazzup |
| Subject | [Csnd] AM question |
aOffset = 0.5 aSine1 poscil 0.3, 40 , 1 aSine2 poscil 0.3, 440, 1 *out (aSine1+aOffset)*aSine2* Ok, i'm trying to visualise how the example above works in csound: *We have a sine wave constantly going along in amplitude from -1 to 1, at a specific frequency; *We then add an offset number, reducing the amplitude to movement between -0.5 to 0.5; *We then multiply the sine wave amplitude ( from -0.5 to 0.5) at various points of the wave depending on the modulating signals frequency. So because the modulating signal is oscilating between -1 and 1, it creates two equally loud (as loud as the carrier)? partials on the carrier wave? Is that right, or am i missing something? Also, from the floss manual: "Ring-modulation is a special-case of AM, without DC-offset (DC-Offset = 0). That means the modulator varies between -1 and +1 like the carrier. The sounding difference to AM is, that RM doesn't contain the carrier frequency. " How does the carrier frequency disappear? I thought it's just being masked by the louder? partials than the fundamental. -- View this message in context: http://csound.1045644.n5.nabble.com/AM-question-tp5735538.html Sent from the Csound - General mailing list archive at Nabble.com. |
| Date | 2014-05-27 01:23 |
| From | Jim Aikin |
| Subject | [Csnd] Re: AM question |
If you add 0.5, you'll get a sine wave with an amplitude that varies between -0.5 and 1.5, not between -0.5 and 0.5. You want to multiply, not add. -- View this message in context: http://csound.1045644.n5.nabble.com/AM-question-tp5735538p5735542.html Sent from the Csound - General mailing list archive at Nabble.com. |
| Date | 2014-05-27 01:44 |
| From | Anthony Palomba |
| Subject | Re: [Csnd] Re: AM question |
Hey Jim, Amplitude modulation is multiplication not addition.http://upload.wikimedia.org/wikipedia/commons/8/8d/Illustration_of_Amplitude_Modulation.png On Mon, May 26, 2014 at 7:23 PM, Jim Aikin <midiguru23@sbcglobal.net> wrote: If you add 0.5, you'll get a sine wave with an amplitude that varies between |
| Date | 2014-05-27 02:48 |
| From | Justin Smith |
| Subject | Re: [Csnd] Re: AM question |
AM is additive / subtractive in the frequency domain, and multiplicative in the amplitude domain.
in this example that was provided: aOffset = 0.5 aSine1 poscil 0.3, 40 , 1 aSine2 poscil 0.3, 440, 1 *out (aSine1+aOffset)*aSine2* We take a 0.3 amplitude sine wave at 40 hz, and a 0.3 amplitude sine wave at 440 hz.
We add 0.5 to the first sine wave, producing a 40 hz sine wave from -0.2 to 0.8. We then multiply the signals, producing the new frequencies of 400hz and 480hz in the range -0.15 to 0.24 as the output.
the addition and subtraction of frequencies in the output is a result of the element by element multiplication of the sample amplitudes, and can be easily shown if you work out an example of say 10 samples or so by hand.
On Mon, May 26, 2014 at 5:44 PM, Anthony Palomba <apalomba@austin.rr.com> wrote:
|
| Date | 2014-05-27 02:51 |
| From | Justin Smith |
| Subject | Re: [Csnd] Re: AM question |
correction: the output will be in the range -0.02 to 0.24 On Mon, May 26, 2014 at 6:48 PM, Justin Smith <noisesmith@gmail.com> wrote:
|
| Date | 2014-05-27 04:20 |
| From | David Bowen |
| Subject | Re: [Csnd] Re: AM question |
Excuse me, but I believe there is still an error in your arithmetic. The original 0.3 amplitude sine waye will vary between -0.3 and +0.3. Adding 0.5 to it will make it vary between 0.2 and 0.8, not the -0.2 to 0.8 you claim. Multiplying by the second sine wave will give a variation between -0.24 and +0.24 (max positive for the first sine aligning with max negative to get the minimum, max positive for the first with the most negative to get the minimum.) David Bowen On Mon, May 26, 2014 at 8:51 PM, Justin Smith <noisesmith@gmail.com> wrote:
|
| Date | 2014-05-27 05:29 |
| From | Askwazzup |
| Subject | [Csnd] Re: AM question |
Thank you for replies, though i was talking in 1 and 1 terms for ease of arithmetics, but ok let's use the numbers provided. David Bowen-3 wrote > Excuse me, but I believe there is still an error in your arithmetic. The > original 0.3 amplitude sine waye will vary between -0.3 and +0.3. Adding > 0.5 to it will make it vary between 0.2 and 0.8, not the -0.2 to 0.8 you > claim. Multiplying by the second sine wave will give a variation between > -0.24 and +0.24 (max positive for the first sine aligning with max > negative > to get the minimum, max positive for the first with the most negative to > get the minimum.) > David Bowen Why -0.24 ? Does it multiply both -0.3 and 0.3 by 0.8? Shouldn't it be -0.18 to 0.24 then? -- View this message in context: http://csound.1045644.n5.nabble.com/AM-question-tp5735538p5735547.html Sent from the Csound - General mailing list archive at Nabble.com. |
| Date | 2014-05-27 06:49 |
| From | Justin Smith |
| Subject | Re: [Csnd] Re: AM question |
On Mon, May 26, 2014 at 9:29 PM, Askwazzup <aistiskaikaris@mail.com> wrote: Thank you for replies, though i was talking in 1 and 1 terms for ease of oh yeah, thanks for catching that.
the highest and lowest of the un-biased output should both be multiplied by the largest magnitude signal from the biased one to show the actual full output range.
|
| Date | 2014-05-27 07:37 |
| From | Askwazzup |
| Subject | [Csnd] Re: AM question |
Justin Smith wrote > the highest and lowest of the un-biased output should both be multiplied > by the largest magnitude signal from the biased one to show the actual > full output range. Does this apply only to multiplication, or does it extend to -; /; +? Is the biased signal always on the right side of the un-biased one? -- View this message in context: http://csound.1045644.n5.nabble.com/AM-question-tp5735538p5735549.html Sent from the Csound - General mailing list archive at Nabble.com. |
| Date | 2014-05-27 16:01 |
| From | Justin Smith |
| Subject | Re: [Csnd] Re: AM question |
Unless the two signals are very strongly related in frequency content, every pair of amplitude values between the two will end up being combined. So to find the greatest peaks you would find the two values in the inputs that would combine to produce the greatest magnitude (regardless of whether the combining operation is +, *, -, / or any other). That being said, 440 and 40 are actually very closely related frequencies, and depending on the initial phase you may not see the full potential maxima in the output. On Mon, May 26, 2014 at 11:37 PM, Askwazzup <aistiskaikaris@mail.com> wrote: Justin Smith wrote |
| Date | 2014-05-27 17:21 |
| From | Askwazzup |
| Subject | [Csnd] Re: AM question |
Thank you great master JS -- View this message in context: http://csound.1045644.n5.nabble.com/AM-question-tp5735538p5735552.html Sent from the Csound - General mailing list archive at Nabble.com. |