From the original SynthMaker forum (2004–2014), with its members’ names and dates.
PaulRiedel ·
Hi!
I want to add PWM to my oscillator. I found a synth here that uses PWM, but I don't just want to copy & paste it rather than understand how to implement it. Could someone please explain to me how I can add PWM to my OSC?
Thanks!!
- Paul
AngularMomentum ·
There is one pulse oscillator primitive that has a PW input...when you go inside you can check the make up of that osc.
PaulRiedel ·
Hmm.. Thanks, but I seem to be blind - what is it called / Where can I find it?
AngularMomentum ·
Check the OSC section on the right of the screen, then select the pulse primitive with 3 inputs, or just type in pulse in the search box...only thing I do not know if it is a standard OSC...I forgot...but if it isn't there, I am happy to send you an OSM with that OSC. 🙂
AngularMomentum ·
PaulRiedel ·
AngularMomentum wrote: Check the OSC section on the right of the screen, then select the pulse primitive with 3 inputs, or just type in pulse in the search box...only thing I do not know if it is a standard OSC...I forgot...but if it isn't there, I am happy to send you an OSM with that OSC. 🙂
Hi!
Pulse is not a standard it seems... Would you be so kind to send the OSM? ([email removed]) Thanks a lot! 🙂
infuzion ·
AngularMomentum wrote:
Sorry, doesn't work; I used a Scope to check.
Jordimatik ·
First sorry, I don't speak english.
Try to connect only the phase of the first saw oscillator. I maked this and work. Put values: -0.49 to 0.49 in the knob and you will have a PW without dead points and with a perfect square wave in center. Use a Scope for view results.
chironex ·
Or, you could do this; it's more flexible:
http://www.mediafire.com/?9mzjzttx2mx
aliasant ·
chironex wrote: Or, you could do this; it's more flexible:
http://www.mediafire.com/?9mzjzttx2mx
Cant get that link to work. It stalls before page finished loading.
Edit: Forget that. After waiting 5 minutes I finally got to the download link =)
stefan ·
chironex wrote: Or, you could do this; it's more flexible:
http://www.mediafire.com/?9mzjzttx2mx
FYI, this kind of PWM oscillator aliases and introduces a DC offset. It has its uses though!
chironex ·
Sorry, I'm afraid I have no clue what you mean.
Though, I know it sounds better when you substitute the SawHB with a normal Saw.
infuzion ·
chironex wrote: Sorry, I'm afraid I have no clue what you mean.
You can Google "DC offset" but in summery, the total sum of a "single wave cycle" must = 0. Stefan & I have a discussion about this in "Ideas & Requests".
chironex ·
You mean, the integral of the waveform must be zero?
Shifrin ·
chironex wrote: You mean, the integral of the waveform must be zero?
Not quite, if you do a sum of all the points in one cycle of the waveform the total should equal zero - this is the case for waveforms generated using an odd function (rotational symmertry about the origin).
Shif.
theun ·
chironex wrote: You mean, the integral of the waveform must be zero?
Shifrin wrote: Not quite, if you do a sum of all the points in one cycle of the waveform the total should equal zero
????? Of course Chironex is right here!
I am sorry Shif, but I don't see what you want to address here. You state Chironex 's understanding is not fully right, and after having said that you are writing a sum of all the points in one cycle of the waveform. Which is a "definition" of an integral: , pls read a =0 and b=2*pi. So, it looks like you agree with Chironex ?
Btw rotational symmetry is certainly not a necessary condition !
Acrobat ·
theun wrote: Btw rotational symmetry is certainly not a necessary condition !
Sorry, what's the effect of that condition? It can't be right or wrong if you don't specify the effect you want to achieve, and I really like to understand more about PWM... 😕
stefan ·
Acrobat wrote: theun wrote: Btw rotational symmetry is certainly not a necessary condition !
Sorry, what's the effect of that condition? It can't be right or wrong if you don't specify the effect you want to achieve, and I really like to understand more about PWM... 😕
I just think he means that it's not necessary to keep DC away. I'm not sure what it has to do with PWM at all, to be honest, but that might just be me not getting what shif meant 🙂
TomC ·
stefan wrote: Acrobat wrote: theun wrote: Btw rotational symmetry is certainly not a necessary condition !
Sorry, what's the effect of that condition? It can't be right or wrong if you don't specify the effect you want to achieve, and I really like to understand more about PWM... 😕
I just think he means that it's not necessary to keep DC away. I'm not sure what it has to do with PWM at all, to be honest, but that might just be me not getting what shif meant 🙂
You can make PWM without having a zero crossing at pi/2 where the integral is still 0. It's quite interesting to have PWM with more than 2 different pulse widths.
Tom
Shifrin ·
Bit of a brain-storm there guys, excuse me - of course the integral will equal zero 😊
What I meant to say re: odd functions, is that they'll always have zero DC component. Nothing to do with PWM and certainly not a criteria for a function to have zero DC component of course!
Appologies 😀
Shif.
stefan ·
TomC wrote: You can make PWM without having a zero crossing at pi/2 where the integral is still 0.
Hm, yes? I don't see how pi/2 came into the picture. Just by adjusting the phase offset you can have the two zero crossings basically wherever.
TomC wrote: It's quite interesting to have PWM with more than 2 different pulse widths. Tom
Now this I don't get at all. Do you mean by doing other waveforms than a pulse wave? Like for instance:
___ ___ ___ ___
different pulsewidth _ _____ _____ _
Sounds cool indeed (and easy enough to try out by mixing two pulse waves), but I wouldn't say it's a pulse wave with pwm or anything 🙂
theun ·
Hi Acrobat,
I'm sorry for causing confusion ...
Stefan is right however, I was reacting to the post made by Shif directly prior to mine, where Shif states a.o. that using waveforms having rotational symmetry about the origin leads to zero DC offset, which is true of course!
However I wanted to remark that there are also waveforms thinkable that do not have this symmetry but also lead to zero DC offset.
And (maybe I should have added this) PWM is a technique that is using waveforms not having rotational symmetry. Rotational symmetry is exactly what we don't want here, because we want to "play" with the width of the pulse! See also TomC's post (thank you, Tom!).
More on PWM can be found in Gordon Reid's "Synth Secrets" published in SOS. Still valuable after all those years, and on-line available. part 10 Modulation, februari 2000 http://www.soundonsound.com/sos/feb00/articles/synthsecrets.htm part 46, Strings, february 2003 http://www.soundonsound.com/sos/Feb03/articles/synthsecrets46.asp part 47, PWM & string sounds, march 2003 http://www.soundonsound.com/sos/Mar03/articles/synthsecrets47.asp
About rotational symmetry: http://en.wikipedia.org/wiki/Rotational_symmetry http://www.mathsisfun.com/geometry/symmetry-rotational.html
TomC ·
stefan wrote: [...]
I know I'm as important as my neighbor's cat so nobody will care, but I'll refuse to post in this forum anymore until the forum software gets either fixed or moved to a more reliable server.
I tried to post a long and detailed answer with beautiful ASCII graphics, but gave up after dozens of server timeouts/not reachable/not logged in anymore/whatnot messages.
Bye.
Tom
stefan ·
TomC wrote: stefan wrote: [...]
I know I'm as important as my neighbor's cat so nobody will care, but I'll refuse to post in this forum anymore until the forum software gets either fixed or moved to a more reliable server.
I tried to post a long and detailed answer with beautiful ASCII graphics, but gave up after dozens of server timeouts/not reachable/not logged in anymore/whatnot messages.
Bye.
Tom
Indeed, it's highly annoying. 🙁
Wrote something here? To have a post removed or your name hidden, write to [email protected].
SynthMaker