MSF: How to setup Rate modulation of LFO module that it is playable like normal OSC (without ripples)?

[] Peter:H []

It's all in the headline. Experimenting with the math module I discovered that the LFO seem to create cleaner output than the "normal" Oscillator. Cleaner in a sense that the "ripples" are missing that are added to normal oscillators to prevent aliasing. You can check it with the Oscilloscope yourself - Compare a Saw or Square output of a "normal" Oscillator and that of an LFO. See my screenshots below.

These ripples might be a cause why setting up FM between two Oscillator modules using the GenCustomMode never is "as clean" as when using the dedicated FM module.

Therefore I tried to figure out how to set up a LFO module so that it follows keys. I was not too successful.

What I have tried:

• "Frequency" does not represent the real "frequency" but a linear interpolation between 20 Hz and 20Khz mapped to range of 0% to 100% according to the manual.
• "Note" does not seem to work either: But I was able to come pretty close to it, but with rather unintuitive settings, see patch below.

Hm, I'm curious. Has anybody done this? What is the correct setup?
I wonder - Why is there no "Follow Note Frequency" in the LFO module like it is in the per-voice LFO modulator.

Code: Select all

$eNqtmFtz2jgUgN-5FRrlNS22fJ+x3ckFdjMNhY1pm31U7AN4IiTWltOwv37HMgZCqOWdFl5s6TtH56LLkcNPr2uGXqAoc8EjbH40MAKeiiznywhXcvHBx5-iQThJRMWzMU2lKLYlSJnzZYkuImxidHF1ewsbuYqw8dEO1M8nvu9idDHOmYRiLBgTPyaCixoxjIYJXMsjFkYXf7A8g3m+hrrXNNTP9DzfUX3iibJyTp+aoZK1EHJ1XQBtxnunrAFGr5sCysann0LjAv6pgKfbDmZCc27q+g0doNVAdIClA2wd4OgAVwPobNSZqLNQZ6DOPk-T72v6g65+kX1fAbAOZJbLdHUNPOti6klZFdCBJFUpad41bef5+qlTxTdgIs3ldrpYnKNGWS5FkeT-QoRN03YuXds9bh3RchvhwDcvPcPF6M+-vm5Kut4wtSMYGN0-JBuWSwmKNiNs+sQ-bSZquU7o6-14es1E+tyMZxGMvrWbTWD6nm-jOEyqpxmrljnf7S9x2Lwmb-eZs9oSYJBKyBqJCBuXJo7DaSU3lTTRiNMnBnVK8LBtJSetjWTZPqBryp9vRMWlGrR++7rJqITJLMIYjSvGvtB6r7ofT5Gyqny+4yoy+3aMHlXD37WKOKx1GMOfeHVRUAkR-mB8NEzbNYjhBEFgWufdjcPJ9FZJoIOJh83XNi0chw-GUWdZpEaEP8O2dnXYSsdhuaIbQMl8NJs9jJLRfBeM4Vsb9w37h-Ic8y6Bg6OcPv5KOsmli3934qZlmjMmylRsoM7g+9aj-Lm6-J1zY3gUt18xjhyml908Tiupns-Z3ETpjmfwakSYKCfI73GiX-If40F4JSVNn03UnueW5wWB69q2a3k4Dkf8BZjYQBzORM4lSiP8wcRozOiyvI6waZsYzbcbiPBEZCoA6nQOHDcgruv4geGq1LfCb2SNU9lXZUFgEtP3PeJ7O20kCEzfMQI-cMhB2xsbXmvVdQQOBg93ru2dJAd30OjLt3YdNZFrkRa29LC1h209bO9hRw87e9jVw+4e9vSwt4d9PezHg5ox43BewFowkfxkE1KQYkkfljSs1Ye1Gtbuw9oN6-RhnYZ1+7Buw3p9WK9h-T5sHd+mYB7xF3QLKVXT3bYCy-Yd4gUBwU2zUqIOHdcjGD0AA1qq9Uq8+m9hdChEsHbs-aDxILwfT010D3xZn0n1jn5CK4JoCUtL2FrC0RKulvC0hN9NPFCeibWa85SXC1GsqcwFP14jvgpiC7YipK8I2YtYfUWsvYjdV6ReDJ9hW6aUQQ9vDuhBjPQXq31q69i+Q57wJwrI-1RwbMEpOjioakv6LqS9GTx2QO0NowNpCn-P7kCai7FBOpDD1dg0O7Dd8u8gvggJ08XiPaHu881W8a7vRnBZCMagKFGS8yWDmxXlXN2sFHCVyvwF6qiCRDeCMbopYV4AqIKsvqU0NfUNlbAUxfZOwvqOLwSa0bL8IYoswnUub6qiAH5GQ1M4HSlIpKqHx0BlVUCphqirzLYAOOGGO81xeDVrrFTn25HVtRPFBj2osr4+-gbhBCQUJYnD73kmV5M4TERVpIBmBSzUF5Ld3apW1SLDVmgQDs99fSnjwX+kN1Ot

[MeldaProduction]

Hehe the ripples are caused by antialiasing, it's what makes the oscillators sound so damn awesome! If you don't believe me, try setting the quality to low, or disable antialiasing in any oscillators . It will sound bad, really bad, like in most other instruments

[ystmrk]

Thank you Peter:H & MeldaProduction !
Off the topic, these posts solved my problem.

I tried antialiasing on/off and noticed a big difference in the harmonics of the saw-wave when lower over-sampling (x1).

[] Peter:H []

Hehe the ripples are caused by antialiasing, it's what makes the oscillators sound so damn awesome! If you don't believe me, try setting the quality to low, or disable antialiasing in any oscillators . It will sound bad, really bad, like in most other instruments

Yes, I know, thanks Melda. In the mean time I found the "enable aliasing" toggle in the Oscillator ... now I feel rippled off (aka dumb)

Still - Why is there no Option in the LFO modules to make them follow note frequency? It's different too the per-voice modulator LFOs? And LFOs and Oscillators are so close in feature but not the same... why not build a super-lfo-oscilltor-everything-in-one-module thing

[dakkra]

Those ripples look like Gibbs effect to me, which implies to me that the anti-aliasing is actually harmonic reconstruction (summing) of the original waveform sample. Is this correct, or at least similar in concept? Also it's possible to alias the oscillator at very high frequencies, even with anti-aliasing enabled. It looks like (when using a saw wave oscillator) that the oscillator has a minimum of 5 harmonics (3 with square wave). So at D8 (4698.636 Hz), the top harmonics start to fold back. Oversampling fixes this of course, but wouldn't it be better if the oscillator would just reduce to one harmonic? I get that the wave shape would no longer be a saw (but a sine instead), but at least there wouldn't be nasty fold back harmonics. Food for thought

[] Peter:H []

Those ripples look like Gibbs effect to me, which implies to me that the anti-aliasing is actually harmonic reconstruction (summing) of the original waveform sample. Is this correct, or at least similar in concept? Also it's possible to alias the oscillator at very high frequencies, even with anti-aliasing enabled. It looks like (when using a saw wave oscillator) that the oscillator has a minimum of 5 harmonics (3 with square wave). So at D8 (4698.636 Hz), the top harmonics start to fold back. Oversampling fixes this of course, but wouldn't it be better if the oscillator would just reduce to one harmonic? I get that the wave shape would no longer be a saw (but a sine instead), but at least there wouldn't be nasty fold back harmonics. Food for thought

Don't know the Gibbs effect ... the Bee Gees?
A mathematically correct saw tooth will introduce very high frequency content at the edges, i.e. if for instance in a ramp up the waveform drops back to zero in about 0 samples length.
The Ripples are there to smoothen the edges.
Read about it here: http://www.martin-finke.de/blog/article ... scillator/

[dakkra]

Yes that's quite literally what I discerned. Gibbs phenomenon is the result of a harmonic sum (adding sine waves) https://en.wikipedia.org/wiki/Gibbs_phenomenon]

Screenshot_1.png

^From the page you linked

This implies that the oscillator is "re-synthesizing" the waveform with harmonic summing (further supported by the existence of the harmonic editor).

An animated example:
https://www.desmos.com/calculator/zdkghkqmgo

[] Peter:H []

Back to the original question: How to setup Rate modulation of a LFO module that it is playable like normal OSC?
The LFO *module* has no "sync to note" feature.

[Whywhy]

Before the lfo per voices had keytracking option, you had a preset for keytracking.
Copy and paste this preset should works. ( from the lfo per voices to the lfo module)

[] Peter:H []

Before the lfo per voices had keytracking option, you had a preset for keytracking.
Copy and paste this preset should works. ( from the lfo per voices to the lfo module)

Many thanks WhyWhy! Found it
Pretty funny modulation depth value: 41.09%. how could I have found that value myself!?

Here's a sample patch I did with it. I used a LFO module following frequency which is input to Phase-modulate a Oscillator. Then there's another LFO module doing little FM to the first LFO module.
(Don't play to high notes with it. Ocatve 0 and 1 give best results in my ears)

Code: Select all

$eNrVW9ty4rgWfc9XqJx5myZI8r0KmEpCSOdM6NCBdGceFRDgiZEYW4SQh-PtpyTZxsZgk+7OVB1SlcTykrS1tLUvkmj98boIwQuN4oCztoHOoAEoG-NJwGZtYyWmDc-4o3PS6g-5ik16ZCx4tImpEAGbxeC0bSADnJ53u3Qp5m0Dnlm++njY8xwDnPaCUNCox8OQr-uccQmBUGN8x3SxaYDT6zCY0FGwoPItguqDXNez1Tv+RMJ4RJ50V8MF52J+EVGi+ys1pgFXr8uIxnpMB0G9iP6zomy8qcD0ScBQ3XtYB6htAdcBzDqAVQew6wBODaBOxjoR6ySsE7BOPrfmvVfz3q96zyff55SGFZBBIMbzC8omVRiplKuIVkCGq1iQoEptR8HiqbKJbzTk40Bs7qbTfairSSB4NAzeaNtAyLc+eaaVL70i8aZt+B765ELHAJ+-PixjsliGyiJAA9wGi0DQSK3H2-vhMgyEoKomahvIR85uMVbQb6mN8ZHnepbRaQ1XT4NwNQtYYlY6Lf04LJqXIQ3pWNCJftk20CdkpNA4-QdcEPZ8yVdMqDry6WE5IYL2B23DAL1VGH4h0sLcxeMgDIngEUAG6JP4+YapYe2+NsCjKv9LNthpyRZh84CIpzcsEAEJB3MSU11yNxbkhcZtoyGfHlgQc3bFyFOYvM+XXPIwJMuYTvSrbzxcaWPoItPotPp33dMhXQSCMxqD7SAzo2sbQDckdc-otO5hisIGiKMxbBuXq1jwhRxyHI1R2zgXgoyf5fPD8pxNunwtmTWanVaz0J3uffC9v69fy7QKvaFyb3tbH3zvd1rDOVlSoDWW0TiWDXqWaYDh6GowuL8aXo0U-y8kXClNVdWL9GcF71aD297dzvzLf+9WmrMMo5QAHakEERFS0DMTIoh937cdJ5k9+SY3JVsCEfT3M5j5Jjnqe3SIYbzDMHhVvsiGjm9ZGd+y+04r1oQX2P3VlOL9lFq7lOJ3UYrPHMu0bOy4juV-NKVoP6XQQg5y-mVKrymjsictWp9P8jqL9+nsNWWXfT4xwN1KLFfihk2olB-lFNmqZb0fsGCxWiS2q79Sap0b2C8YR05RrH2KcmAcOKc99eO4nBPGpNPGHzEoHdtekDgYy2kpFeYciFcraq9tOGfY8wxw+lWqG7bsn5cN4KJYeZcGjxGpgc6QLckbbZa0bXweAKcgVTPzwmWVL3n3k5zDfzzS18NP5k-4+qt-ViQM3mi0nZ+sKMeFWUNFp3UdkeW801rygIm8qwVC8XLL18M5Dad6kZ252HcgsrCFlQku15IqfIZsD7nYs7GHKlC257iWhT1kelvUY9swzyCyTWS70Pftg-XNM8syPc9FpmfjLSqTekDi2EmQnost27MdF-pqjpMxn0eUxKBPXoNF8Ebv+TpOls05I+HmjUbI3HHi0DbAkK+iMVUL73vAJnydKBBhTM5nRKdKIQM267QGSqZeSGbxRdtAljS5u260eQhm7cKaxcbPB3BLCjLAgJLnWKvanK8vecijbzLASEfV3A7rF9rxLg3Jpn+xVcOkoGDekDZNmVradSs0Cxv7q1AEF4RN+lTQKFYNXEY8jvkLjfp8IifCmlw0+VjkXugpkePlMwN0o813KrTpQemjSvl8x0Imxhi7rl-MHuAnWE4cVOFtwJ51slBKECDcnx-oIY44D59IlIuIG3tzh2wU6XzhPVr4f6Z65TF1Wk+ETWBqkSdSa9QmiWeavmP7PlKbJFNKJ09kLBk-c1QiOEtSSBe6vmW7jgUdOylGSTmCrgFOb+lU3E2nsZz6hswVZcp3ulA6MwjYbMDZLClAUm2nZBUKA5wuCeMRWZAk-Ti9pzFnhI2T7OY+mM2zdlWz0rMNiVhFRKS7Mn76cR0pSrxhMjNRXNz27mL1GwKtv194tCChnEASkYVUc7W87gmbURU2ZOV6gcnVABBogOGGjecRZ8Gb6liV3VI2E3MDfNOpBUz+uyDj59VSkWhhz3aR+i3FuWIvNORL2uy0hsGMkbAc88l+koal2ZPzKcVXg0BHDQKWBqEeAXqHBEm3SHWL-+1userW-Ne6LfNt6j9xp3XJmYh4GNKo8FBckjl5cJU8zUJrytxm4FIBBNcRl3qU7F0a4PrhRltJSQzOmJHlt3z8nG0KFHhLPUW0aa6pMMBwSRPHnqqtttTqSdcc0GhMmYgrlkmfvG61fmfA3WgjbRNZ0jxHCVyGQepdag7HyhDZBhi-6n1by7RcDzvQc42CrWwWaziuayIPIcf30rqe5-vQw7bjeShfFxqJUX9EOhNT1sY206WJDKDNoe9giG0TOdYhe62dgKQJmNrIq7p6dpNxNTM+jjYz3+XMZISq5LCoC7vKgfYph-IFbQNCt+d2e71KhRmShYzCDurLIDHMBriLAh1DSwO9R3tcu6A8FxswDaJYgGUq6w9Y2xFZqr9bKSrIQbvkYJA4mK2MBf-1jnVVR5MKwID0h1uiUImmD2DmWnVZwQreZcXaZcWo0CF04fYgrNKhfWx8DmbzxpLE8Ucy0A8YmGZ7MFUUWLsU2DsUoJ+j4Aj9uOXrhJB-RTv65PVIbuxdbtzSotkJrA4zJXn6aaa2Qd1HEpTvpYIdd5cdL5+jmTpHk03DXDCblqG2cRmSWG3dpGU4HwYnTCrbUeZM713Ig4Prh5sBYTTUTieJHWTyJ+mc87UeG0z5lH3FKZ2yaI+5RruUPhH2DBhZ0AKZV6-iXicU7zRLutuM2c5v56VPdzZYJ2WD80bBW+7Mgrc7C34NdRfBTBfmti8OknjYnItgQfe6OtPx9LGyiz7Q6SWDSwPcM4wrddUvxQbwSJoOK1fNWu0lieJeklTm+L5oso6RbX9VYUA5SEKFnRVUWLZWYcHi3JPc525i2yms3lxlU73fPlvyOdecLZ+97bMjn1GuOVcWmLkWPFng5JrwVRXsGeByFUWUCXniS8VNt20MOYlo9F+9X6yyl63B2LXgltZWaPvOL7A6e6yM7H+vEtg-aWdumKDRkofq3AYW8o0z2-vBPD1bUTAN-C0sr5ds17ePjc5v3sB8oHfPCzjyv64unJ6DPv81sLti2vXQo+CN1KBZ3d--epr-2V33o-nAuvjn5fNwdH33tbH+j0duffhs--37zHWm18O36XIAz-9Ezpf1gPLG55vH6OFpRt2Rl7TUXQ-W7XaVKUSlIBfhmhkdBSKk4GklhHR2NbM6XMtrBzXTin7Vut43SaWy4mI-soudnFqlECr9xqCRpuzHNVyV4R7sBe7pBe2zVaXoHJkfN529iNK3vR7thyc072HfG2v9sBdIhlHlA8wSr1be6A9FROSWYsH2j6JgGVIRF+x9lwshJzFn9EerBPZRllQdqNRO068L2vYtw5ESYX8At+6-yZ-BaGB3L-3hfPz37zeZNfS6XrUFK2VjyK5hMpc9HB-CyZ1owNWWcf3eheW7HxfHFSSpUtpSLoacn2XmiCBO7a0fzZQNrY9jqihKFVVOiSq3vACXZCwPZouD3ptT-iKT1+eTlbrsFesUCEhvoy8mNcBQ0CWIdU5OI9DQ-QMvF48gbFcP2y0VxZ3W+VgEL1RHhmmcmC+7YVMuzw-3vErnQeWo8rxtOI4oZfrcr3CNrNlp6VLcaX0PJmIur1ypU1mQP6ZK5U4hzazS4cPP4079HzsnLX3LDIH0cq9veqbvI9t3fHk1Kt3mB3cvNIqCCX1ggYgf24ZvgCEjyxG-joKJvriyfU4uDmV7yg1kHNzrxdtd3gy-u8VchL+m89o82MX+OqbcT93ohXjwzDC-65yOXh46a54yxnCOm6sv39KDD33MkUJSsFkPNjOwVQ+2MrBdD7YzsFMPdjKwWw92M7BXD-Y6JxKDOq1RRBc85MMDN7MUSGHxMVisseYxWFNjrWOwlsbax2BtjXWOwToa6x6DdTXWOwYr+dUbiFfsBXTpWJ9DY1f+mAa4pyFVV1-hmW2pHwPkbjIjR97Jq+kma79zog9N7-XlP7h7JVRfALzPXQBEOxcAIfSxK3PC6nuAUiTC4qn0NDKQgnn10rdg7mFy4e9eXfjTh6sn+nR1e-pYGs2JPgitQ1i1CLsW4dQi3FqEV424J2zCF2iXrjxbnmYrAaZV8LFVcFbFPLaKmVWxjq0iV92fdBOPSUiPGM0Wuq2Gj68mx5RexT+2yx38TgP4nQ3kJdiFnmybSr+VUAVJv9zwWAFKvyRRAdHfXXCtCoj+bg-EFZDtt3sQqoAlBqgC8YULejedlhHqK0naUJXe5e4CgGHAZiHN7pnq5VKM75KgbBRRnRHKkFdHMJdE0BmPNjeCLmTMB+R1vDWPJm1DzmUS-pVb0EFgroGhUMapR2XiTmPVRXqpQN1eKOKaScvyipKWUjnSnNRyENEyNcBqUD8VTJ60mvu+QBZ3Tv4H4T-bsA==