New modes for the E352 Cloud Terrarium

Cwejman, Livewire, TipTop Audio, Doepfer etc... Get your euro on!

Moderators: Kent, Joe., analogdigital, infradead, lisa, parasitk, plord

Post Reply
User avatar
MATSmile
Super Deluxe Wiggler
Posts: 1418
Joined: Wed Feb 26, 2014 7:59 pm
Location: In Your Mind

Post by MATSmile » Wed May 31, 2017 1:40 am

Hey Paul,
On E370 2-OP FM is fixed sine or you can assign any waveform to it?

Dogma
Super Deluxe Wiggler
Posts: 4132
Joined: Wed Aug 28, 2013 10:46 pm

Post by Dogma » Wed May 31, 2017 3:53 am

MATSmile wrote:Hey Paul,
On E370 2-OP FM is fixed sine or you can assign any waveform to it?
Why do you keep referencing the e370 in all the e352 threads?????????????????????????????????????????????????????????????????

User avatar
listentoaheartbeat
Super Deluxe Wiggler
Posts: 2571
Joined: Sun Mar 30, 2014 7:42 am
Location: Berlin

Post by listentoaheartbeat » Wed May 31, 2017 3:59 am

listentoaheartbeat wrote:Morphing modulator, sine carrier would be cool, with the second output carrying the modulator like on the e330. Or is it the other way round?
Bumping this out of interest. :tu:

paults
mice nuts
Posts: 4450
Joined: Fri Jan 16, 2009 12:46 am

Post by paults » Wed May 31, 2017 7:58 am

On E370 2-OP FM is fixed sine or you can assign any waveform to it?
Yes, for both E352 and E370, you can assign a sine or a fixed wave from the wavetables. And THEN, you can morph them. There is another thread here where I demo this.

User avatar
MATSmile
Super Deluxe Wiggler
Posts: 1418
Joined: Wed Feb 26, 2014 7:59 pm
Location: In Your Mind

Post by MATSmile » Wed May 31, 2017 11:36 am

paults wrote:
On E370 2-OP FM is fixed sine or you can assign any waveform to it?
Yes, for both E352 and E370, you can assign a sine or a fixed wave from the wavetables. And THEN, you can morph them. There is another thread here where I demo this.
Thanks :tu:

paults
mice nuts
Posts: 4450
Joined: Fri Jan 16, 2009 12:46 am

Post by paults » Wed May 31, 2017 12:24 pm

The assignment is by "dialing in" the wave using Z Morph knob/CV.

In the Cloud Mode, you can use the menu system to pick it directly from the loaded wavetables.

Perhaps in a future SW rev, this 'menu pick' can be for all 3 modes, not just the 1.

User avatar
MATSmile
Super Deluxe Wiggler
Posts: 1418
Joined: Wed Feb 26, 2014 7:59 pm
Location: In Your Mind

Post by MATSmile » Wed May 31, 2017 12:56 pm

I have an idea for noise algo. What about implementing simple decay envelope triggered by Sync input with adjustable time in SW.

Brian Odey 1
Learning to Wiggle
Posts: 18
Joined: Thu May 11, 2017 11:29 am
Location: UK

Post by Brian Odey 1 » Wed May 31, 2017 3:39 pm

paults wrote:Well, I'm glad SOMEBODY is :lol:

Perhaps the person in question has never managed a project of this magnitude: 30,000 lines of C code, multiple, parallel HW developments and responding to beta testers who find things we never saw and suggest new ways of doing things that require extensive re-writes of device drivers so that we can stay within the 96KHz sample rate.



This is not a little VCA or mixer built from random Google schematics :deadbanana:
Think about what you are saying here. 96kHz (not KHz) shows that you are trying to impress rather than actually understanding audio. Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you? So go with 10,000 lines of code and fewer re-writes.

paults
mice nuts
Posts: 4450
Joined: Fri Jan 16, 2009 12:46 am

Post by paults » Wed May 31, 2017 3:53 pm

This is Mike Peake, isn't it?

And you are 100% right, I really don't understand audio. The 2 EE degrees, the 10 patents, the 39 years of design.....all a fraud. Thanks for setting the record straight. I feel so much better now.

User avatar
emeb
Common Wiggler
Posts: 201
Joined: Mon Jul 13, 2009 12:35 pm
Location: Mesa, Arizona

Post by emeb » Wed May 31, 2017 5:23 pm

Brian Odey 1 wrote:Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you?
All snark aside, there are good reasons why we choose to run our oscillators at 96kHz vs lower rates. The main one is that it allows simpler synthesis techniques to be used while minimizing aliasing, but there are also issues of responsiveness to modulation that improve with higher sample rates.

You're right in that classical sampling theory requires only that you sample at 2x the bandwidth of the signal, but there's more going on here than Nyquist dreamt of.

User avatar
gonkulator
Super Deluxe Wiggler
Posts: 1422
Joined: Wed Mar 26, 2014 3:56 pm
Location: Edmonds, WA, USA

Post by gonkulator » Wed May 31, 2017 5:45 pm

paults wrote:This is Mike Peake, isn't it?
:hihi:
Racks
Solaris Site
Stuff for sale
Stuff for sale shown on Modulargrid
WTB: Black panels


Unflappably nonchalant, with some exceptions

User avatar
Pighood
Super Deluxe Wiggler
Posts: 1937
Joined: Fri Sep 14, 2012 8:48 pm

Post by Pighood » Wed May 31, 2017 5:49 pm

Sometimes I miss Walters9515 or whatever his name was.
*phnert*

User avatar
Bob Borries
Veteran Wiggler
Posts: 686
Joined: Sun Jul 11, 2010 11:41 pm
Location: L.A.

Post by Bob Borries » Wed May 31, 2017 6:00 pm

I expect a very low S/N ratio on your digital noise, please.

User avatar
emeb
Common Wiggler
Posts: 201
Joined: Mon Jul 13, 2009 12:35 pm
Location: Mesa, Arizona

Post by emeb » Wed May 31, 2017 10:17 pm

Bob Borries wrote:I expect a very low S/N ratio on your digital noise, please.
But the noise is the signal! AAAAAaaaahhhhhhhh.....

User avatar
2disbetter
Ultra Wiggler
Posts: 930
Joined: Wed Jul 16, 2014 9:40 pm
Location: Germany

Post by 2disbetter » Thu Jun 01, 2017 12:38 am

emeb wrote:
Brian Odey 1 wrote:Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you?
All snark aside, there are good reasons why we choose to run our oscillators at 96kHz vs lower rates. The main one is that it allows simpler synthesis techniques to be used while minimizing aliasing, but there are also issues of responsiveness to modulation that improve with higher sample rates.

You're right in that classical sampling theory requires only that you sample at 2x the bandwidth of the signal, but there's more going on here than Nyquist dreamt of.
This! More is better!

2d

Brian Odey 1
Learning to Wiggle
Posts: 18
Joined: Thu May 11, 2017 11:29 am
Location: UK

Post by Brian Odey 1 » Thu Jun 01, 2017 12:56 am

paults wrote:This is Mike Peake, isn't it?

And you are 100% right, I really don't understand audio. The 2 EE degrees, the 10 patents, the 39 years of design.....all a fraud. Thanks for setting the record straight. I feel so much better now.
Categorically not Mike! I have trawled through all your patents, and whilst they are sort of granted or maintained, they are not entirely relevant; only one really relates to analog audio work. Even that is not that groundbreaking, and quite old. Coin operated machines? I 'spose they make a sound.

Dogma
Super Deluxe Wiggler
Posts: 4132
Joined: Wed Aug 28, 2013 10:46 pm

Post by Dogma » Thu Jun 01, 2017 1:01 am

Brian Odey 1 wrote:
paults wrote:This is Mike Peake, isn't it?

And you are 100% right, I really don't understand audio. The 2 EE degrees, the 10 patents, the 39 years of design.....all a fraud. Thanks for setting the record straight. I feel so much better now.
Categorically not Mike! I have trawled through all your patents, and whilst they are sort of granted or maintained, they are not entirely relevant; only one really relates to analog audio work. Even that is not that groundbreaking, and quite old. Coin operated machines? I 'spose they make a sound.
Hey Brian! Its pretty obvious there is some other motive to your trolliing so have with it!

Or go make some music or something instead of trawling through someones old patents :help: :help: :help:

User avatar
Dragonaut
Veteran Wiggler
Posts: 686
Joined: Thu Dec 29, 2016 12:00 pm
Location: Lake Placid, NY

Post by Dragonaut » Thu Jun 01, 2017 1:05 am

I'd sure like a not-that-groundbreaking patent. As well as my own personal internet stalker.
https://soundcloud.com/acaciabridge (IDM, Ambient, Bass, Downtempo)
www.instagram.com/dr_science_phd (synths, hikes, and views.)

slow_riot
Super Deluxe Wiggler
Posts: 4002
Joined: Tue Jun 30, 2009 8:58 pm
Location: uk

Post by slow_riot » Thu Jun 01, 2017 2:46 am

Brian Odey 1 wrote: Think about what you are saying here. 96kHz (not KHz) shows that you are trying to impress rather than actually understanding audio. Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you? So go with 10,000 lines of code and fewer re-writes.
For one thing, Nyquist is only related to crude detection of simple tones, not signal fidelity, 30Khz sample rate represents a 15Khz sine wave using a single square wave. In engineering terms that is called 100% harmonic distortion.

Aliasing is also a huge concern in digital audio processing, and implicitly related to sample rate. There is very rarely any reason to run at a lower resolution assuming the system can handle it. (One issue with high resolution processing in an audio system is the high current draw. Older criticisms of high sample rates claimed that clock jitter became unmanageable at 192Khz but those claims are now debunked I think).

Dogma
Super Deluxe Wiggler
Posts: 4132
Joined: Wed Aug 28, 2013 10:46 pm

Post by Dogma » Thu Jun 01, 2017 2:53 am

slow_riot wrote:
Brian Odey 1 wrote: Think about what you are saying here. 96kHz (not KHz) shows that you are trying to impress rather than actually understanding audio. Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you? So go with 10,000 lines of code and fewer re-writes.
For one thing, Nyquist is only related to crude detection of simple tones, not signal fidelity, 30Khz sample rate represents a 15Khz sine wave using a single square wave. In engineering terms that is called 100% harmonic distortion.

Aliasing is also a huge concern in digital audio processing, and implicitly related to sample rate. There is very rarely any reason to run at a lower resolution assuming the system can handle it. (One issue with high resolution processing in an audio system is the high current draw. Older criticisms of high sample rates claimed that clock jitter became unmanageable at 192Khz but those claims are now debunked I think).
its ssoooo obvious he knows next to nothing about digital processing.....

User avatar
akrylik
Super Deluxe Wiggler
Posts: 1907
Joined: Sat Feb 18, 2012 9:13 am

Post by akrylik » Thu Jun 01, 2017 3:16 am

slow_riot wrote:
Brian Odey 1 wrote: Think about what you are saying here. 96kHz (not KHz) shows that you are trying to impress rather than actually understanding audio. Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you? So go with 10,000 lines of code and fewer re-writes.
For one thing, Nyquist is only related to crude detection of simple tones, not signal fidelity, 30Khz sample rate represents a 15Khz sine wave using a single square wave. In engineering terms that is called 100% harmonic distortion.

Aliasing is also a huge concern in digital audio processing, and implicitly related to sample rate. There is very rarely any reason to run at a lower resolution assuming the system can handle it. (One issue with high resolution processing in an audio system is the high current draw. Older criticisms of high sample rates claimed that clock jitter became unmanageable at 192Khz but those claims are now debunked I think).
Although the "spirit" of your post is correct the details are incorrect. A 30kHz sampling rate IS able to represent a 15kHz sine wave perfectly . It's the 15kHz square wave (which has lots of upper harmonics) or any other waveform for that matter that cannot be represented with a 30kHz sampling rate.

Brian Odey 1
Learning to Wiggle
Posts: 18
Joined: Thu May 11, 2017 11:29 am
Location: UK

Post by Brian Odey 1 » Thu Jun 01, 2017 5:53 am

akrylik wrote:
slow_riot wrote:
Brian Odey 1 wrote: Think about what you are saying here. 96kHz (not KHz) shows that you are trying to impress rather than actually understanding audio. Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you? So go with 10,000 lines of code and fewer re-writes.
For one thing, Nyquist is only related to crude detection of simple tones, not signal fidelity, 30Khz sample rate represents a 15Khz sine wave using a single square wave. In engineering terms that is called 100% harmonic distortion.

Aliasing is also a huge concern in digital audio processing, and implicitly related to sample rate. There is very rarely any reason to run at a lower resolution assuming the system can handle it. (One issue with high resolution processing in an audio system is the high current draw. Older criticisms of high sample rates claimed that clock jitter became unmanageable at 192Khz but those claims are now debunked I think).
Although the "spirit" of your post is correct the details are incorrect. A 30kHz sampling rate IS able to represent a 15kHz sine wave perfectly . It's the 15kHz square wave (which has lots of upper harmonics) or any other waveform for that matter that cannot be represented with a 30kHz sampling rate.
Quite right akrylik, but like I mentioned, if you can't hear, clearly, above 15kHz, then harmonics can't be detected either - so it becomes pointless and expensive to exceed. Going for 44kHz for a sound source among other sounds, is as high as you need to go. 96 kHz sampling is unnecessary and just a figure for sales brochures. Agreed, there may be those that can tell the difference between 44 and 96 kHz for the final master mix, but ask those same people to pin point a 44 kHz sound from a 96 kHz sound, and they would not be able to do it repetitively.

User avatar
akrylik
Super Deluxe Wiggler
Posts: 1907
Joined: Sat Feb 18, 2012 9:13 am

Post by akrylik » Thu Jun 01, 2017 6:07 am

Brian Odey 1 wrote:
akrylik wrote:
slow_riot wrote:
Brian Odey 1 wrote: Think about what you are saying here. 96kHz (not KHz) shows that you are trying to impress rather than actually understanding audio. Most adult people can't hear above 15kHz in a track flanked by HH's, Cymbals and Snares, so 30 kHz sample rate is fine, and 44kHz is as far as you need to go. With a given composition, can you actually hear the difference between 44 kHz and 96kHz from any one VCO? Well can you? So go with 10,000 lines of code and fewer re-writes.
For one thing, Nyquist is only related to crude detection of simple tones, not signal fidelity, 30Khz sample rate represents a 15Khz sine wave using a single square wave. In engineering terms that is called 100% harmonic distortion.

Aliasing is also a huge concern in digital audio processing, and implicitly related to sample rate. There is very rarely any reason to run at a lower resolution assuming the system can handle it. (One issue with high resolution processing in an audio system is the high current draw. Older criticisms of high sample rates claimed that clock jitter became unmanageable at 192Khz but those claims are now debunked I think).
Although the "spirit" of your post is correct the details are incorrect. A 30kHz sampling rate IS able to represent a 15kHz sine wave perfectly . It's the 15kHz square wave (which has lots of upper harmonics) or any other waveform for that matter that cannot be represented with a 30kHz sampling rate.
Quite right akrylik, but like I mentioned, if you can't hear, clearly, above 15kHz, then harmonics can't be detected either - so it becomes pointless and expensive to exceed. Going for 44kHz for a sound source among other sounds, is as high as you need to go. 96 kHz sampling is unnecessary and just a figure for sales brochures. Agreed, there may be those that can tell the difference between 44 and 96 kHz for the final master mix, but ask those same people to pin point a 44 kHz sound from a 96 kHz sound, and they would not be able to do it repetitively.
It is not pointless unfortunately. Those higher harmonics of a 15kHz square wave (or any non-sine) will alias down into the lower frequencies and you will hear them very well. You are confusing the requirements of the encoding of a sound with the requirements of the synthesis of a sound.

User avatar
Vortico
Wiggling with Experience
Posts: 333
Joined: Mon Nov 05, 2012 1:29 am
Location: Tennessee, USA
Contact:

Post by Vortico » Thu Jun 01, 2017 6:10 am

Supersonic sample rates are nice when driving other oscillators with via FM at audio rates (especially if your modulator is not a sine wave), because waveshaping at high frequencies changes the medium-frequency harmonics of lower frequency waveforms. Many nonlinear analog processors like tube distortion can also push the effects of supersonics into the audible range, which is the reason distortion pedal emulations suck. It's not that you hear the supersonics directly, but you can treat the digital modules more like actual analog circuits when pulling the above tricks.

The real reason though is that the DSP programmer doesn't need to worry about generating a perfect bandlimited signal, which requires lots of CPU, but can instead leave it up to an analog filter with shallower slope on the DAC output.
Last edited by Vortico on Thu Jun 01, 2017 6:18 am, edited 2 times in total.

User avatar
2disbetter
Ultra Wiggler
Posts: 930
Joined: Wed Jul 16, 2014 9:40 pm
Location: Germany

Post by 2disbetter » Thu Jun 01, 2017 6:11 am

Brian Odey 1 wrote: Quite right akrylik, but like I mentioned, if you can't hear, clearly, above 15kHz, then harmonics can't be detected either - so it becomes pointless and expensive to exceed. Going for 44kHz for a sound source among other sounds, is as high as you need to go. 96 kHz sampling is unnecessary and just a figure for sales brochures. Agreed, there may be those that can tell the difference between 44 and 96 kHz for the final master mix, but ask those same people to pin point a 44 kHz sound from a 96 kHz sound, and they would not be able to do it repetitively.
Just incase you missed this:
emeb wrote: All snark aside, there are good reasons why we choose to run our oscillators at 96kHz vs lower rates. The main one is that it allows simpler synthesis techniques to be used while minimizing aliasing, but there are also issues of responsiveness to modulation that improve with higher sample rates.
2d

Post Reply

Return to “Eurorack Modules”