KVR Audio

lasteno · Post by **lasteno** » Thu May 24, 2018 2:30 pm

andy-cytomic wrote:This entire mutation / randomisation panel is currently only a mockup, but when there is something that is feature complete then I'll make it available as to use as a beta

can't wait for that update. testing the demo sounds amazing¡ just amazing. giving you promotion to your amazing work Cytomic. amazing sound.

Yorrrrrr · Post by **Yorrrrrr** » Tue Jun 12, 2018 4:52 pm

God, these things take forever..

Vertion · Post by **Vertion** » Tue Jun 12, 2018 5:35 pm

Good things to those who wait. After Andy finishes coding his plugins, Andy let's his plugins cook and simmer on the stove for months, then he adds fresh carrots, corn and potatoes, into the mix. After realizing he just cooked his laptop for months, he looks at the melted mess of plastic and vegetables, he takes a picture, converts it to DLL... and voila!

andy-cytomic · Post by **andy-cytomic** » Wed Jun 13, 2018 10:58 am

Silence usually means I'm working hard, which is a good thing

In the last few weeks I've sorted out a breakthrough algorithm to tackle arbitrarily complex circuits with a fixed amount of cpu and small memory overhead, which is very big news indeed. No one has published such a method in any scientific papers I know of. I need to profile the method and compare it to the other methods, it could mean I can keep slew rate limiting in the main drive section op-amp of The Scream, which previously I was going to have to ditch. I also hope to be able to use this new method to tackle the MD mode of a fuzz pedal emulation

handed · Post by **handed** » Wed Jun 13, 2018 11:31 am

Fascinating, best of luck with it...

Liero · Post by **Liero** » Thu Jun 14, 2018 8:41 am

If I ever win a lottery there'll be two things I'll do:

1) Pay a million bucks to hire 10 of the best code slaves for Cytomic, so I can get more distortion pedals faster.

2) Pay another million for the Sonic Charge brothers so they'll release Synplant 2.

((( 3) go back to being a poor musician )))

Vertion · Post by **Vertion** » Fri Jun 15, 2018 8:56 pm

andy-cytomic wrote:I need to profile the method and compare it to the other methods

Corr:
https://arxiv.org/corr/home

Submit a new paper:
https://arxiv.org/corr/submithelp

andy-cytomic · Post by **andy-cytomic** » Thu Jul 05, 2018 2:27 pm

News: cytomic.com is moving to a new scalable hosting service. If all goes well we should be back up and running in around a day. Although the web page is in maintenance mode you can still authorise, but there may be a few hours outage during the transfer.

andy-cytomic · Post by **andy-cytomic** » Thu Jul 05, 2018 5:16 pm

The transfer to the new web page is complete

Architeuthis · Post by **Architeuthis** » Sat Jul 28, 2018 12:49 am

I'm going to restate about The Scream what I said in The Drop thread. Basically using mods I got a setting that was nice for a filtered sawtooth and wondering if there are mods to induce feedback or other craziness.

And I have a quick question.

Are you planning on using this mod scheme for a filter product in the future? Our products might get into a bit of a competition. Because I am working on a filter product that could have tons of tweakability, not like 100 analog components like The Scream. More like a handful of macro parameters, internal clipping, internal bias, internal feedback, internal noise level, internal compression/signal boosting, resonance VS frequency scaling, internal "passive filtering" frequency/resonance (yes my filters will have filters, would you like a filter with your filter? yes please). Our filter models will sound very different, which will be fun.

I expect modding features to show up in the synthesizer you are making. Excited to see what you come up with. I hope I am right in saying we are on the same wavelength. I expect a very reactive/responsive/great sounding synth from you that does things no other synth does (but analog synths have been doing forever).

andy-cytomic · Post by **andy-cytomic** » Sat Jul 28, 2018 9:34 am

Architeuthis wrote:I'm going to restate about The Scream what I said in The Drop thread. Basically using mods I got a setting that was nice for a filtered sawtooth and wondering if there are mods to induce feedback or other craziness.

And I have a quick question.

Are you planning on using this mod scheme for a filter product in the future? Our products might get into a bit of a competition. Because I am working on a filter product that could have tons of tweakability, not like 100 analog components like The Scream. More like a handful of macro parameters, internal clipping, internal bias, internal feedback, internal noise level, internal compression/signal boosting, resonance VS frequency scaling, internal "passive filtering" frequency/resonance (yes my filters will have filters, would you like a filter with your filter? yes please). Our filter models will sound very different, which will be fun.

I expect modding features to show up in the synthesizer you are making. Excited to see what you come up with. I hope I am right in saying we are on the same wavelength. I expect a very reactive/responsive/great sounding synth from you that does things no other synth does (but analog synths have been doing forever).

If you have a look at the drive op-amp in The Scream you will see it's already a macro model, it's a Boyle macro model which offers useful parameters like slew rate in mS, unity gain bandwidth, and open loop gain. Inside are multiple circuit components whose component values are controlled by the macro model parameters to bring about the desired results.

Where it makes sense to offer macro model values to allow for easier control of the circuit components inside I will do so, otherwise I'll just have all the individual components available for editing directly.

As I write new products, or add new functionality to existing ones, I'll see how best I can offer this sort of feature on a case by case basis, but until I've done the work and I'm happy with it I won't make any promises.

As for generating self oscillation with The Scream, just lower the resistance of the switching JFET resistor 'rjfet_off". This will create a low impedance feedback path between the output of the drive and tone sections back to the input of the drive section. Varying the drive knob will push the circuit into self oscillation, and also change the cutoff frequency slightly due to the RC network formed with C104. The internal op-amp stability capacitance can also be varied by changing the unity gain bandwidth parameter "fu" which stands for "Frequency at Unity gain". Check out the preset "Resonizer" in this zip file as an example of where to start:

https://cytomic.com/files/scream-beta-presets-2.zip

sqigls · Post by **sqigls** » Sat Jul 28, 2018 10:02 am

thanks Andy

Architeuthis · Post by **Architeuthis** » Sat Jul 28, 2018 5:40 pm

HAHA. I KNEW IT.

This confirms for me that what I'm doing for my own filter models has similarities to analog modelling. I get a lot of the behaviors I am seeing with The Scream, but I'm not doing any modelling, I am simply stringing together digital components like oscillators, filters, clippers. It's clear to me that each digital component like a macro analog component.

Andy, I think we are achieving similar results but using vastly different methods. I am bad at math, I vaguely know how analog circuits work, yet I am able to recreate the sound and behavior of many analog filters through trial and error, brute force experimentation.

andy-cytomic · Post by **andy-cytomic** » Tue Jul 31, 2018 8:02 am

Architeuthis wrote:HAHA. I KNEW IT.
...
This confirms for me that what I'm doing for my own filter models has similarities to analog modelling. I get a lot of the behaviors I am seeing with The Scream, but I'm not doing any modelling, I am simply stringing together digital components like oscillators, filters, clippers. It's clear to me that each digital component like a macro analog component.

Andy, I think we are achieving similar results but using vastly different methods. I am bad at math, I vaguely know how analog circuits work, yet I am able to recreate the sound and behavior of many analog filters through trial and error, brute force experimentation.

I'd like to point out the major difference between traditional DSP (explicit) and analog modelling DSP (implicit).

In traditional DSP each processing element is a self contained block which you can chain together with other blocks to get the desired result. This means that none of the blocks interact with each other.

In analog modelling DSP you can set things up with buffers to make them into traditional DSP, but typically all the elements interact in complicated ways with each other to get the result.

The easiest example to consider is a one pole low pass filter combined with a diode clipper. In the analog modelled version the low pass filter and the clipper interact with each other to produce a great sounding result which is used in many guitar stomp boxes. Using traditional DSP you just can't get the same result, and people tend to like the sound of the more complicated interaction analog modelling gives.

Here are some plots to show what is going on:

Architeuthis · Post by **Architeuthis** » Tue Jul 31, 2018 3:35 pm

I really want to understand what you're saying here before I respond and show you what I'm doing.

Question 1:

What is the buffer component?

Question 2:

In your bottom example, is that basically just:

Code: Select all

Lowpass -> Soft Clipper -> Output

And your top example is better because it's:

Code: Select all

Lowpass <-> Soft Clipper -> Output

The <-> representing some form of interaction between components. So basically you're saying the filter and clipper interact in the analog-style example, and they are separate and boring in the bottom traditional DSP example?

Question 3:

How does your diode diagram differ from a traditional DSP soft clipper?

Question 4:

How do you integrate a soft clipper into a filter? Is that your secret sauce?

Cytomic "The Scream" stomp box distortion plugin