Support

If you have a problem or need to report a bug please email : support@dsprobotics.com

There are 3 sections to this support area:

DOWNLOADS: access to product manuals, support files and drivers

HELP & INFORMATION: tutorials and example files for learning or finding pre-made modules for your projects

USER FORUMS: meet with other users and exchange ideas, you can also get help and assistance here

NEW REGISTRATIONS - please contact us if you wish to register on the forum

Users are reminded of the forum rules they sign up to which prohibits any activity that violates any laws including posting material covered by copyright

Smothing signal (similar way to vacuum tubes, analog)

DSP related issues, mathematics, processing and techniques

Smothing signal (similar way to vacuum tubes, analog)

Postby mayo » Sun Jun 14, 2020 9:29 am

In analog domain (vacuum tubes, analog chain, transformers) I noticed signal is smoothened - this removes some harshness, digital sound, sharp transients and make transients peaks rounder. There is algorithm to emulate vacuum tube but smooths only very very very little.

Could it be possible to make some real time algorithm to smooth signal? Here are examples, also visual examples:

https://www.mathworks.com/help/signal/e ... thing.html

https://en.wikipedia.org/wiki/Smoothing

http://terpconnect.umd.edu/~toh/spectrum/Smoothing.html
mayo
 
Posts: 36
Joined: Fri Oct 19, 2007 2:53 pm

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby k brown » Sun Jun 14, 2020 4:33 pm

Partly that's because a properly-designed and not overloaded vacuum tube circuit does not 'round' anything; they are very fast and very accurate. The 'rounding and 'smoothing' of old tube gear is mostly due to euphonic distortions caused by sonic limitations of the early input/output transformers used. Other than that, tube gear sounds smoother not because of any 'smoothing' or 'rounding' inherent in tubes, but rather their lack of various non-linear distortions of (mostly earlier) solid state circuitry.

So if you're wanting to model vintage (rather than modern) tube 'smoothness', you're really wanting to model old audio tranformer distortion/saturation.
Last edited by k brown on Mon Jun 15, 2020 10:05 am, edited 1 time in total.
Website for the plugins : http://kbrownsynthplugins.weebly.com/
k brown
 
Posts: 1198
Joined: Tue Aug 16, 2016 7:10 pm
Location: San Francisco, CA USA

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby mayo » Mon Jun 15, 2020 10:02 am

k brown wrote:Partly that's because a properly-designed and not overloaded vacuum tube circuit does not 'round' anything; they are very fast and very accurate. The 'rounding and 'smoothing' of old tube gear is mostly due to euphonic distortions caused by sonic limitations of the early input/outputtransformers used. Other than that, tube gear sounds smoother not because of any 'smoothing' or 'rounding' inherent in tubes, but rather their lack of various non-linear distortions of (mostly earlier) solid state circuitry.

So if you're wanting to model vintage (rather than modern) tube 'smoothness', you're really wanting to model old audio tranformer distortion/saturation.


yes it makes sense, I though it is caused by vacuum tubes because signal is "flying in the air" not going thru wires/cables but electrons are flying in air from anode to catode, but as you said it is probably caused by transformers.
Do you have ideas / schematics how to make/simulate that? (transformers etc.) I have some tube emulation codes but nothing for transformers.

thx
mayo
 
Posts: 36
Joined: Fri Oct 19, 2007 2:53 pm

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby k brown » Mon Jun 15, 2020 10:08 am

Well, the toolbox Analog Toolkit Saturation would probably be a rough stab (for sawtooth waves anyway), but different transformers saturate differently, of course.
Website for the plugins : http://kbrownsynthplugins.weebly.com/
k brown
 
Posts: 1198
Joined: Tue Aug 16, 2016 7:10 pm
Location: San Francisco, CA USA

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby MichaelBenjamin » Mon Jun 15, 2020 2:07 pm

.
Last edited by MichaelBenjamin on Mon Sep 21, 2020 10:50 am, edited 1 time in total.
MichaelBenjamin
 
Posts: 275
Joined: Tue Jul 13, 2010 1:32 pm

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby wlangfor@uoguelph.ca » Sun Jun 21, 2020 2:34 pm

It will take you a long time to really learn how to do it yourself, two year learning curve.
My youtube channel: DSPplug
My Websites: www.dspplug.com KVRaudio flowstone products
User avatar
wlangfor@uoguelph.ca
 
Posts: 912
Joined: Tue Apr 03, 2018 5:50 pm
Location: North Bay, Ontario, Canada

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby mayo » Mon Jun 22, 2020 2:28 pm

wlangfor@uoguelph.ca wrote:It will take you a long time to really learn how to do it yourself, two year learning curve.


that is why Im asking for help / example if anybody more experienced can help?

Because classic lowpass only cut highs not really smooth all signal, transients etc. and make signal simpler/smoother
mayo
 
Posts: 36
Joined: Fri Oct 19, 2007 2:53 pm

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby wlangfor@uoguelph.ca » Mon Jun 22, 2020 2:38 pm

The ways though are much more complicated than that.
My youtube channel: DSPplug
My Websites: www.dspplug.com KVRaudio flowstone products
User avatar
wlangfor@uoguelph.ca
 
Posts: 912
Joined: Tue Apr 03, 2018 5:50 pm
Location: North Bay, Ontario, Canada

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby mayo » Mon Jun 22, 2020 5:57 pm

wlangfor@uoguelph.ca wrote:The ways though are much more complicated than that.


did you tried personally? you was sucesfull?
mayo
 
Posts: 36
Joined: Fri Oct 19, 2007 2:53 pm

Re: Smothing signal (similar way to vacuum tubes, analog)

Postby wlangfor@uoguelph.ca » Mon Jun 22, 2020 6:55 pm

Yes, but what it comes down to ultimately is understanding how a discrete time signal is affected by the 44,100 samples per second or 192,000 samples relates to latency. It's ultimately the speed of that connection that will affect the sound; so the question is how do you affect that; and what does that mean? It really only means something after you consider latency and whether or not VST allows dropping the graphical load onto a GPU.

So, even if you came up with an effect that made a smooth sound, how then would you decide exactly every other component, even if someone helped you. Regardless of whether you feel that something works or doesn't work, it still comes down to the details, because what if it sounded smooth when you started but then sounds gritty after you add the graphics?

Sound implausible? Try this plugin I made:
https://www.kvraudio.com/product/vstplug-governor-by-king-oz-records/details

You see? it sounds a bit terrible, interestingly that gritty effect is achieved by getting the volume and then averaging to a ceiling with mean averaging in a dsp code. it's pretty basic, you see: dry+dry+ceiling+ceiling/4 = new volume level and then you just make sure the new volume level is similar to the original and you've created a tube like effect. But how do you sell it after it's all said and done, and how to make it smooth? That will get you started.

What is smooth? :) EDIT: I wanted to add: plainly; a limiter or destructive effect might change the sound; so trying to find a less destructive substitute might be better, like mean averaging as I'd said with the volume level. After all; why just throw on limiting and saturation and call it a tube. You might look up stereo saturation, I forget what it's called hmm... it's a word that means spacial or something like that; maybe someone will mention it. GL.

But in the end, many things you do in the end, even if not effect will change the sound in sometimes in-explicable ways, and by not knowing it's like a guess. Smoothness could mean a destructive, truncating effect that basically takes away a great deal of data and like RMS averages; or it could mean to saturate and reduce peaks, maybe in a unique way.

Either way, people use unique methods and it's really all guesswork. Because if it wasn't it would be something to technical or long winded to talk about it in a forum post either way.

You could try the demo of this product, it ended up sounding fairly smooth; and that was done by an extreme method of limiting, avoiding distortion from a true peak going over a 0dBFS threshold.

https://www.kvraudio.com/product/dspplug-inter-sample-peak-limiter-by-king-oz-records/details
My youtube channel: DSPplug
My Websites: www.dspplug.com KVRaudio flowstone products
User avatar
wlangfor@uoguelph.ca
 
Posts: 912
Joined: Tue Apr 03, 2018 5:50 pm
Location: North Bay, Ontario, Canada

Next

Return to DSP

Who is online

Users browsing this forum: No registered users and 27 guests