Noise source and its impact on sample and hold.

Discuss John Bowen Synths - Solaris
Post Reply
normanion
Posts: 4
Joined: Sun Jan 29, 2017 9:31 am

Noise source and its impact on sample and hold.

Post by normanion » Tue Apr 18, 2017 11:32 am

Hi All,

At the beginning: I want to tell you that this is my first post at this forum and I am glad to be one of "to be owners" as I see you all as group of emotionally mature people and have not seen examples of rudeness here. And sorry for my bad english, as it is not my native language and I do not use it often.

As I am waiting for my Solaris, I am thinking of new patches. During those fantasies I came at one aspect...

Can anyone tell me what is the source of white noise?

My second question: is it used for S/H?

Why am I asking? If S/H uses white noise from prerecorded waveform, it could be possible to set sampling interval to achieve periodic pattern (to be precise: it's always periodic from prerecorded waveform, but at sampling rate of length of wavetable divided by natural number it could be short enough to be easily heard). Could this be achieved? If assumption of prerecorded white noise is true, is it long enough to guarantee gaussian distribution of value over long period of time?

I know that from perspective of sound designing the only problem that matters could be periodic pattern at certain setting, that can be easily heard, but (i) I would like to know as much as possible about tool I am getting and also (ii) white noise generated from basic physic phenomenon could be used in other applications.

I hope I am not getting too confusing. If anything is not clear, please tell me and I shall try to formulate my thoughts other way.

Thank you in advance!

Best Regards,
Norman

John Bowen
Site Admin
Posts: 1592
Joined: Sat Apr 07, 2007 3:00 am

Re: Noise source and its impact on sample and hold.

Post by John Bowen » Tue Apr 18, 2017 4:19 pm

Hi Norman,

Good question! I would have to have someone look at the code now to determine this (the original coder is no longer around). I suspect it is a 'pseudo-random generator'. I would not expect you to be able to achieve a periodic pattern, though.

Also, maybe a bit of clarification as to what the Solaris has that's labeled 'S&H'. The S/H used in the LFOs is the random, stepped output that you are talking about. The "sampling interval" in this case is just the Rate of the LFO.

There is another S/H provided in the Oscillator Type MM1. This waveshape provides a noise source that is controlled by the keyboard, so you can affect the 'color' by controlling the osc frequency. Using one of the audio oscs in Low Frequency mode does give the same result as using the LFO's S/H, except that the rates are different as you play up and down the keyboard (unless you set the Osc to the No Track setting).
Unfortunately, neither one allows retriggering of the phase for S&H, so you don't get a random pattern across an entire chord always.

normanion
Posts: 4
Joined: Sun Jan 29, 2017 9:31 am

Re: Noise source and its impact on sample and hold.

Post by normanion » Wed Apr 19, 2017 12:00 pm

Hi John,

Thank you for your answers as well as aditional information that is also very usefull.

When you will know the answer, please let us know.

Can you tell us what's the distribution of S&H either in LFO or OSC? Is it gaussian (typical S&H )? Is it uniform? Or other? Sorry for asking such irrelevant questions but my head piles with ideas of using some outputs to control other applications (uniform distribution can be assigned to position cursor in drum sequencing - for example in music domain).

If you have to waste yours coders time for some single persons query, and if it's not any top secret, I can look at parts responsible for noise generation. I'm not a coder and it's not my cup of tea, but as a physicist I had to code during my college years and deal with randomisation in monte carlo method, so maybe I could answer this questions without troubling anyone. Regarding second question, it's easy to check without looking into code if someone has necessary tools. I would use discriminators (not sure their name in English, I'm translating directly word to word from Polish - they are tools that count pulses in voltage range set by user during predetermined time) and oscilloscope for convenience.

BTW: Off topic question, related to some remarks you made in other thread: have you read "Solaris" by Lem?

Best regards,
Norman

John Bowen
Site Admin
Posts: 1592
Joined: Sat Apr 07, 2007 3:00 am

Re: Noise source and its impact on sample and hold.

Post by John Bowen » Wed Apr 19, 2017 12:07 pm

BTW: Off topic question, related to some remarks you made in other thread: have you read "Solaris" by Lem?
I have not read the novel, but I saw Tarkovsky's film version of it in 1976, and it had a profound affect on me, having just lost the love of my life, so the name 'Solaris' has always had a special meaning.

Post Reply

Who is online

Users browsing this forum: No registered users and 3 guests