Alternate synthesis method?

Software simulator of the unique Russian synthesizer
Post Reply
GENNA
Posts: 25
Joined: Wed Apr 25, 2018 7:42 pm
Contact:

Alternate synthesis method?

Post by GENNA » Wed Aug 21, 2019 9:50 pm

Hello Alexander,

I apologize if this is not the proper forum (may be better of as a Pixilang thread), but I am interested in your opinion on something.

Firstly, thank you so much for VANS, I may have said it before but this software is really amazing for experimental music. I truly love it!

That being said, I've noticed it's not really possible to get a convincing white/pink/brown noise sound out of it. I can understand why, because the synthesis method is based on a finite number of sine oscillators. So I was wondering, would it be possible (perhaps in Pixilang) to, instead of using oscillators, have a finite number of extremely narrow-band filters with white noise as the source? That way, instead of drawing the amplitude of the oscillators, you're drawing the amplitude of the filters. And I suppose, theoretically, you could have any sound as the source but white noise would provide the most bandwidth.

What I'm not sure about however, is the CPU expense. I'm assuming it would be greatly increased - although, for this purpose I'm more interested in the composition aspect of the program and less the real-time synthesize use.

Anyway, this was just an idea I had. If you have any thoughts I would love the hear them, although I'm sure you're a busy man.

Thanks again!!

John

GENNA
Posts: 25
Joined: Wed Apr 25, 2018 7:42 pm
Contact:

Re: Alternate synthesis method?

Post by GENNA » Thu Aug 22, 2019 2:59 am

It just occurred to me that I never explained my rationale for asking the question in the first place.

IMHO, most additive-based synthesizers produce similar artifacts when creating complex sounds. Again, I attribute this to the fact that it's not easy to synthesize noise from simple sine oscillators. I'm merely wondering if it's possible to achieve more natural sounds, or even more accurate reproductions of existing sounds if building from a noise source instead.

Cheers!

John

User avatar
NightRadio
Site Admin
Posts: 3134
Joined: Fri Jan 23, 2004 12:28 am
Location: Ekaterinburg. Russia
Contact:

Re: Alternate synthesis method?

Post by NightRadio » Mon Aug 26, 2019 9:56 am

would it be possible (perhaps in Pixilang) to, instead of using oscillators, have a finite number of extremely narrow-band filters with white noise as the source?
Yes, but this will require major changes in the ANS engine. We can't just replace the sine oscillators by the filters, because it will slow down the program too much. More tricky algorithms are needed here. I sometimes think about it. Perhaps I will try to do something in the next updates.

GENNA
Posts: 25
Joined: Wed Apr 25, 2018 7:42 pm
Contact:

Re: Alternate synthesis method?

Post by GENNA » Wed Aug 28, 2019 7:49 pm

Thanks for the reply.

Please know that this isn't something that I'm expecting to be implemented in the future. I was mainly curious, knowing that VANS was created in Pixilang, if it was something easy to achieve. Now I understand it's not as straightforward as I thought. Thank you for confirming that.

John

Post Reply