Author |
Message |
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
|
Back to top
|
|
|
EdisonRex
Site Admin
Joined: Mar 07, 2007 Posts: 4579 Location: London, UK
Audio files: 172
|
Posted: Thu Dec 27, 2007 4:36 pm Post subject:
|
|
|
frijitz wrote: | ScottG wrote: | Last night I tried using only the harmonics that are prime numbers and I got what I would call the most interesting evolving sound yet (this method). |
Wow, that is too cool! Like a wind chime in a way.
I wonder if there is something special about the primes, or if another set of harmonics with a similar distribution would also be interesting. I hope Mr. Mathteacher is thinking about this.
Ian |
Yes, there is. _________________ Garret: It's so retro.
EGM: What does retro mean to you?
Parker: Like, old and outdated.
Home,My Studio,and another view |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Thu Dec 27, 2007 6:19 pm Post subject:
|
|
|
Can you explain what the connection is? |
|
Back to top
|
|
|
EdisonRex
Site Admin
Joined: Mar 07, 2007 Posts: 4579 Location: London, UK
Audio files: 172
|
Posted: Mon Dec 31, 2007 4:40 am Post subject:
|
|
|
frijitz wrote: |
Can you explain what the connection is? |
Sorry Ian, I kind of forgot about this over the weekend. Too busy decoding the inside of my mixing console.
There are two subjects - harmonic content (timbre) and melodic content (scales/modes). I have long been curious about the relationship between harmonic periods and prime numbers. The du Sautoy article, although tortuously written, does note that the sum of harmonics follows the prime number staircase rather uncannily. Which implies that the amplitude of a given harmonic follows a prime number. This becomes interesting when multiplying harmonics (hence the bell tones produced by careful choice of harmonic relationship). Somewhere I have read a much more academic discussion of the use of prime number relationships to achieve cymbals using AM (ring mod). I'll try to dig it up if there's interest.
The second subject has a more quirky research history. Paul Dysart has an interesting take on using prime numbers in scales. The Tonnetz stuff comes up a lot, although the relationship of prime numbers to Western musical scales is limited at best. Chris Caldwell has some stuff on it too, although he takes the novel approach of using MIDI notes to index the prime numbers. _________________ Garret: It's so retro.
EGM: What does retro mean to you?
Parker: Like, old and outdated.
Home,My Studio,and another view |
|
Back to top
|
|
|
frijitz
Joined: May 04, 2007 Posts: 1734 Location: NM USA
Audio files: 54
|
Posted: Mon Dec 31, 2007 7:57 am Post subject:
|
|
|
EdisonRex wrote: |
There are two subjects - harmonic content (timbre) and melodic content (scales/modes). I have long been curious about the relationship between harmonic periods and prime numbers. The du Sautoy article, although tortuously written, does note that the sum of harmonics follows the prime number staircase rather uncannily. Which implies that the amplitude of a given harmonic follows a prime number. |
Hmmmm ... I'll need to think about that some more.
I don't see anything related to the Zeta function in Scott's clip. He used prime number-labeled harmonics, as I understand it. N = 1, 3, 5, 7, 11, ....
My WAG is that its unique sound may have to do with how the ear-brain works. Anharmonic sounds are often perceived as having a pitch, which the ear-brain constructs somehow. For example two sinusoids in a 2:3 relation will make you think you are hearing the fundamental. Maybe having the prime-numbered harmonics prevents a pitch perception because there are no small number ratios present. When I listed to Scott's clip I hear a bunch of tones that do not quite blend into a cohesive sound. So that's what I'm thinking. But maybe this impression is just due to having such a sparse harmonic content.
Ian |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Mon Dec 31, 2007 9:43 am Post subject:
|
|
|
frijitz wrote: | I don't see anything related to the Zeta function in Scott's clip. He used prime number-labeled harmonics, as I understand it. N = 1, 3, 5, 7, 11, ....
My WAG is that its unique sound may have to do with how the ear-brain works. Anharmonic sounds are often perceived as having a pitch, which the ear-brain constructs somehow. For example two sinusoids in a 2:3 relation will make you think you are hearing the fundamental. Maybe having the prime-numbered harmonics prevents a pitch perception because there are no small number ratios present. When I listed to Scott's clip I hear a bunch of tones that do not quite blend into a cohesive sound. So that's what I'm thinking. But maybe this impression is just due to having such a sparse harmonic content.
Ian |
Yes, the frequencies chosen were those who's integer multiplier of the fundamental is a prime number, so 1 (for fund.), 2, 3, 5, 7,9,11,13, etc. I did this so that there were no common factors (other than unity) between any of the frequencies used. I, too, am not sure if this fact is the reason for the increased "interesting" quality or if it's simply sparseness of harmonic content. However, if one constructs a signal from several even harmonics (for example), even with sparseness, it's not very interesting in comparison.
Ian already knows this, but I'm nearing completion of the 32 NCO FPGA implementation. Currently, I'm mitigating the fact that the IIR reduces the amplitude of the noise fed into it. As more and more heavy filtering is applied (to get slower changing modulation signals), the amplitude is so severely reduced that the precision is reduced to unacceptable levels. What I mean by precision is that the signal is as if it were fewer bits wide than the 18 bits it started with. I am testing the use of a 22 bit wide noise signal into a 22 bit wide IIR and windowing the 18 bits I need. Hopefully, this will allow me to generate nice full scale 18 bit modulation signals. I noticed the problem yesterday while testing and I heard really low-fi trash coming from something that should most certainly not be low-fi. Oscope revealed large stepping in the output.
In fact, the filter method I am using is a digital implementation of a simple passive RC lowpass circuit frequently used to limit high frequencies from a noise generator. In the analog domain, we can usually just buffer/amplify, however in the digital domain, this doesn't work very well, especially if the filter attenuation is severe. In this case, you can push a bigger signal into the RC circuit (IIR) and as in an analog wold, the output also increases. |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Sat Jan 05, 2008 5:27 pm Post subject:
|
|
|
ScottG wrote: | I've gotten my FPGA 32 NCO sine synth working. - later. |
All I can say is - W W !!
Also - when can I HAVE one?? Something based on THAT as it is now, with controllability would be durned-tootin'-right-killer! |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sat Jan 05, 2008 5:39 pm Post subject:
|
|
|
Rykhaard wrote: | All I can say is - W W !!
Also - when can I HAVE one?? Something based on THAT as it is now, with controllability would be durned-tootin'-right-killer! |
Thanks I still need to finish the noise filter configuration stuff, I sorta cheated for now so this one filter parameter applies to all 32 filters. I also need to work on the patch editor more, but I have a contract to do some VB business report generation junk that I had better get busy on. Once I get that done, I'll finish all this and post it at the FPGA-Synth website and drop a message on here. This project was implemented using a _dev_board_, so the hardware is easily available. I use a Xilinx Spartan-3E Start Kit, one of the least expensive boards with enough on board stuff to make a synth. All I had to add to the board was a 6N138 for MIDI and a resistor-cap dealie for the output of the on board DAC. Oh - this is _only_ a 12 bit DAC BTW. |
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Sat Jan 05, 2008 7:03 pm Post subject:
|
|
|
ScottG wrote: |
Thanks I still need to finish the noise filter configuration stuff, I sorta cheated for now so this one filter parameter applies to all 32 filters. I also need to work on the patch editor more, but I have a contract to do some VB business report generation junk that I had better get busy on. Once I get that done, I'll finish all this and post it at the FPGA-Synth website and drop a message on here. This project was implemented using a _dev_board_, so the hardware is easily available. I use a Xilinx Spartan-3E Start Kit, one of the least expensive boards with enough on board stuff to make a synth. All I had to add to the board was a 6N138 for MIDI and a resistor-cap dealie for the output of the on board DAC. Oh - this is _only_ a 12 bit DAC BTW. |
Whoa, wait a minute! I DO hope that I'm not confusing issues here! But - you mentioned VB! By CHANCE - you're not programming this magnificical device _IN_ VB are you?? (Major hopes - as I've played in VB 6 and .NET 2.0) ??
From the examples of it that I've heard so far, if coding for it wouldn't be too hard to teach myself, I'd be HIGHLY more'n willing to buy the devboard 'kit' to play with it as well, if I could continue what you're working on myself as well, for progging chips that I could then plop into a board for what/whichever other required hardware!
Minorly off-subject: would it be more proper to now move this thread to the new sub-forum that the laddies opened up?
Edit: Thot it might be partly useful to mention as well: I've been a HUGE Additive synthesis fan, since I had bought my 1st of 4, Kawai K5's (and a K5000r) in the early 90's. Been missing that programming structure for synthesis for a long, long time. |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Sat Jan 05, 2008 7:16 pm Post subject:
|
|
|
Rykhaard wrote: | Minorly off-subject: would it be more proper to now move this thread to the new sub-forum that the laddies opened up? |
Somehow I think the laddies most definitely have it in the picture for such a shuffle _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sat Jan 05, 2008 7:18 pm Post subject:
|
|
|
Rykhaard wrote: |
Whoa, wait a minute! I DO hope that I'm not confusing issues here! But - you mentioned VB! By CHANCE - you're not programming this magnificical device _IN_ VB are you?? (Major hopes - as I've played in VB 6 and .NET 2.0) ??
From the examples of it that I've heard so far, if coding for it wouldn't be too hard to teach myself, I'd be HIGHLY more'n willing to buy the devboard 'kit' to play with it as well, if I could continue what you're working on myself as well, for progging chips that I could then plop into a board for what/whichever other required hardware!
Minorly off-subject: would it be more proper to now move this thread to the new sub-forum that the laddies opened up? |
The patch editor is written in VB.NET (2008), but the synth FPGA "hardware" is written in Verilog. I learned Verilog using free web tutorials and PDF manuals. It has unlocked a whole world of digital design that I could never do with a soldering iron (32 sinewave NCOs??).
The Xilinx board I have has a 100 pin (Hirose) connector and several smaller connectors that have access to FPGA pins, so more stuff can be added. FPGAs are based on RAM and must be reloaded at power up. That's what I like about them, one minute they can be a polyphonic string synth, the next a strange demented monosynth. This board (like most I assume) has a "platform" flash for power up that can be reprogrammed to suit your needs. My boards all power up as some kind of synth or another, but can be reprogrammed via USB. |
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Sat Jan 05, 2008 7:20 pm Post subject:
|
|
|
Blue Hell wrote: | Rykhaard wrote: | Minorly off-subject: would it be more proper to now move this thread to the new sub-forum that the laddies opened up? |
Somehow I think the laddies most definitely have it in the picture for such a shuffle |
Well, right then! I do abouts most quickly think, thinely related laddies shouldst be promoted to Top Laddies, at once! |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sat Jan 05, 2008 7:30 pm Post subject:
|
|
|
Rykhaard wrote: | Blue Hell wrote: | Rykhaard wrote: | Minorly off-subject: would it be more proper to now move this thread to the new sub-forum that the laddies opened up? |
Somehow I think the laddies most definitely have it in the picture for such a shuffle |
Well, right then! I do abouts most quickly think, thinely related laddies shouldst be promoted to Top Laddies, at once! |
new sub-forum? where?
EDIT: Oh wait, I found it. (duh) |
|
Back to top
|
|
|
blue hell
Site Admin
Joined: Apr 03, 2004 Posts: 24079 Location: The Netherlands, Enschede
Audio files: 278
G2 patch files: 320
|
Posted: Sat Jan 05, 2008 7:36 pm Post subject:
|
|
|
Need to determine the point where this went from OT to OT .. lookes like a hijacked thread ...
@ Scott : http://electro-music.com/forum/forum-158.html _________________ Jan
also .. could someone please turn down the thermostat a bit.
|
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sat Jan 05, 2008 7:49 pm Post subject:
|
|
|
Blue Hell wrote: | Need to determine the point where this went from OT to OT .. lookes like a hijacked thread ...
|
My plan from the start but now it seems I've been found out and thwarted. |
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Sat Jan 05, 2008 8:01 pm Post subject:
|
|
|
(Rather'n quoting) .... VB .NET 2008. Urg. I'll have to see how far along the path I can free upgrade from .NET 2.0. (Full - received free FROM M-soft, for reviewing videos of theres, in 2005).
Scott - have you been documenting all that's happening with the NCO on your site? If so, I'll start attending it regularly to see what's up and where. My interests in it have increased phenomenally, with your latest posting.
Do you have links avail. up there as to where from you learned Verilog? If I follow / try to learn in your footsteps, I'd like to follow as closely relatedly to this as possible, to learn, before I veer off in my typical standards of modifying. lol
Not rem'bering if I've your site bookmarked, could you let me know what it is, please?
(Don't foresee having many troubles learning .NET 2008 and Verilog. )
Finally - a GREAT amount of thanks to you for all of the work that you've been doing on this! |
|
Back to top
|
|
|
JovianPyx
Joined: Nov 20, 2007 Posts: 1988 Location: West Red Spot, Jupiter
Audio files: 224
|
Posted: Sat Jan 05, 2008 8:12 pm Post subject:
|
|
|
Rykhaard wrote: | (Rather'n quoting) .... VB .NET 2008. Urg. I'll have to see how far along the path I can free upgrade from .NET 2.0. (Full - received free FROM M-soft, for reviewing videos of theres, in 2005).
Scott - have you been documenting all that's happening with the NCO on your site? If so, I'll start attending it regularly to see what's up and where. My interests in it have increased phenomenally, with your latest posting.
Do you have links avail. up there as to where from you learned Verilog? If I follow / try to learn in your footsteps, I'd like to follow as closely relatedly to this as possible, to learn, before I veer off in my typical standards of modifying. lol
Not rem'bering if I've your site bookmarked, could you let me know what it is, please?
(Don't foresee having many troubles learning .NET 2008 and Verilog. )
Finally - a GREAT amount of thanks to you for all of the work that you've been doing on this! |
I should have said I use VB.NET 2008 - Express Edition - the completely legitimately free version that you can download from the Microsoft site. It's actually pretty nice, not limited from what I can tell (at least for standalone programs).
I put stuff up at http://www.fpga.synth.net/ in the wiki section. There are a couple of my synth designs there and some other random articles I did. There isn't anything there yet about this particular synth because it's really not quite finished. I'll put the Verilog code and a description of it and samples there. I also will post new versions there if I find big booboos in the code.
And sorry about the hijack, it wasn't intentional at all - in fact, this thread was the catalyst for putting this design together, it's a synth that produces animated sounds. |
|
Back to top
|
|
|
Rykhaard
Joined: Sep 02, 2007 Posts: 1290 Location: Canada
|
Posted: Sat Jan 05, 2008 8:28 pm Post subject:
|
|
|
I should have said I use VB.NET 2008 - Express Edition - the completely legitimately free version that you can download from the Microsoft site. It's actually pretty nice, not limited from what I can tell (at least for standalone programs).
[/quote]
Whoa! Ex'llent! Compilable?? For some very odd reason, I never WAS able to figure out WHY I wasn't able to compile ANYthing from .NET 2.0. I wrote a couple of pieces of software for Numerology and Astrology that I dearly wished ( / wish) to compile, but couldn't ever get that to happen. If 2008 IS compilable, I'll see about importing my old code, for that purpose as well as learning the latest implementation of it. MANY thanks for mentioning that!
Quote: |
I put stuff up at http://www.fpga.synth.net/ in the wiki section. There are a couple of my synth designs there and some other random articles I did. There isn't anything there yet about this particular synth because it's really not quite finished. I'll put the Verilog code and a description of it and samples there. I also will post new versions there if I find big booboos in the code.
|
Thankee for that info as well. I'll read through all that's there to try coming to more of a grip as to what's where / when and how. (I've been wanting to go the PIC / etc. etc. route for quite some time but haven't. This could also be my starting point.
Shall bookmark as well. |
|
Back to top
|
|
|
|