Hey hey everyone,
so it turns out that the internet has a few more open resources, or last time 
I didn't search properly. So I found a Csound example of "reset inhibit sync" 
and I added one of "phase advance sync".

I've been wondering though: is there a way to perform "overlap sync" in 
Csound?
( https://en.wikipedia.org/wiki/Oscillator_sync#Overlap_Sync )

I thought about using some kind of granular synthesis technique with table 
writing opcodes.

Could anyone suggest ways to improve my versions of reset inhibit or phase 
advance sync?

*** sync.csd ***


-o sync.wav -W

; ==============================================


sr	=	48000
ksmps	=	1
nchnls	=	2
0dbfs	=	1

instr 1 ; Reset inhibit sync
 	istart_ratio = p5
 	istop_ratio = p6
 	ifreq = p4
 	kmod line istart_ratio, p3, istop_ratio
 	khi = floor(kmod) ; shorten the actual value to the next lowest integer harmonic
 		; This removes some aliasing, but introduces stepping
 		; filtering with port reintroduces aliasing
 	aph phasor ifreq
 	aread limit (aph * kmod), 0, khi
 	asound table3 aread, 1, 1, 0, 1
 	aout dcblock asound
 	outs aout*0.9, aout*0.9
endin

instr 2 ; phase reversal sync
 	ifreq = p4
 	istart_ratio = p5
 	istop_ratio = p6
 	kmfreq line ifreq*istart_ratio, p3, ifreq*istop_ratio
 	amod poscil3 1, ifreq, 2
 	asound poscil3 0.7, kmfreq*amod, 1
 	aout dcblock asound
 	outs aout, aout
endin

instr 3 ; Naive sync (this is BAD!)
 	ifreq = p4
 	istart_ratio = p5
 	istop_ratio = p6
 	kmod line istart_ratio, p3, istop_ratio
 	aph phasor ifreq
 	asound table (aph * kmod), 1, 1, 0, 1
 	aout dcblock asound
 	outs aout*0.8, aout*0.8
endin


; ==============================================

f1 0 32768 10 1
;f2 0 32768 -23 "sync.ftable"
f2 0 8192 -27 0 1 4095 1 4096 -1 8191 -1

i1 0 1 220 16 1
i1 1 4 220 32 1
i2 5 1 220 16 1
i2 6 4 220 32 1
i3 10 1 220 16 1
i3 11 4 220 32 1
e


*** sync.csd ***

I hope you enjoy it and even more so that you could improve it. :)
Ta-ta,
----
Ffanci
* Homepage: https://freeshell.de/~silvain
* Twitter:  http://twitter.com/ffanci_silvain
* GitHub:   https://github.com/fsilvain

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

Hi Ffanci,

Regarding Overlap sync, it's certainly possible to write code to do
it. The way I'd go about it would be:

1. Use two counters that track phases for the two oscillators.
2. For the master, advance the phasor as normal and wrap as it hits
the limits for given frequency.
3. Advance the slave phasor as in 2 but for the related frequency.
However, when 1 resets, additional reset the phase of the slave.
4. For the master, you could use it directly with tablei or table3 to
generate the waveform.
5. For the slave, track whenever the phase resets.  When it does,
write the entire grain of the wavetable to a ring buffer.
6. For the slave's audio, read from the ring buffer at a constant rate.

You could probably use an array for the ring buffer, and would have to
make sure it's long enough for the maximum size of the grain.

I'm curious about this, and will see if I can write up an
implementation to demonstrate.

steven


On Tue, Jul 26, 2016 at 12:59 PM, Ffanci Silvain  wrote:
> Hey hey everyone,
> so it turns out that the internet has a few more open resources, or last
> time I didn't search properly. So I found a Csound example of "reset inhibit
> sync" and I added one of "phase advance sync".
>
> I've been wondering though: is there a way to perform "overlap sync" in
> Csound?
> ( https://en.wikipedia.org/wiki/Oscillator_sync#Overlap_Sync )
>
> I thought about using some kind of granular synthesis technique with table
> writing opcodes.
>
> Could anyone suggest ways to improve my versions of reset inhibit or phase
> advance sync?
>
> *** sync.csd ***
> 
> 
> -o sync.wav -W
> 
> ; ==============================================
> 
>
> sr      =       48000
> ksmps   =       1
> nchnls  =       2
> 0dbfs   =       1
>
> instr 1 ; Reset inhibit sync
>         istart_ratio = p5
>         istop_ratio = p6
>         ifreq = p4
>         kmod line istart_ratio, p3, istop_ratio
>         khi = floor(kmod) ; shorten the actual value to the next lowest
> integer harmonic
>                 ; This removes some aliasing, but introduces stepping
>                 ; filtering with port reintroduces aliasing
>         aph phasor ifreq
>         aread limit (aph * kmod), 0, khi
>         asound table3 aread, 1, 1, 0, 1
>         aout dcblock asound
>         outs aout*0.9, aout*0.9
> endin
>
> instr 2 ; phase reversal sync
>         ifreq = p4
>         istart_ratio = p5
>         istop_ratio = p6
>         kmfreq line ifreq*istart_ratio, p3, ifreq*istop_ratio
>         amod poscil3 1, ifreq, 2
>         asound poscil3 0.7, kmfreq*amod, 1
>         aout dcblock asound
>         outs aout, aout
> endin
>
> instr 3 ; Naive sync (this is BAD!)
>         ifreq = p4
>         istart_ratio = p5
>         istop_ratio = p6
>         kmod line istart_ratio, p3, istop_ratio
>         aph phasor ifreq
>         asound table (aph * kmod), 1, 1, 0, 1
>         aout dcblock asound
>         outs aout*0.8, aout*0.8
> endin
>
> 
> ; ==============================================
> 
> f1 0 32768 10 1
> ;f2 0 32768 -23 "sync.ftable"
> f2 0 8192 -27 0 1 4095 1 4096 -1 8191 -1
>
> i1 0 1 220 16 1
> i1 1 4 220 32 1
> i2 5 1 220 16 1
> i2 6 4 220 32 1
> i3 10 1 220 16 1
> i3 11 4 220 32 1
> e
> 
> 
> *** sync.csd ***
>
> I hope you enjoy it and even more so that you could improve it. :)
> Ta-ta,
> ----
> Ffanci
> * Homepage: https://freeshell.de/~silvain
> * Twitter:  http://twitter.com/ffanci_silvain
> * GitHub:   https://github.com/fsilvain
>
> Csound mailing list
> Csound@listserv.heanet.ie
> https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
> Send bugs reports to
>        https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

Hi Ffanci,

I've attached my attempt at an overlap-sync implementation.  It uses a
k-rate while loop to look for trigger points within the 2nd phase
signal.  When it does, it mixes in one grain into the ring buffer.
Both the master and slave oscillators use table3 for reading from the
ftable. One issue with the overlap sync is that the gain is going to
fluctuate with the number of overlaps, so should be accounted for in
some way.  (The code I wrote does not really do that.) This
implementation is a bit simpler than what I wrote in the previous
email in regards to the phase calculations.

Hope that's useful!
steven


On Tue, Jul 26, 2016 at 7:16 PM, Steven Yi  wrote:
> Hi Ffanci,
>
> Regarding Overlap sync, it's certainly possible to write code to do
> it. The way I'd go about it would be:
>
> 1. Use two counters that track phases for the two oscillators.
> 2. For the master, advance the phasor as normal and wrap as it hits
> the limits for given frequency.
> 3. Advance the slave phasor as in 2 but for the related frequency.
> However, when 1 resets, additional reset the phase of the slave.
> 4. For the master, you could use it directly with tablei or table3 to
> generate the waveform.
> 5. For the slave, track whenever the phase resets.  When it does,
> write the entire grain of the wavetable to a ring buffer.
> 6. For the slave's audio, read from the ring buffer at a constant rate.
>
> You could probably use an array for the ring buffer, and would have to
> make sure it's long enough for the maximum size of the grain.
>
> I'm curious about this, and will see if I can write up an
> implementation to demonstrate.
>
> steven
>
>
> On Tue, Jul 26, 2016 at 12:59 PM, Ffanci Silvain  wrote:
>> Hey hey everyone,
>> so it turns out that the internet has a few more open resources, or last
>> time I didn't search properly. So I found a Csound example of "reset inhibit
>> sync" and I added one of "phase advance sync".
>>
>> I've been wondering though: is there a way to perform "overlap sync" in
>> Csound?
>> ( https://en.wikipedia.org/wiki/Oscillator_sync#Overlap_Sync )
>>
>> I thought about using some kind of granular synthesis technique with table
>> writing opcodes.
>>
>> Could anyone suggest ways to improve my versions of reset inhibit or phase
>> advance sync?
>>
>> *** sync.csd ***
>> 
>> 
>> -o sync.wav -W
>> 
>> ; ==============================================
>> 
>>
>> sr      =       48000
>> ksmps   =       1
>> nchnls  =       2
>> 0dbfs   =       1
>>
>> instr 1 ; Reset inhibit sync
>>         istart_ratio = p5
>>         istop_ratio = p6
>>         ifreq = p4
>>         kmod line istart_ratio, p3, istop_ratio
>>         khi = floor(kmod) ; shorten the actual value to the next lowest
>> integer harmonic
>>                 ; This removes some aliasing, but introduces stepping
>>                 ; filtering with port reintroduces aliasing
>>         aph phasor ifreq
>>         aread limit (aph * kmod), 0, khi
>>         asound table3 aread, 1, 1, 0, 1
>>         aout dcblock asound
>>         outs aout*0.9, aout*0.9
>> endin
>>
>> instr 2 ; phase reversal sync
>>         ifreq = p4
>>         istart_ratio = p5
>>         istop_ratio = p6
>>         kmfreq line ifreq*istart_ratio, p3, ifreq*istop_ratio
>>         amod poscil3 1, ifreq, 2
>>         asound poscil3 0.7, kmfreq*amod, 1
>>         aout dcblock asound
>>         outs aout, aout
>> endin
>>
>> instr 3 ; Naive sync (this is BAD!)
>>         ifreq = p4
>>         istart_ratio = p5
>>         istop_ratio = p6
>>         kmod line istart_ratio, p3, istop_ratio
>>         aph phasor ifreq
>>         asound table (aph * kmod), 1, 1, 0, 1
>>         aout dcblock asound
>>         outs aout*0.8, aout*0.8
>> endin
>>
>> 
>> ; ==============================================
>> 
>> f1 0 32768 10 1
>> ;f2 0 32768 -23 "sync.ftable"
>> f2 0 8192 -27 0 1 4095 1 4096 -1 8191 -1
>>
>> i1 0 1 220 16 1
>> i1 1 4 220 32 1
>> i2 5 1 220 16 1
>> i2 6 4 220 32 1
>> i3 10 1 220 16 1
>> i3 11 4 220 32 1
>> e
>> 
>> 
>> *** sync.csd ***
>>
>> I hope you enjoy it and even more so that you could improve it. :)
>> Ta-ta,
>> ----
>> Ffanci
>> * Homepage: https://freeshell.de/~silvain
>> * Twitter:  http://twitter.com/ffanci_silvain
>> * GitHub:   https://github.com/fsilvain
>>
>> Csound mailing list
>> Csound@listserv.heanet.ie
>> https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
>> Send bugs reports to
>>        https://github.com/csound/csound/issues
>> Discussions of bugs and features can be posted here

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

Steven Yi, Jul 27 2016:

> Hi Ffanci,
>
> I've attached my attempt at an overlap-sync implementation.
Hey Steven,
thank you very much. It's been an interesting read. I'll have to compare your description of the steps with the actual code. I know that there are a few things to improve the quality, but I'm wondering how much CPU that would cost. I might be stuck with extra oscillators and volume crossfades. That withstanding your overlap sync sounds really good and smooth and might be the way to go.

Thanks for your code and all the other nice UDOs!
...

Ta-ta,
----
Ffanci
* Homepage: https://freeshell.de/~silvain
* Twitter:  http://twitter.com/ffanci_silvain
* GitHub:   https://github.com/fsilvain

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here

Happy that was useful!  If you have any questions on the code, I'm
happy to explain any part of it.

Cheers!
steven

On Sun, Jul 31, 2016 at 9:44 AM, Ffanci Silvain  wrote:
> Steven Yi, Jul 27 2016:
>
>> Hi Ffanci,
>>
>> I've attached my attempt at an overlap-sync implementation.
>
> Hey Steven,
> thank you very much. It's been an interesting read. I'll have to compare
> your description of the steps with the actual code. I know that there are a
> few things to improve the quality, but I'm wondering how much CPU that would
> cost. I might be stuck with extra oscillators and volume crossfades. That
> withstanding your overlap sync sounds really good and smooth and might be
> the way to go.
>
> Thanks for your code and all the other nice UDOs!
> ...
>
>
> Ta-ta,
> ----
> Ffanci
> * Homepage: https://freeshell.de/~silvain
> * Twitter:  http://twitter.com/ffanci_silvain
> * GitHub:   https://github.com/fsilvain
>
> Csound mailing list
> Csound@listserv.heanet.ie
> https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
> Send bugs reports to
>        https://github.com/csound/csound/issues
> Discussions of bugs and features can be posted here

Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
        https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here