[sdiy] Kawai K3 Waveform ROM's - Update

[Tom Wiltshire]

As I understood it, the idea of running one of the waves at twice the 
frequency of the other was exactly to make sure that it has all the 
even harmonics.

Imagine:

One wave with odd harmonics 1,3,5,7,9,11, etc

One wave with all harmonics 1,2,3,4,5,6,7,8,etc

Now we double the frequency of the second wave, which means that with 
reference to the first wave we now have 2,4,6,8,10,12,14,16,etc

Also, rather than the red wave being the derivative of the green, I 
think the green may be the integral of the red. The reason for thinking 
it's this way round and not the other is the Stilston and Smith paper 
that Aaron Lanterman mentions in his lectures. The red waves look like 
the BLITs from that paper (pg11). The idea of those is that instead of 
storing the waveform per se, you actually store the rate of change of 
the waveform, particularly for the discontinuities which is where the 
aliasing problems come from. You can then keep a running total of the 
current sample output (an integral) to derive the actual waveform.

	"Alias free digital synthesis of classic analog waveforms" Stilson and 
Smith

	www-ccrma.stanford.edu/~stilti/papers/blit.pdf

Still, it's only a clue. I don't get exactly what they're up to in the 
K3.

Tom



On 14 Mar 2007, at 08:14, Johannes Öberg wrote:

> But one of the waves is supposed to run at 2x the frequency (which
> wouldn't make much of a difference in the odd/even department I
> guess). Now, I still don't really understand what the patent talks
> about but it... Notice however that the two waves are out of phase.
>
> I think the patent was written the way it was so that competitors
> wouldn't figure out the method from it.
>
> FWIW, it looks to me that the red wave is the derivate of the green
> (albeit attenuated).
>
> /Johannes
>
> On 3/13/07, Plutoniq9 • <plutonique9 at hotmail.com> wrote:
>> The patent though talks about one wave having all harmonics present & 
>> the
>> second only having the odd harmonics, the FFT graphs show that they 
>> both
>> have the same odd/even harmonics.
>>
>> Ryan
>
>
>
++++ Electric Druid ++++
Web Design & Development
http://www.electricdruid.net