Page 1 of 1

DDS and the Electronic Music Box - April 2012

Posted: Wed Mar 21, 2012 5:06 pm
by JavaBen
First, I very much enjoyed the article on DDS by Craig Lindlay.

I do have a question, mainly to make sure I'm understanding DDS process.

If, as in the example, M is set to 295,279,002, and the sample frequency is as given at 32Khz, and also as given, I have a wave table with 256 entries (2^8), and I truncate the 32 N bit accumulator word to an address word of 8 bits to address the wave table, then at 2.2Khz output, isn't my increment into the wave taable going to be about 17.something (incrementing the address by 17.something), and at 256/360 degrees, isn't that going to give me a sine wave increment of about 24 degrees for each increment into the wave table?

In other words, aren't I incrementing an actual wave form by about 24 degrees each time?

That doesn't seem to be too accurate, so what am I missing?

Thanks!

Re: DDS and the Electronic Music Box - April 2012

Posted: Mon Apr 09, 2012 2:58 pm
by craiglindley
JavaBen,

How I envision this in my head is that since M is a real fixed point number (comprised of an integer part and a fractional part) which is continually added to the phase accumulator each sample time, the sample address is continually advancing returning a new sample from the wave table providing an accurate representation of the waveform in the table.

Hope this helps