[sdiy] I2C DAC

Scott Gravenhorst music.maker at gte.net
Tue Jun 22 19:57:21 CEST 2010

Bruce Duncan <modcan at sympatico.ca> wrote:
>Thanks guys..great information.
>I wasn't getting the complete picture on the benefits of the audio dac.
>I am not trying to recreate the PPG sound so I will try and get a 
>non-instumentation dac working on this one.
>I am band-limiting synthesized waves by using additive sine technique 
>so no ugly square waves here hopefully.
>I used to have an Ensoniq EPS which was 12bit. Wonder if it used an 
>interpolating dac?

I've used Linear Tech LTC2624 12 bit DAC which is part of an FPGA development board I play
with.  (Spartan-3E Starter Kit)

You can hear it in action:
http://www.fpga.synth.net/pmwiki/uploads/FPGASynth/PolyDaWG8-2.mp3 in a Karplus-Strong
physical model synth.  I do not believe that this is an interpolating DAC.

>> > Would I be correct in saying that these features are more important when
>> > playing back sampled material vs generating signals like a digital OSC?
>> > Oversampling a 12bit sine wave at x256 that is 48kHz and interpolating
>> > between steps and then outputting at the same sample rate 48K does not
>> > create any new data..or does it.
>>I am not sure what you refer to, but I don't know any conversion
>>device in which a signal is oversampled and then immediately
>>- Sigma-delta converters internally oversample (interpolate) the
>>signal in the digital domain to generate a high-speed 1-bit stream
>>which approximates the original signal once low-pass filtered.
>>- Some "traditional" (ie not 1 bit internally) converters might
>>oversample to get more resolution or to make reconstruction filtering
>>easier (the internal oversampling makes the steps finer and easier to
>>filter: you don't need a super steep filter to remove the aliasing).
>>I don't see how there would be a difference between synthesized audio
>>and samples read from a ROM. None of them are meant to contain data
>>above sr/2, so both of them will need a reconstruction filter.
>>However, if you're after the sound of early digital synths (PPG,
>>Prophet VS...) you'd better go with the rawest control/instrumentation
>>DACs - because those early machines did not have reconstruction
>>filters - which would have been tricky to implement I guess on devices
>>like the PPG wave in which sample transposition is achieved by
>>adjusting the clock rate rather than resampling (source:
>>http://www.electricdruid.net/index.php?page=info.wavetableoscs). And
>>no, a SSM2044 doesn't count as a reconstruction filter ;)
>> > Would the sound be any better using an audio DAC vs a voltage out control
>> > type dac given the same sample rate and bit depth?
>>Might depend on what kind of "fidelity" (high fidelity to the
>>"theoretical" waveform, or high fidelity to classic digital synths)
>>you're after!
>>Another thing: more than DAC quality, something that will directly
>>impact the quality of your synthetic waveforms is whether they are
>>band-limited or not in the first place. If you work at 48kHz,
>>(square_output = phase_counter_16bits < 0x8000 ? 0 : 4095) is going to
>>sound ugly no matter what kind of DAC you put behind.
>Synth-diy mailing list
>Synth-diy at dropmix.xs4all.nl

-- ScottG
-- Scott Gravenhorst
-- FPGA MIDI Synthesizer Information: home1.gte.net/res0658s/FPGA_synth/
-- FatMan: home1.gte.net/res0658s/fatman/
-- NonFatMan: home1.gte.net/res0658s/electronics/
-- When the going gets tough, the tough use the command line.

More information about the Synth-diy mailing list