Re: Sega sound boards...

> Ugh...someone please tell me why I'm reading my email/newsgroups
> when I have a big project due ~3pm :-( Anyways...

because you are supposed to? well, it is a reason... :>

> No, quantizing is the "Resolution" of those stair steps (i.e.
>how many different places they can be. There are only a finite number
>of digital values that you can quantize to.

Well, Quantizing isn't the "resolution", but the PROCESS of assigning a
digital value to the individual voltage steps. And that (the resolution)
is limited by the word length (16 bits for CD players).

> When you're not dealing with an oversampled converter, I don't see
> any inherent benefit of oversampling other than to make the design of your
> final, analog low-pass filter after your DAC easier (i.e. lower order.)
> Probably not worth schleping around all the additional samples, as I think
> the stuff in our chips is 2nd or 3rd order Butterworth (i.e. pretty
> straightforward) anyways (No, you don't all have to go sign NDAs, that
> stuff is in our datasheets ;-) )

Ok, so the idea is to put as little in the signal path as possible.
instead of desiging a better filter, you design a better D/A convertor.

> > I though oversampling was just for aiding error correction at playback...
> > in conjunction with that process that scatters the data acoss the disc so
> > a scratch will not wipe out a good chunk of the data (what was this
> > process called again?)
>
> Ahhh, could be. I'm not too familiar with that kind of stuff.
makes sense to me.

> > > Apply a little analog filtering to smooth out the edges, and presumably
> > >you
> > > have a nice signal. Apply too much, and things sound like a dog shitting in a
> > > swimming pool (all muddy).
> >
> > I thought there was just another low-pass (anti-aliasing) filter after
> > the D/A convertor.
>
> There is, but its purpose isn't to prevent aliasing. Aliasing
> can't occur in D -> A conversion. If you go through the math, basically
> that low-pass filter replaces all the "sharp corners" of the stairstep
> analog waveform out of the DAC with sync functions (basically sin(x)/x),
> which, if you sampled the signal corrrectly, will all add up to give you
> the proper (i.e. non-aliased) waveform.

That is just what I thought..
 
> I probably just confused you more, so I should probably just keep
> my mouth shut in the future.....

no, I understand it. Speak UP! I understand a lot of the tech-talk that
goes on in here, but I dont know much about the specifics, like functions
or pinouts of individual chips... so I usually keep my mouth shut as I
probably couldn't contribute anything worthwile. :>

Jeff

-- 
http://idt.net/~mayday19