Xref: utzoo comp.dsp:114 sci.electronics:7984 rec.music.synth:9231 Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!porter!schabtac From: schabtac@porter.uucp (Adam Schabtach) Newsgroups: comp.dsp,sci.electronics,rec.music.synth Subject: Re: Digital Mixer Using Burr-Brown 200 Khz ADC Message-ID: <4513@ncar.ucar.edu> Date: 28 Sep 89 16:23:22 GMT References: <9231@pyr.gatech.EDU> Sender: news@ncar.ucar.edu Reply-To: schabtac@porter.UCAR.EDU (Adam Schabtach) Organization: not until I move into a new apartment. Lines: 53 In article <9231@pyr.gatech.EDU> byron@pyr.gatech.edu.gatech.edu (Byron A Jeff) writes: >[This is a repost. I got absolutly no response to the first posting.] > It's possible you got no response because your first posting was a bit vague. For instance, it would help if you told us which chip you have in mind -- that is, give us the part number (BB makes quite a few ADCs :-). I'm assuming you're talking about the chip that was featured on the cover of a recent electronics trade journal. >How difficult a proposition would it be to use this BB ADC to build a >multi-channel digital mixer? I haven't read the specs on it yet (I will >today at the Tech library) but if it's dual 200Khz ADC then it should >be possible to sample 8 audio channels at CD (44.1 Khz) or DAT (48 Khz) >quality. So how feasible is something like this: > > [ diagram showing ADC->DSP->DAC deleted ] It depends on what you mean by "difficult" -- i.e. how are you with digital prototyping? The circuit you describe would require a lot of work to get together, debugged, and running. (I think I can generalize that way and say that almost anyone would consider it "a lot" of work, even if they worked for Fairlight or some place like that.) As far as using the ADC for 8 channels, from skimming the aforementioned article, it seemed to me the chip was best suited for just doing two channels. It could be set up with a multiplexer, but since it's pretty cheap (under $30 in lots of $1000, so probably under $50 to you and me), you would probably save yourself time and frustration (and hence money) by using several chips instead of a multiplexer. > >With the output being either high quality audio or DAT compatable digital. >The DSP would handle all mixing, EQ, effects, etc. > You'd probably end up using several DSPs. It demands most of the processing power of one current DSP chip (say the Motorola 56000) to implement a fairly complex reverb, OR a graphic EQ, so if you wanted EQ, effects, etc. for each channel, I think you'd end up using at least one DSP per channel. >If samples can be obtained for ~$50 then a box for between $200 to $300 >could be realized. If the DSP is fast enough then another ADC could >easily be added for a grand total of 16 channels. > Um, I think you're underestimating a bit. The socket for a 56000 costs about $20, and it looks like you'll need at least 8 of those. That's $160. Etc... Don't forget to throw in your own personal cost in design, labor, debugging, etc. >Any comments? > There they are. I don't mean to be discouraging, but it sounds like a pretty big project (albeit very interesting -- I wish I had the time to tackle something like it). --Adam * * * KILL YOUR TELEVISION * * * schabtac@stout.ucar.edu