home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.electronics
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!sdd.hp.com!wupost!eclnews!wuee1!anbo
- From: anbo@wuee1.wustl.edu (Albrecht Jander)
- Subject: Re: 1 Bit DAC on CD players
- Message-ID: <1992Aug12.145928.882@wuecl.wustl.edu>
- Sender: usenet@wuecl.wustl.edu (Usenet Administrator)
- Nntp-Posting-Host: wuee1
- Organization: Washington University, St. Louis Mo.
- References: <1992Aug10.024538.19830@sactoh0.sac.ca.us> <8135@sirius.ucs.adelaide.edu.au>
- Date: Wed, 12 Aug 1992 14:59:28 GMT
- Lines: 61
-
- In article <8135@sirius.ucs.adelaide.edu.au> francis@cs.adelaide.edu.au writes:
- >In article <1992Aug10.024538.19830@sactoh0.sac.ca.us>, sap@sactoh0.sac.ca.us (Stephan A. Puga) writes:
- >|> Why is a 1 bit DAC on portable CD players better? I see an awful
- >|> lot of ads and such touting the greatness of 1bit DAC. It seems to
- >|> me you could get faster accessing if you had an 8bit DAC, but I
- >|> don't know. If anyone could shed some light on this matter I would
- >|> be very grateful.
- >
- >Very briefly: all players decode the 16 bit x 2 information on a CD.
- >The one bit DAs are simply just one of a number of ways to make a
- >DA that converts the 44.1kHz x 16 bit signal into sound.
- >
- >Although the data format of a CD does at one point involve eight bit
- >data this is a long way from where it gets turned into sound. Speed
- >of accessing makes litte sense in the context of audio CDs.
- >
- >One-bit/bitstream/MASH designs are simply ones that involve a different
- >set of compromises to some other designs. They require more
- >computational power in the digital filtering section and uses a simpler
- >analogue part. Since we are getting better and better at making cheap
- >fast digital bits, but progress is a lot slower on the analogue design
- >front the one-bit designs seem to have the edge in the tradeoff game.
- >They are cheaper to make and produce better results; for the moment.
- >
- >The ultra high end of hifi still has adherents to the more conventional
- >16 bit or wider converters, claiming that if you put the money and effort
- >onto these designs they sound better than the one-bit designs. Serious money
- >and effort, and you would buy an argument about the technical merits.
- >High end hifi does not go by the same rules as the rest of us.
- >
- >Of course it is newer and therefore better so far as advertising copywriters
- >go. Next year it will be some other techno-flash technology. At least
- >one-bit DA design has some merit unlike a lot of the other hype we see.
- >
- >
- > Francis Vaughan
-
- One bit DAC's are inherently linear. (You can't put two points in space
- without being able to draw a straight line through them.) The problem with
- traditional designs of high resolution DAC's has been to achieve linearity,
- or even monotinicity (a bigger number in always should produce a bigger
- voltage out). Most traditional DAC's use some kind of switch to switch
- current sources onto the output pin. The problem is that it's really
- hard to match current sources in one part in 2^15 unless you laser trim
- after manufacture. This is what makes good 16 bit converters expensive.
-
- It turns out that with proper filtering, you can trade bits for sample rate
- without losing any of the signal. A one bit DAC is very course, but it is
- run at a couple of megahertz rather than 44.1KHz. All you need is fast
- digital circuitry and some simple analog processing to make a very linear
- 16 bit DAC. That's alot cheaper than laser trimmed analog circuits.
-
- I am by no means an expert on the subject, but I hope that clears up some
- questions.
-
- Albrecht
-
-
-
- ---------------------------------------------------------------------------
- Peltier was a cool dude!!
-