DSPRelated.com
Forums

dolby "B", "C"

Started by nospam October 16, 2003
	I've got some old cassettes with unique and
original recordings that I'de like to move onto
computer.  They were recorded with Dolby B.  Some
were recorded with dolby C.  I'de like to transfer the
tapes to computer and then "decode" the dolby on the
wav files after.
	My understanding is that since dolby is an
analog system, it should be simple to emulate it.  I
already have a feeling that an expander and a low pass
filter (whose strength would follow the loudness contour
of the audio) might be the way to go.
	Any thoughts?


Just curious, why do you want to do it this way instead of using a tapedeck
with the decoding built-in?  Is it for educational/hobby purposes or because
you feel better results can be obtained on the PC?  Sorry I can't help more
as I don't know the specifics of those systems.

"nospam" <PeterC@nym.alias.net.almost> wrote in message
news:3f8eb88c.4137400@localhost...
> I've got some old cassettes with unique and > original recordings that I'de like to move onto > computer. They were recorded with Dolby B. Some > were recorded with dolby C. I'de like to transfer the > tapes to computer and then "decode" the dolby on the > wav files after. > My understanding is that since dolby is an > analog system, it should be simple to emulate it. I > already have a feeling that an expander and a low pass > filter (whose strength would follow the loudness contour > of the audio) might be the way to go. > Any thoughts? > >

nospam wrote:
> > I've got some old cassettes with unique and > original recordings that I'de like to move onto > computer. They were recorded with Dolby B. Some > were recorded with dolby C. I'de like to transfer the > tapes to computer and then "decode" the dolby on the > wav files after. > My understanding is that since dolby is an > analog system, it should be simple to emulate it. I > already have a feeling that an expander and a low pass > filter (whose strength would follow the loudness contour > of the audio) might be the way to go. > Any thoughts?
Dolby B/C is a compander with adaptive lowpass filtering. The response time const is adaptive also. The operation of Dolby strongly depends on the input signal level, so the input should be normalized. You should have some reference level recorded at the tape, The results especially with Dolby C are going to be miserable unless you follow the compander specs exactly. Vladimir Vassilevsky DSP and Mixed Signal Design Consultant http://www.abvolt.com
On Thu, 16 Oct 2003 15:45:25 -0700, "Jon Harris"
<goldentully@hotmail.com> wrote:

>Just curious, why do you want to do it this way instead of using a tapedeck >with the decoding built-in? Is it for educational/hobby purposes or because >you feel better results can be obtained on the PC? Sorry I can't help more >as I don't know the specifics of those systems. >
Hard experiece has demonstrated that dolby recorded on one cassette deck does not seem to decode right when played on another cassette deck. The tapes sound *very* muffly when the dolby circuit is activated because my current cassette deck is not the one they were recorded on. There must be some adjustment in the analog dolby circuitry that drifts out of wack over time, as analog systems will do (ever play with a real analog synthizizer? You gotta re-tune the oscillators every week or so). Decoding in software would allow me to use trial & error to "tune" the dolby decoding process for each recording.
On Thu, 16 Oct 2003 23:39:59 GMT, Vladimir Vassilevsky
<vlv@abvolt.com> wrote:

>Dolby B/C is a compander with adaptive lowpass filtering. The response >time const is adaptive also. The operation of Dolby strongly depends on >the input signal level, so the input should be normalized. You should >have some reference level recorded at the tape, The results especially >with Dolby C are going to be miserable unless you follow the compander >specs exactly. > >Vladimir Vassilevsky > >DSP and Mixed Signal Design Consultant > >http://www.abvolt.com
Thank you. That might be enough to get me started on my bumbling way. :D BTW, years ago someone once told me that dolby C is just dolby B applied twice in series. Does that sound right?
"nospam" <PeterC@nym.alias.net.almost> wrote in message
news:3f944862.21826270@localhost...
> On Thu, 16 Oct 2003 15:45:25 -0700, "Jon Harris" > <goldentully@hotmail.com> wrote: > > >Just curious, why do you want to do it this way instead of using a
tapedeck
> >with the decoding built-in? Is it for educational/hobby purposes or
because
> >you feel better results can be obtained on the PC? Sorry I can't help
more
> >as I don't know the specifics of those systems. > > > > Hard experiece has demonstrated that dolby recorded > on one cassette deck does not seem to decode right > when played on another cassette deck. The tapes > sound *very* muffly when the dolby circuit is > activated because my current cassette deck is not > the one they were recorded on. There must be some > adjustment in the analog dolby circuitry that > drifts out of wack over time, as analog systems > will do (ever play with a real analog synthizizer? > You gotta re-tune the oscillators every week or so).
Dolby is fairly sensitive to the signal level. You should get a Dolby calibration tape, which has a tone of a carefully controlled amplitude. You play that tape, and adjust the gain on the playback circuit going into the Dolby decoder. Though that assumes that they were recorded at the right level in the first place. -- glen
"nospam" <PeterC@nym.alias.net.almost> wrote in message
news:3f944b8b.22635418@localhost...
> On Thu, 16 Oct 2003 23:39:59 GMT, Vladimir Vassilevsky > <vlv@abvolt.com> wrote: > > >Dolby B/C is a compander with adaptive lowpass filtering. The response > >time const is adaptive also. The operation of Dolby strongly depends on > >the input signal level, so the input should be normalized. You should > >have some reference level recorded at the tape, The results especially > >with Dolby C are going to be miserable unless you follow the compander > >specs exactly.
(snip)
> BTW, years ago someone once told me that dolby C is just > dolby B applied twice in series. Does that sound right?
It isn't exactly right, but when C first came out there were implementations that did it using two B chips. Dolby B and C use a compression of only the higher frequencies, and only below a specified level. I think it was 10dB for B and 20dB for C. DBX uses a non-level sensitive compression of the whole signal, so it isn't level or frequency sensitive. -- glen
Glen Herrmannsfeldt wrote:
> > "nospam" <PeterC@nym.alias.net.almost> wrote in message > news:3f944b8b.22635418@localhost... > > On Thu, 16 Oct 2003 23:39:59 GMT, Vladimir Vassilevsky > > <vlv@abvolt.com> wrote: > > > > >Dolby B/C is a compander with adaptive lowpass filtering. The response > > >time const is adaptive also. The operation of Dolby strongly depends on > > >the input signal level, so the input should be normalized. You should > > >have some reference level recorded at the tape, The results especially > > >with Dolby C are going to be miserable unless you follow the compander > > >specs exactly. > > (snip) > > > BTW, years ago someone once told me that dolby C is just > > dolby B applied twice in series. Does that sound right? > > It isn't exactly right, but when C first came out there were implementations > that did it using two B chips. > > Dolby B and C use a compression of only the higher frequencies, and only > below a specified level. I think it was 10dB for B and 20dB for C. > > DBX uses a non-level sensitive compression of the whole signal, so it isn't > level or frequency sensitive. >
<pedantry mode on> DBX does encode with the high frequencies emphasized, and decodes with the inverse, but the compansion isn't frequency sensitive beyond that. </pedantry>
> -- glen
-- Les Cargill
On Mon, 20 Oct 2003 20:54:02 GMT, PeterC@nym.alias.net.almost (nospam)
wrote:

>On Thu, 16 Oct 2003 15:45:25 -0700, "Jon Harris" ><goldentully@hotmail.com> wrote: > >>Just curious, why do you want to do it this way instead of using a tapedeck >>with the decoding built-in? Is it for educational/hobby purposes or because >>you feel better results can be obtained on the PC? Sorry I can't help more >>as I don't know the specifics of those systems. >> > >Hard experiece has demonstrated that dolby recorded >on one cassette deck does not seem to decode right >when played on another cassette deck. The tapes >sound *very* muffly when the dolby circuit is >activated because my current cassette deck is not >the one they were recorded on.
Couldn't this be symptomatic of slightly different head azimuth alignment? Someone else mentioned that Dolby is very level sensitive so a slight alignment error which may not normally be audible could possibly cause a problem with the Dolby system. Best Regards John McCabe To reply by email replace 'nospam' with 'assen'
"Les Cargill" <lcargill@worldnet.att.net> wrote in message
news:3F94CC6A.65505D30@worldnet.att.net...
> Glen Herrmannsfeldt wrote:
(snip)
> > Dolby B and C use a compression of only the higher frequencies, and only > > below a specified level. I think it was 10dB for B and 20dB for C. > > > > DBX uses a non-level sensitive compression of the whole signal, so it
isn't
> > level or frequency sensitive. > > > > <pedantry mode on> > DBX does encode with the high frequencies emphasized, and decodes > with the inverse, but the compansion isn't frequency sensitive beyond > that. > </pedantry>
That may be true. But then there is pre-emphasis/de-emphasis in the whole tape path, separate from any compression/expansion system. I do have a tape deck with Dolby B, C, and DBX all included, and have used all of them. Because of the level sensitivity it is not recommended to use external dolby decoders, but external DBX boxes work just fine. The design of dolby B was that it wouldn't sound too bad if played without decoding, with a slight decrease on the treble control. Dolby C may be too much for that, though. I know sometimes I can hear the effects of using the wrong decoder. -- glen