DBX Noise Reduction with Digital?

  • Thread starter Thread starter camus
  • Start date Start date
C

camus

New member
I read somewhere that racking a dbx noise reduction unit in front of your DAT can improve your dynamic range and prevent quantization noise. Has anybody tried this? I never thought a digital recordr would be able to "read" dbx encoding.....is this even possible?
 
Well, you could encode and decode DBX on a DAT but, IMHO, I can't imagine any benifit. Where did you read that?
 
Say what??? Why on earth would you wanna do that???

A DAT's dynamic range is upwards of 95db -- you want more???

And on a related-note... does it make ANY sense to have all the dynamic range at our disposal, then have to squash the shit out of the mix, reducing the dynamic range to about 3 db anyway?!??!?!

But anyways, even if it did help the quantize noise.... if you have any sort of decent converters at all, you should have to worry about it!

Bruce
 
heheh....I guess that'll teach me to stop pokin me nose around some of these "home audiophile" messageboards....*cough cough*
 
Oh Jezus! Did you read anything about green majic markers and CDs?
 
Blue Bear Sound said:
if you have any sort of decent converters at all, you should have to worry about it!
Click to expand...
Damn drunk Bears!
(no, not really...) ;)

But that should have read "you shouldn't have to worry about it!"...

Bruce
 
Blue Bear Sound said:
Say what??? Why on earth would you wanna do that???

A DAT's dynamic range is upwards of 95db -- you want more???

And on a related-note... does it make ANY sense to have all the dynamic range at our disposal, then have to squash the shit out of the mix, reducing the dynamic range to about 3 db anyway?!??!?!

But anyways, even if it did help the quantize noise.... if you have any sort of decent converters at all, you should have to worry about it!

Bruce
Click to expand...
I thought DATs (on a good day :D) would yield high 80's to low 90's of dynamic range. Additionally, if I am not mistaken, the DBX noise reduction units do not use compression in any form, so you would not be decreasing the dynamic range in the music any. I was under the impression that they encode the signal(by boosting the high frequencies) and then decode upon playback...decoding would attenuate the same coresponding frequencies..thus attenuating the noise in said frequencies at the same time.... (at least my dinosaur, type I units do)....signal altering, yes...but compression no. ...well now that I think about it, they may use [selected band] compression and expansion, but I don't think so...but I'm to lazy to look it up. Additionally, isn't *real-life-human-hearing* in the order of 120dB+ dynamic range, so if recordings could reflect that, it would be a step closer to "live." ...not that the DBX
unit are going to help get you there...but I'd certainly prefer that preference [of achieving better dynamic range] over the MP3 rage.....right?? As far as [quantization] noise, I believe the DBX units are also more geared for high freq. noise, specifically tape hiss, rather than as a general noise reduction.

Interesting theory...actually might warrent a try, but the bass altering effects would probably negate any positive results, as that usually was the problem if calibration was off just the slightest (or not at all!:D ) of all the machines involved in the signal chain....tape recorder especially.
 
Last edited:
The way I understand the DBX process is that it compresses ( the encoding part) and then expands (the decoding part) signal. I'll have to dig out my old manual for a unit I used to have where it explained the process. It's been a while. Personally, I could hear artifacts of the thing working albiet, that was on analog tape. Kind of a halo of fuzz sound.
 
Great - so even if it were to work, on mixdown/mastering/post-processing you're still going to squash the dynamic range down to maybe 10db at best!

Plus... it would have to be an expensive DBX encoder/decoder to not affect the high-sound quality of the digital signal. As Track Rat said, the DBX stuff has a problem with artifacts at the crossover threshold of being on/off, so they would actually introduce MORE noise into the signal at a much higher level than any quantization noise.

Personally, I'd just color the cable with green marker and let THAT take care of the "quanitization noise" (which again, with good converters SHOULDN'T be a problem to begin with).

God... ya gotta *love* this bullshit audio mythology that sprouts up.... :rolleyes:

Yikes.................

(not saying it's you, Camus - just whoever you heard this from!)


Bruce
 
Last edited:
Well, I've been thinking a little more about it. The "tip" (ahem) came from an audiophile mailing list, so I would think it would actually apply more in terms of improving what you HEAR rather than what was recorded.

I guess the whole theory would revolve around the dbx squashing the incoming signal into the DAT, compressing the dynamic range so you would be able to record at "hotter" levels than normal, thus getting more bits for yer buck. Upon playback, the dbx would then expand the recorded signal back again, in theory, giving you back the dynamic range and at the same time improving your recorded S/N ratio.

Interesting, but it works only as a "playback" noise remover I guess. Doesn't actually help in the final mixdown/mastering process!
 
So let's say it DOES work... you've replaced the "cleanup" of quantization noise with artifacts from the DBX decoding process.... you're actually worse off!

Bruce
 
I can virtually garantee that the artifacts from a DBX unit would be way more audible that quantization noise ffom even first generation DAT A/D/A convertors. You want a big improvement? Pick up an Apogee Rosetta or a Lucid for a front end for a DAT.
 
Blue Bear Sound said:
Plus... it would have to be an expensive DBX encoder/decoder to not affect the high-sound quality of the digital signal. As Track Rat said, the DBX stuff has a problem with artifacts at the crossover threshold of being on/off, so they would actually introduce MORE noise into the signal at a much higher level than any quantization noise.

Bruce
Click to expand...

I don't think expense has anything to do with the dbx units. At the time of their manufacture, there wasn't a lineup of economy to best. ...just what they manufactured. They did come out with a variety of units during their heyday. I don't understand the statement about the crossover threshold of on/off. There is no such thing with dbx noise reduction. You either encode the signal, or you turn off/bypass the process. Anyway, I wasn't really a big fan of dbx either,.... but back *when*, it made my 1/2" 8 track sound better...I preferred the dimished tape hiss over the bass "pumping" it wanted to do....The vertical rack mount units (remember all the Allison Gain Brains and Keypex's??)seemed to work pretty good on a precisely calibrated pro format reel to reel.....but a lot of pros preferred the 30ips, etc...etc...ya know the story, I'm sure.
 
What I meant was the frequency-dependent "shhh" noise artifacts that surrounded transients... I beleive it was related to the crossover points of the DBX encoding process....

There were different types of DBX encoders/decoders.... consumer decks equipped with DBX were fitted with Type I while professional studios used Type II or eventually Type III... (if memory serves me correctly!)

There was a significant difference in performance between the consumer flavour and the pro version...........

On an aside, my DAT (DA-40) is spec'd at 93db dynamic range - the model up from it (DA-45HR) is spec'd at 113db...

The point is moot anyways regarding DBX..... DBX will cause more problems than it fixes...... I agree with Track Rat - if Camus wants to see an improvement, he should pick up a good converter!

Bruce
 
You must log in or register to reply here.
Back
Top