Bouncing without DBX on.

  • Thread starter Thread starter Victory Pete
  • Start date Start date
Victory Pete

Victory Pete

Banned
Has anyone bounced tracks either in the same machine or to another machine with the DBX off on both machines? The tracks would have been originally recorded with DBX on. I am wondering if because the tape was first encoded then you could do a transfer with it off, after all what is on the tape is an encoded signal. So on playback from the other machine you would then decode with DBX. If this actually works you would have only one generation of DBX "artifacts." I am going to try an experiment with this soon.
VP:cool:
 
Yes, bouncing with dbx off is actually the preferred way, as it reduces distortion and low frequency build-up.

:)
 
Beck said:
Yes, bouncing with dbx off is actually the preferred way, as it reduces distortion and low frequency build-up.
Click to expand...

I'm not doubting you or second-guessing, but do you have anything to back this up with? I'm curious to know why this is.

For me, my guess (and it's only a guess) would be that the combining/mixing of the high frequencies from not-decoded bounce "source" tracks would confuse the dbx decoder during playback of the bounced track and cause weird dynamics fluctuations, especially in the 8k-16k range. I'm guessing this because I think there is the possibility of either the "source" tracks being slightly out of phase with each other, or mixing the not-decoded source signals together would affect the dynamics of the raw track and cause the dbx NR to produce weird anomalies/artifacts while expanding/decoding.

Is that a valid concern? Or, is my understanding of dbx noise reduction a little off?
 
Last edited:
lo.fi.love said:
I'm not doubting you or second-guessing, but do you have anything to back this up with? I'm curious to know why this is.

For me, my guess (and it's only a guess) would be that the combining/mixing of the high frequencies from not-decoded bounce "source" tracks would confuse the dbx decoder during playback of the bounced track and cause weird dynamics fluctuations, especially in the 8k-16k range. I'm guessing this because I think there is either the possibility of either the "source" tracks being slightly out of phase with each other, or that mixing the not-decoded source signals together would affect the dynamics of the raw track and cause the dbx NR to produce weird anomalies/artifacts while expanding/decoding.

Is that a valid concern? Or, is my understanding of dbx noise reduction a little off?
Click to expand...

That is a good point! I initially meant for the bounces to be single track only I think combining tracks would cause problems on decoding. I read some place you arent suppose to do this method when going to a 2 track machine, basically you cant mix any multiple tracks down to 1 or 2 tracks, it will cause DBX errors.
VP:D
 
I agree with what Beck has to say, but would have to add that I imagine that this assumes that the DBX is in perfectly calibrated on both machines for an external bounce, and that each channel/track of DBX is perfectly calibrated for an internal bounce.

For that matter, all levels would have to be perfectly calibrated. For instance, if one track is outputting a little too hot, couldn't this screw up the level of DBX on the recode end?

I personally have never had a problem with any DBX artifacts and on the one occasion that I bounced, which was an internal bounce on an MSR-16, the machine performed much better than I had expected. I took a track of dialogue, added a lot of compressions and some spooky reverb, bounced that, added some more effects, and then bounced again. I was surprised that there was no audible loss in quality of the original dialogue, nor any additional hiss thanks to using DBX throughout.

Let us know how your experiment goes!
-MD
 
lo.fi.love said:
I'm not doubting you or second-guessing, but do you have anything to back this up with? I'm curious to know why this is.
Click to expand...


Yeah... 30 years of experience with dbx and other NR systems in both commercial and home studio environments. ;)

Compared to any of the Dolby's dbx is more forgiving as far as levels are concerned. If you continue to layer encoding upon encoding with dbx you’ll increase distortion and see emphasis of head bump frequencies. One bounce maybe no big deal. More than that though and you'll hear it.

If you have on-board dbx and you want to overdub while bouncing all you need is an outboard dbx unit like the 150X (Type I) or 140X (Type II) depending on what your deck uses.

:)
 
Wait, I think there's a crossed wire here...

I thought the question was basically, "Can I play back several dbx-encoded signals WITHOUT decoding them, mix them down to one or two tracks, and then play back the resulting track through a dbx unit and get a usable result?"

My answer was "No"

I think we're talking about different things... ?
 
Hey Pete, what DBX units are you running? I have the Manuel for the 150 if you need it. PM me.
 
lo.fi.love said:
Wait, I think there's a crossed wire here...

I thought the question was basically, "Can I play back several dbx-encoded signals WITHOUT decoding them, mix them down to one or two tracks, and then play back the resulting track through a dbx unit and get a usable result?"
Click to expand...

Yes... same answer. If you bounce several tracks down to one they are still encoded (compressed) if they were already recorded with dbx.

:)
 
hmmmm, i just tried it both ways with my tascam 414 and when i bounce without the dbx on i get a lot more highs then i started with as well as more noise, and it sounds more compressed. (yes i am switching it back on when im done)
when i bounce with it on it kills a little high end but sounds a lot closer to what i started with.
i'll have to play around with this a bit more.
 
doug deeper said:
hmmmm, i just tried it both ways with my tascam 414 and when i bounce without the dbx on i get a lot more highs then i started with as well as more noise, and it sounds more compressed. (yes i am switching it back on when im done)
when i bounce with it on it kills a little high end but sounds a lot closer to what i started with.
i'll have to play around with this a bit more.
Click to expand...

Are you just bouncing single tracks? no mixdown?
VP
 
Keep experimenting with the record levels while bouncing.

Not to discourage, but units like the 414 really are best suited as composition tools. The frequency response of the 414 is only 40 Hz to 10 kHz +/- 3 dB, and the THD is 2.0%... that's twice what it is for the 246, and more than twice what it is for an open-reel deck. It has very little headroom compared to more expensive decks.

I can say machines like the TSR-8, 38 w/outboard DBX, MSR-16 and older portastudios like the Tascam 246 work flawlessly bouncing with DBX off, either one-to-one transfers or consolidating several tracks to one.

I forgot to ask what decks people are using. That should be my first question before proceeding. DBX implementation varies greatly from the lower cost portastudios to the open-reel machines.

I should not make a blanket statement that applies to all decks from any manufacturer.
 
I have the TSR-8, MSR24 with built in DBX and an MS-16 with its 2 DBX units.
 
You must log in or register to reply here.
Back
Top