To SRC, avoid Pro Tools HD bounce-to-disk?

[Mike H]

I've seen so many conflicting comments regarding degradation of mix sound quality from using Pro Tools bounce-to-disk (BTD) that I would really appreciate others' experiences on this.

I normally mix at 44.1/24. My mix chain is:

Pro Tools HD --> Lavry Blue DAC's --> outboard processing of tracks--> 8816/8804 analog summing/mixing --> mix processing --> ATR 102 tape --> Lavry Blue ADC's --> Pro Tools HD.

So, I end up with the mix on tape, and also as a 44.1/24 stereo track in the Pro Tools session.

(A) If I am going to send the mix to a mastering engineer, I can send the tape or the 44.1/24 mix WAV file. When sending the WAV file, I can either:
(1) Automate the fades and then bounce the automated stereo track to a new (stereo interleaved, or multi-mono) WAV file, or
(2) Just copy the existing (multi-mono) stereo WAV file (w/o automation) and avoid the bounce.

(B) If it must stay in-house and I burn a sample CD, I put the POW-r dither plug-in (16 bits, option 1) on a master fader track and BTD (16-bits, stereo interleaved, convert after bounce) to create a 44.1/16 AIFF file for CD.

(C) If I must end up with an MP3 file for the Internet, I BTD using the Pro Tools MP3 option.

Should I be avoiding BTD at all costs?
What are the best alternate methods?

FYI, the other tools I currently have available are:
A Masterlink, and DSP-Quattro.

Thanks!!

[Greg Wells]

never EVER bounce to disk. digidesign does us all a disservice by continuing to have this old code in the software. i've learned this lesson the hard way.

print your mix to a stereo track in PT and then EXPORT that track. for mastering, simply send the mastering engineer a PT session with your stereo mix track in a session and let them do the external conversion.

if you want an automated fade or any other automation on the mix itself, return the mix to a stereo aux input track, write the automation on that, then bus the output of that to a stereo audio track. hope this makes sense.

[RoundBadge]

Originally Posted by Greg Wells

never EVER bounce to disk. digidesign does us all a disservice by continuing to have this old code in the software. i've learned this lesson the hard way.

print your mix to a stereo track in PT and then EXPORT that track. for mastering, simply send the mastering engineer a PT session with your stereo mix track in a session and let them do the external conversion.

if you want an automated fade or any other automation on the mix itself, return the mix to a stereo aux input track, write the automation on that, then bus the output of that to a stereo audio track. hope this makes sense.

Good advice..

Also ..if you need to SRC/dither for cd ref's or make mp3's as well..
and have to convert your exported untouched file,use a program like Barbabatch,etc to do all the work..NOT Protools conversion..
sounds better than PT processing and it will be the same volume as your original exported file..
.. as where if you bounce the same file from PT using PT conversion/dither...
..to me it sounds F'd up/lower volume in comparison

try it: export the file untouched/convert & dither with Barbabatch...
..then also do a bounce with the PT junk ..

[Mike H]

Sorry guys, as I'm sure you noticed I mis-titled this thread. I am concerned about dither, not SRC...................brain momentarily out of gear.

But I believe your answer is the same: Never bounce, always go outside to another program.

RoundBadge, you've concluded Barbabatch is the best program for Mac? If I'm going to go thru the process, I might as well use the best program available.

[Dan]

How about a file comparison on this one?

[Mike H]

A further thought, as I ponder this:

Wouldn't I be ahead to just run directly from the 1/2" tape into a new Pro tools 44.1/16 session, and avoid any dither from the original 44.1/24 session? I could then take that 44.1/16 file into DSP-Quattro for burning a CD or making an MP3.

I would also still go from tape back into the 44.1/24 session, to create a mix file for the mastering engineer.

[Steve MacMillan]

I just got the new version 3 of Audiofile Engineering's Sample Manager. They have licensed the iZotope sample rate conversion algorithm. I haven't done a real careful comparison with BarbaBatch, but it sure sounded good on the bit I heard. I have seen the white paper and the iZotope algorithm looks like it is a real advancement over anything else.

STeve

[nlc201]

Originally Posted by Mike H

A further thought, as I ponder this:

Wouldn't I be ahead to just run directly from the 1/2" tape into a new Pro tools 44.1/16 session, and avoid any dither from the original 44.1/24 session? I could then take that 44.1/16 file into DSP-Quattro for burning a CD or making an MP3.

Interesting. I would be curious to see what adding dither to an non-"maximized" tape transfer would do, if anything. Since the noise coming off of the tape would most likely be at a higher level than the LSB where the dither would live. Would you be "maximizing" the mix for reference purposes, like a faux master type thing? That would change things for sure. In that case, you're better off working at 24-bit and using whatever digital limiter's onboard dither at the final output. Now keep in mind that I'm talking REFERENCE mixes here, not final deliveries. I honestly don't think that it would matter. When was the last time that the client came back and told you they didn't like the dither or lack of it (unless you're working with Steely Dan or something)?

Originally Posted by Mike H

I would also still go from tape back into the 44.1/24 session, to create a mix file for the mastering engineer.

Why? I would deliver on tape! Let the mastering engineer do the conversion. It would also allow analog processing without the extra A/D - D/A conversion. I'm sure most mastering engineers would prefer this unless they don't have a good machine sitting around (which is pretty rare for decent mastering engineers to not have one!). Some MEs I know also like to spend a fair amount of time tweaking the deck on the output as part of the mastering process. Deliver the tape and have the files available as a backup.


I think that if you transfer to 24-bit / 44.1k off the tape you'll be fine. If you're not maximizing the level, you can probably get away with not dithering for reference purposes. If you are maximizing for reference purposes, just use the built in dither on whatever maximizer you use. It should be fine. Keep the high resolution original file around as mastering backup. Sorry if I'm repeating myself.....

[RoundBadge]

Originally Posted by Steve MacMillan

I just got the new version 3 of Audiofile Engineering's Sample Manager. They have licensed the iZotope sample rate conversion algorithm. I haven't done a real careful comparison with BarbaBatch, but it sure sounded good on the bit I heard. I have seen the white paper and the iZotope algorithm looks like it is a real advancement over anything else.

STeve

I'd like to check that one out,

i've got the IZo Ozone plug in PT .
the dither is pretty good..

Sounds cool..Id like to try thier SRC as well.

Steve,were did you get it?

[RoundBadge]

Originally Posted by Mike H

A further thought, as I ponder this:

Wouldn't I be ahead to just run directly from the 1/2" tape into a new Pro tools 44.1/16 session, and avoid any dither from the original 44.1/24 session? I could then take that 44.1/16 file into DSP-Quattro for burning a CD or making an MP3.

yeah,that 's an interesting idea, for tape print version ref's and stuff.

Originally Posted by Mike H

I would also still go from tape back into the 44.1/24 session, to create a mix file for the mastering engineer.

I usaully let the ME do that[Lavry Golds]

[RoundBadge]

Originally Posted by Mike H

Sorry guys, as I'm sure you noticed I mis-titled this thread. I am concerned about dither, not SRC...................brain momentarily out of gear.

But I believe your answer is the same: Never bounce, always go outside to another program.

RoundBadge, you've concluded Barbabatch is the best program for Mac? If I'm going to go thru the process, I might as well use the best program available.

I think the IZo-ozone ultra dither is pretty good..

I've used various modes with the internal PT power dith[1,2,3.] but they seem to alter the original sound in ways I don't dig, either the highs get kinda crispy or dull depending on the setting.

The B-Batch just seems to alter the sound a little less than the others..

[Steve MacMillan]

Originally Posted by RoundBadge

I'd like to check that one out,

i've got the IZo Ozone plug in PT .
the dither is pretty good..

Sounds cool..Id like to try thier SRC as well.

Steve,were did you get it?

http://audiofile-engineering.com/sample_manager.php

Cheers

[RoundBadge]

Originally Posted by Steve MacMillan

http://audiofile-engineering.com/sample_manager.php

Cheers

Bought it.
Thanks bro.thumbsup

[danasti]

Originally Posted by Greg Wells

never EVER bounce to disk. digidesign does us all a disservice by continuing to have this old code in the software. i've learned this lesson the hard way.

I think they do it so you can quickly have something to take with you in the car or something unimportant. I've also learned not to BTD the hard way and I know I'm not alone. I agree that they should just remove it completely - save everyone the confusion...

[Mike H]

Well, that's a pretty clear message: DON'T bounce! tutt

Let me ask this again, to be sure I'm not missing something:

Since I always mix down to 1/2" tape, can't I just always go back into a Pro Tools session at the sample rate & bit rate that I want to end up at, so I then never have to SRC or dither?

For example, if I am going to burn a CD, just go back into a new 44.1/16 session:
ATR 102 --> Lavry Blue ADC -- PT session (44.1/16)

Then I import the wav file (in the PT session folder) into an audio file processor (Barbabatch, Sample Manager, etc.) and convert from wav to AIFF.

No bouncing, no SRC, no dither.

Am I on the right track now?
Is there something wrong with this approach?

[RoundBadge]

Originally Posted by Mike H

Well, that's a pretty clear message: DON'T bounce! tutt

Let me ask this again, to be sure I'm not missing something:

Since I always mix down to 1/2" tape, can't I just always go back into a Pro Tools session at the sample rate & bit rate that I want to end up at, so I then never have to SRC or dither?

For example, if I am going to burn a CD, just go back into a new 44.1/16 session:
ATR 102 --> Lavry Blue ADC -- PT session (44.1/16)

Then I import the wav file (in the PT session folder) into an audio file processor (Barbabatch, Sample Manager, etc.) and convert from wav to AIFF.

No bouncing, no SRC, no dither.


Am I on the right track now?
Is there something wrong with this approach?

Sure,I don't see why not.Or just record it into pt as an aiff so no format conversion is needed either..
just export the thing straight out and plop it into the burning program
at that point,any PT/ ITB mastering/tweaking stuff would happen at 44.1/16 though
maybe some mastering dudes could chime in.

What the Hell,worth a go.. I'll try it too

[RoundBadge]

BTW, the Sample manger is looking pretty cool..70 bucks.thumbsup
looks like a new "go to" conversion program fo me
the guys that make it definitely care.
i'll have to do a little b-batch shootout.

[picksail]

I still bounce to disc.

In the past, when I was using anything other than PTHD, I chose to 'print' the mix.

[cajonezzz]

Originally Posted by Greg Wells

never EVER bounce to disk. digidesign does us all a disservice by continuing to have this old code in the software. i've learned this lesson the hard way.

print your mix to a stereo track in PT and then EXPORT that track. for mastering, simply send the mastering engineer a PT session with your stereo mix track in a session and let them do the external conversion.

if you want an automated fade or any other automation on the mix itself, return the mix to a stereo aux input track, write the automation on that, then bus the output of that to a stereo audio track. hope this makes sense.

Hey Greg- when you print back to PT, can you rundown the exact process? I've usually got a master fader with sometimes something on it.... do you buss the Master Fader?

Or are you just monitoring a STero audio track as a "master fader" the whole time...?
I"ve been bouncing to disk for some time as well.... sometimes out to a masterlink via AES, but I know a ton of guys printing back... i've just not done it.

?

[mpr3]

I wish this was made into a cut and dry practice, but with the many tools available and the many ways to move between formats, sample rates and bit depths, coming up with the 'best' practice is not easy.

Quick question for a digital guru:

Is it true that most all AD converters perform real time SRC to arrive at the destination SR? (I've read repeatedly that AD converters start the conversion process at a very high SR, and then move down to the destination SR, even 96khz.)

If so, wouldn't it make sense to record our 'analog mixes' back to digital at higher sample rates and then rely on high quality 'offline' SRC algorithms that potentially offer a cleaner filtering down to the edge of our hearing range? Afterall, how could a converter priced under $5k achieve, in realtime, a conversion quality that rivals what an algorithm can do with all the DSP and time it needs to get the job done with the least amount of artifact?

[Ruudman]

In PT you have an on/off setting for RTSRC

ruudman

[elarreal]

[QUOTE=Greg Wells;1044253]never EVER bounce to disk. digidesign does us all a disservice by continuing to have this old code in the software. i've learned this lesson the hard way.

Hi Greg,

Does Logic Pro 7.2 behave in the same manner as PT when it comes to the SRC, BTD & Dithering?

I ask because I believe you are also a Logic user. (correct me if I'm wrong)


Thanks

[K Bartos]

Hi Guys !

What exactly is the difference between BTD and "printing" to disk ? I assume that printing is playing the mix back and recording it onto anoter track in realtime.
Don't the two processes do the same thing ?
I was under the impression that PT had a zero tolerance for any errors during both playback and BTD and that's whats makes it more prone to halt due to "System held off interrupts" and buffer errors and the like.
The impression I've gotten from reading the DUC is that other software tolerate errors during playback (which is now a feature that can be implemented in the prefs in PT 7.3), while PT doesn't. That would guarantee and playback and BTD to be error free.
But your are talking about the ITB summing algo that happens while BTD ?
Can u guys please elaborate ?

Kristian

[mpr3]

They are referring to the Bounce to Disk feature in PT that *should* be identical to bussing the mastering fader out to another stereo audio track and recording to that manually.

I have never done the phase null test, but one would assume that both methods would result in a perfect bit for bit null.

I have not had any problems with our quickie refs printed using BTD. I do mostly use an outboard mix bus compressor on final mixes and so naturally end up printing mixes to a stereo track and have never noticed an improvement or difference aside from the qualities imparted from the compressor (2500 or 33609).

What exactly are you guys noticing that causes you to avoid BTD?

[Ruudman]

If you are to do a phase null test, please make sure you don't use any fx plugs on the mix.



ruudman

[Steve MacMillan]

I've been using ProTools every day since v1. Back in the MIX days, there was a hot debate over BTD where many people thought that the stereo image collapsed and the mix generally didn't sound as good as printing thru AES or internally thru a stereo buss. I was never quite sure myself because some of the best sounding mixes I did in the box in those days were BTDs. However I did notice that complicated automated vocal eq and automated filter sweeps weren't hitting their marks when I would BTD, and if I printed thru AES to a Masterlink everything sounded fine. You had to record to a Masterlink because printing back into the session with a fully automated mix with every last bit of DSP in play was impossible. So I did some careful tests, and sure enough, BTD was not nearly as accurate as AES with automation. So I stopped caring about the sound quality question and never did I BTD.

A few months back I was recommending to a client that they never use BTD, because of the automation problems. I realized that it had been a long time since I had done my tests, and maybe I should check it out again with my HD system (running 7.2 at the time). So again I set up an automation torture test, and to my surprise BTD outperformed internal recording by a very noticeable margin. I then I thought I would do some listening tests, and to be honest I couldn't tell the difference with some very hifi material.

So until I learn otherwise, I've dropped my long standing dislike of BTD. I still think there are many advantages to recording internally. I like being able to punch in on mix prints, etc.

STeve

[RoundBadge]

Originally Posted by cajonezzz

Hey Greg- when you print back to PT, can you rundown the exact process? I've usually got a master fader with sometimes something on it.... do you buss the Master Fader?

Or are you just monitoring a STero audio track as a "master fader" the whole time...?
I"ve been bouncing to disk for some time as well.... sometimes out to a masterlink via AES, but I know a ton of guys printing back... i've just not done it.

?

For an all ITB internal print,I tried ditching the master fader and bussing the entire mix to a stereo aux with all the 2 buss plugs there then printed to another stereo track[everthing at unity of course]then exported as an unconverted stereo file.
..seems ok.

[picksail]

Steve, this is exactly the same experience that I had. I've been on PT since '94ish. I really never endorsed the idea of BTD until HD was introduced. Oh, the MIX+++ days...

Thank goodness for HD.

[K Bartos]

I know this is High-End but I still need to ask;
Does the improved BTD since the Mix-days also apply to PTLE ?
I'm fully aware that there are differences in the audio processing structures in the two.
Is there a difference between HD and LE in this respect ?

[C.Lambrechts]

Originally Posted by K Bartos

I know this is High-End but I still need to ask;
Does the improved BTD since the Mix-days also apply to PTLE ?
I'm fully aware that there are differences in the audio processing structures in the two.
Is there a difference between HD and LE in this respect ?

I doubt there's a difference between BTD in LE or HD software. It is basically the same software code with restrictions towards hardware and some features. The code is exactly the same.

Either that or I'm just talking out of the back of my neck ... I'm not a software writer but LE and HD using the same code would make sense.


That said ... as far as I have been able to test ... which is record the mix to an internal track through a bus or BTD and then compare those in a 'phase null test' they have allways 'nulled' out for me. Nothing wrong with BTD imvho.

There have been automation problems in the past with the use of certain plugins in certain conditions (can't remember exactly how and what) but they applied on playback as well so basically it wouldn't make a difference wether you bounced or rerecorded .... the problem would be there.

I tried the 'null' test on a couple of recent mixes using a variety of plugins, track numbers , density in automation, outboard used etc etc .... Both the internal rerecorded track and the bounced files 'null' out perfectly. These BTD problems are a myth ... if you can hear a difference between a rerecorded mix and a BTD mix then something else is wrong ...