Is DP's summing wack?

[goldenarmes]

Hmmm...Great discussion. I am no where near being a mix engineer. I do all the editing for a project studio am am trying to learn more and want to give the clients the best quality I can give them. So here's my question:

If I under stand this right, I am to set the master fader to -6db when bouncing to disk? Then would it be acceptable to then import the bounce into something like PEAK and do the compression/ limiting in there or would I then be running into the same problems?

Now then that raises one more question for me. If I were going to take a mix to a mastering engineer how much headroom should i leave... -6db?

Thank you everyone for all you time and help over the years

[amerecordingstudio]

The thread was a few months ago and is definately worth reading. I think Lynn's site is: www.3daudio.com

Anybody actually find this thread? i can't, but i'd love to read it.

[chrispick]

If I under stand this right, I am to set the master fader to -6db when bouncing to disk? Then would it be acceptable to then import the bounce into something like PEAK and do the compression/ limiting in there or would I then be running into the same problems?

Now then that raises one more question for me. If I were going to take a mix to a mastering engineer how much headroom should i leave... -6db?

Answering your questions in order:

Forgive me if I overstate stuff. Just trying to be clear...

1.) No, you probably don't want to set your master fader to -6dB. Leave it at 0 (unity, as they say). What you do want is for the levels in your master fader channel -- your mix levels -- to peak at -6dBFS. This equals out to between +3 and +4 dB on DP's master fader, roughly speaking.

To clarify: The FS in dBFS stands for "Full Scale" and refers to the maximum level a digital file can withstand before clipping occurs. 0 dbFS equals +6 dB on DP's master fader (approximately, that is -- DP actually leaves a teeny bit of headroom). If you hit red on DP's master fader, you've exceeded 0 dBFS and you've clipped your audio.

To note: Most contemporary CD have peak levels around -0.1 dBFS to get things as loud and possible. Doesn't mean you have to pump it up that loud. Just giving you reference.

If you want precise metering of your master fader, I suggest picking up Elemental Audio's Inspector XL. Set its meter to read "linear" and you'll know exactly where your dBFS peak levels hit.

2.) Yeah, it's acceptable to master by bouncing your mix to disk, then doing the mastering work in Peak or other high-grade stereo audio editing app. Lots of people do it. It's what I do. And yeah, that's when you perform all the finishing adjustments like final limiting (i.e., pushing peak levels up to -0.1).

BTW: Technically, this last-stage limiting/compressing/EQing/whatever is really called premastering, although few people use this term. Mastering is actually the step of transferring the data to media (e.g., burning a CD).

I recommend getting some good mastering plugs to use in Peak to do it. I use Izotope's Ozone 3 (in conjunction with some other plugs); I think it's a great bang-for-the-buck. Other great mastering plugs include those by UA (UAD-1), PSP, WaveArts and Waves.

3.) You should confer with your mastering engineer to see what specs he/she demands. The -6dBFS gives them a lot of headroom certainly, but many of these guys use digital and analog tools to do their deed, so they'll have specific standards for you to meet.

Hope this helps.

[Fibes]

I've been busy and it's good to see some lively discussion...

In a DAW (just like with DAT) don't mix for level, your mixes reatain their imaging, punch and you won't get that fog that so many people complain about. Those of us who came from Analog realize this is counterintuitive and the dudes raised during the loudness wars are trying to keep up with the Jones' so initially we were both screwed.

Leave the level stuff for the mastering stage where you (or a qualified M.E.) can use mastering procedures to get your level to where you want it. That doesn't mean you can't use a mix buss comp but limiting isn't gonna help if it's merely for the sake of level.

The day i figured this out I realized the summing buss marketing hype was just that. Things can punch ITB but you have to work in a way that allows it to happen.

Does that mean i wouldn't prefer an ATR 2" and an API board? No, but the urge to pint the finger at the lifeless mixes no longer goes to the DAW it goes between the chair and the DAW.

Yes, you do have to take different steps to get great sounds in digital that are sometimes very different than analog. From mic selection and placement to compression and from recording levels to mix levels. Only a fool would think that more is always more with music; right?

Have a wonderful Thanksgiving from the most violent vegetarian in georgia.

[Tritonemusic]

...

[Timeline]

Does that mean i wouldn't prefer an ATR 2" and an API board? No, but the urge to point the finger at the lifeless mixes no longer goes to the DAW it goes between the chair and the DAW.

Beautiful.

I see your point but there's no doubt that having to get around a DAW when producing music takes some life out of the creative process musically even though you can clean up •••• better later.

All in all I guess I like having both now that ADDA conversion tones are better. Maybe the sound will start to turn me on more and all the tech manuvering won't matter in the future, i don't know, but as a producer, I've been thinking about hiring recordists so i can do my thing better as opposed to being my own. After all, it is all about the music....

[stickwolf]

What if for a particular project, I plan to do the limited mastering inside DP? Should I mix to -6dBFS, export, and then import into a fresh sequence for mastering, or should I just run the mastering stuff on the master fader? It'd seem that I'd end up in the same exact place. And should I or should I not assume that if I do a very basic mastering of running ONE limiter plug-in over the stereo mix, that it would be IDENTICAL to do in DP or Peak or whatever, with the same plug in?

Is this all related to dithering at all? Assuming I'm using DP alone, and serious projects I'll send out for mastering but others I'll do myself, well, what difference does it make about exporting or anything? Should I just try to have the levels that are coming into the master fader at -6dBFS?

[blue]

In a DAW (just like with DAT) don't mix for level, your mixes reatain their imaging, punch and you won't get that fog that so many people complain about. Those of us who came from Analog realize this is counterintuitive and the dudes raised during the loudness wars are trying to keep up with the Jones' so initially we were both screwed.

Leave the level stuff for the mastering stage where you (or a qualified M.E.) can use mastering procedures to get your level to where you want it. That doesn't mean you can't use a mix buss comp but limiting isn't gonna help if it's merely for the sake of level.

i'm not sure i follow. are you saying there's a sonic difference between applying mastering procedures at the summing stage and doing it afterwards? even if you're using the same plugs to get the right levels?

i'd love to know more about this. i don't always get the opportunity to have my stuff mastered professionally, so anything that would improve my final mixes would be great.

[stickwolf]

I'm not saying there *is* anything. I'm ASKING.

My impression from previous posts by others is:
follow the rule of thumb of leaving keeping the level in the master faders at -6dBFS or just lower (which is around +3 or +4 according to DP's numbers).

My questions are: A. why? and B. if this is suggested for exporting to then master in Peak, then wouldn't it also be suggested to do for mastering in DP? And also, wouldn't mastering in DP vs. Peak be identical if the same plugins are used? And finally, would exporting and then importing back into DP for mastering be any different than just running the mastering plugs before exporting?

[chrispick]

I'm not saying there *is* anything. I'm ASKING.

My impression from previous posts by others is:
follow the rule of thumb of leaving keeping the level in the master faders at -6dBFS or just lower (which is around +3 or +4 according to DP's numbers).

My questions are: A. why? and B. if this is suggested for exporting to then master in Peak, then wouldn't it also be suggested to do for mastering in DP? And also, wouldn't mastering in DP vs. Peak be identical if the same plugins are used? And finally, would exporting and then importing back into DP for mastering be any different than just running the mastering plugs before exporting?

A.) Well, let me start with this caveat: I don't have any science to back up the -6dbFS philosophy. It's a tip I learned from a mastering dude, and my experience has been that it produces clearer mixes. For me, pushing the mix levels to the hilt tends to make the mix a bit muddier.

My uneducated guess is this -6 deal has something to do with additively blending multiple tracks with their many effect plugs into a stereo track. Keeping the master fader peaks at -6dBFS keeps all the in-coming faders at similar or lower peak levels, and that gives ample headroom for all to calculate efficiently.

B.) Yeah, you don't have to do your mastering pass in Peak or other external stereo editor; you could do it in DP. I think the key is to do it as a second pass, after you've already mixed your stereo master stem (for the above reasons, I guess).

I move things into Peak to master because it's a simplified screen layout, I can output multiple formats from there quickly and easily, and operations like fades, silencing, edge-trimming and disc-burning are super-easy.

I also think it sounds a little clearer when mastering in Peak instead of DP, but I've never empirically tested this. This is why I'm curious about DP's summing capabilities. It's not an ITB vs. OTB issue for me. It's a "does-DP-sum-as-well-as-other-software" question.

[stickwolf]

So you ARE suggesting that exporting to a stereo master file and then reimporting to do the mastering is recommended. Yes?

Obviously if anybody gets actual data about ANY of this, we all want to know!

[blue]

I'm not saying there *is* anything. I'm ASKING.

i hope you didn't think i was responding to your question! i was responding to fibes post.

[stickwolf]

I guess I erred. I thought you were responding to my post since you were asking basically the same thing, so I figured you didn't understand my question, but maybe we just posted at about the same time. Oops.

[qo]

I'm also curious about what it is, technically, that recommends -6dBFS. FWIW, I did use InspectorXL on these mixes and there are zero overs of any kind on any track, nor at the Master fader. So, if there are no overs, then the reason to use -6dBFS can't be to avoid overs when multiple tracks are summed to stereo, right? Also, if the -6dBFS trick is to help with summing, then wouldn't it be dependent on the number of tracks in a project? That is, wouldn't the per-track level have to be different when summing 16 tracks, vs 200 tracks?

I'm not disputing the -6dBFS advice. Rather, I'm just curious about what's going on in mathland behind the scenes that makes -6dBFS better. And, wonder if there are scenarios where, say, -10dBFS would be better? If this is the case, then shouldn't MOTU change the overload indicators to adapt to the number of tracks being summed?

Thanks!

qo

[toodamnhip]

I've been suspicious of DP's summing for a while. After recently acquiring a Soundcraft Ghost 32x8 console, I decided to take a recent project that was mixed in DP and start remixing it with the Ghost. ALL plugins were removed, DP's faders were double-clicked to snap it to 0db, same for the pan pots to snap them to centered. Each channel output was then routed to a separate DA channel and these were run into the Ghost Tape/MixB inputs.

The difference (again, with no plugins) is, frankly, remarkable. It's like a "fog" is removed from the audio. If there is interest, I'll post a "before" and "after" snippet from one of the tunes and we can have a blind test with ya'll deciding which of the two mixes sound more transparent, focused, etc. Not only is the difference very noticable, but it seems much easier to get a decent mix more quickly. Things just seem to fall into place and "work" without a whole lot of effort.

So, what's the deal here? Is it the plugins (e.g. EAS Eqium/Firium, Sonalksis comps, Altiverb)? Or DP's summing? Or DP's pan-law choice? Or some interaction between DP and Apogee AD/DA CoreAudio? Or something else? I'm substituting two ELI Distressors for the comps, and a TC M-ONE for the verb and using the Ghost's EQ to replace Eqium/Firium. The styles in this project are e.g. urban/folk/rock.

I used to work with tape and consoles years ago and do remember that things seemed generally easier back then. But, I took a hiatus from music production for about a decade and went straight to DAWs upon my return, so didn't really have any sort of back-to-back OTB/ITB experiences to draw from.

Again, I can post some examples this weekend if there's interest in this. But, my main reason for posting this is to see if you folks have thoughts about where the source of the fog may be, or if anyone else has experienced the same thing. I'm using DP 4.6 (which includes plugin delay-compensation), and Apogee AD/DA with native CoreAudio drivers.

For now, I'm swearing off plugins of any sort and am going to use DP as a simple tape recorder (no automation, no nuttin!). Since the Ghost only has mute automation, this is going to be a major change in the workflow. It means that I'm gonna have to finish mixes right then and there, as opposed to saving them for a second listen a few days later. But, really, the sacrifice is worth it given the initial experience of the last couple days.

Thanks!

qo

Thi summing test is not a fair test in my opinion for the "obvoius" reason that, you're adding in the sound of all the consoles inputs, whether pre-amp based or a supposedly benign, flat input.
It is very possible that you just enjoy the sound of your consoles electronics as applys to several channel of audio.

There is really no way to make a fair test as the moment you sum into another system, you are adding the console's sound