Is DP's summing wack?

[sdemott]

If I remember correctly the issue with the DAW and the -6dBFS threshold has to do with adding ANY plugin to a signal that is over -6.00dBFS. There was a long thread about this on Lynn Fuston's site with some highly respected digital experts chiming in. They even explained the test to check (and I don't remember it now, but it was enlightening). From what I remember the processing of the signal causes distortion and overs that are not captured on the meters at all but have a profound effect on headroom and the overall dimension of the mix. As long as the signal was under -6dBFS you were fine.

Ever since then I have kept alll signals under -6dBFs even at the Master Fader and have noticed a very big difference in the dimension and realism of my mixes.

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

[buzzsmith]

http://www.3daudioinc.com/

[giles117]

Ever since then I have kept alll signals under -6dBFs even at the Master Fader and have noticed a very big difference in the dimension and realism of my mixes.

Becuase I believe in ITB mixing I have been mixing like this for quite some time.

When I cut I tend to cut no louder than -9dBFS

Keeps the mixes warmer and allows for alot of dimension.

Lot of arguments, but a lot of satisified cutomers is my bottom line....

[qo]

Warning, potentially stupid questions approaching...

If -6dbfs is the answer for DAWs, even at the master fader, how does one go about getting a "hot" mix using a DAW? I'm not a big fan of nuked (heavily compressed) mixes, but I do like to strive for a reasonable level (though not to the detriment of the material). Do you -6db'ers then send your work to an OTB mastering facility (or master yourself OTB and, if so, how?)? If mastering is done ITB, then wouldn't it be limited to -6dbfs as well, or does this -6dbfs thing only pertain to summing?

And, if -6dbfs is the point at which a mix collapses, why don't DAW vendors move their overload indications down to this level? MOTU?

Part of what I'm liking about this initial experience with DAW+external console is that the levels within the console (and between the console and external hardware-based processors) can be pushed pretty hard without worrying about a few db here and a few db there when I'm really wanting to concentrate more on the music. It just seems like a much more tolerant/forgiving signal chain (assuming -6dbfs is the answer for achieving ITB nirvana). But, if DAWs (and plugins) would warn at -6dbfs, and if there's still a way to get a reasonable level in the final product, then maybe DAWs can be made to be just as forgiving?

[giles117]

I send all my work out to a mastering lab. The dictate at that point is the clients budget....

[Jaysplace101]

I think you can keep it at -6 until you print the stereo mix. Then, you normalize and throw on the compression. The problem is, I like to put some comression on the master while I'm mixing, so that makes it a little more tricky.

j

[croyal]

I find that -6dBFS is a good ref because of the unseen clipping going on. MagicDave (I think) talked about the clipping internally, but not being a problem as long as the Master didn't clip. Good advice, but things sound so much better ITB when there's no clipping going on behind the scenes. For me, individual tracks can be recorded up to 0dBFS, but when there's any summing going on I'll limit myself to -6dBFS master output (as well as any summing Auxes) for extra headroom.

Also, Magic Dave pointed out that the metering on DP is related to the originally recorded signal (meaning the fader at "0" means no attenuation and +2 means you are adding 2db to the original signal)

So to see digital levels in DP you need to open a plugin with digital metering (such as Trim) to see relationships to 0dBFS. In other words, don't rely on the Master fader metering for anything other than a clip indicator or signal present.

I notice I can get the "better" mixes in ITB when I used better analog stuff on the front end. The little analog distortions are cumulative and are a huge part of the OTB experience. So Q, try tracking through your Ghost (as hot as you wish) into a Compressor or two (set to very low compression, 1:1 or 2:1, but not bypased) then into your AD (limiting as neccessary). Thus when you mix ITB the sonic signature of the Ghost [etc.] is already acting as glue to each track. IMO it is much easier to mix when good things were done while tracking. For instance, gentle Eq in the Ghost on the way into DP is not bad idea.

Chris

[Tritonemusic]

..

[chrispick]

I find that -6dBFS is a good ref because of the unseen clipping going on. MagicDave (I think) talked about the clipping internally, but not being a problem as long as the Master didn't clip. Good advice, but things sound so much better ITB when there's no clipping going on behind the scenes. For me, individual tracks can be recorded up to 0dBFS, but when there's any summing going on I'll limit myself to -6dBFS master output (as well as any summing Auxes) for extra headroom.

Also, Magic Dave pointed out that the metering on DP is related to the originally recorded signal (meaning the fader at "0" means no attenuation and +2 means you are adding 2db to the original signal)

So to see digital levels in DP you need to open a plugin with digital metering (such as Trim) to see relationships to 0dBFS. In other words, don't rely on the Master fader metering for anything other than a clip indicator or signal present.

The above is what I've learned as well. I think I ran across the -6dBFS info on recording.org from one the frequent posters who's a mastering engineer. The headroom really helps I've found.

Thereafter, I take the summed master stems and perform all the premastering stuff I do (a little UAD-1 Precision EQ, a little PSP VintageWarmer, a healthy hunk of Ozone3 limiting, exciting, sub-bass roll-off, etc.).

BTW: Inspector XL will allow you to meter all of this with greater precision. I always keep one in my master buss for this reason (among others).

[qo]

So Q, try tracking through your Ghost (as hot as you wish) into a Compressor or two (set to very low compression, 1:1 or 2:1, but not bypased) then into your AD (limiting as neccessary). Thus when you mix ITB the sonic signature of the Ghost [etc.] is already acting as glue to each track. IMO it is much easier to mix when good things were done while tracking. For instance, gentle Eq in the Ghost on the way into DP is not bad idea.

Thanks Chris! I'll definitely give this a shot since I'd really really hate to lose the automation/recall that DP provides. Sounds like there are many months of experimentation in my future...so what else is new?

[wylie]

So Q, try tracking through your Ghost (as hot as you wish) into a Compressor or two (set to very low compression, 1:1 or 2:1, but not bypased) then into your AD (limiting as neccessary). Thus when you mix ITB the sonic signature of the Ghost [etc.] is already acting as glue to each track. IMO it is much easier to mix when good things were done while tracking. For instance, gentle Eq in the Ghost on the way into DP is not bad idea.

Thanks Chris! I'll definitely give this a shot since I'd really really hate to lose the automation/recall that DP provides. Sounds like there are many months of experimentation in my future...so what else is new?

Same here.
Great topic Q.

[danny]

your Ghost will whip any workstation, imo, pretty much anything will.

[Jaysplace101]

Then, you normalize and throw on the compression.

Why normalize? That's a destructive process which will degrade your audio. It's absolutely unnecessary. Use your compressor to to bring the levels up. Normalizing just makes audio quality worse. Stay in the floating point environment until your very last step (when bouncing and dithering to 16-bit), if possible. If you care about your audio, you'll never normalize again.

All normalizing does is brings everything up to zero (or whatever level you want it to bring it up to.) It's basically just changing the gain for the entire file. Sure it would degrade the audio if you pushed it past zero db, but as far as I know, if you just normalize up to zero db, it's just analyzing the file to see how much lower than zero it is at the peaks and then adding that much gain to the entire file. Please tell me if I'm wrong (for all these years) I certainly don't HEAR any degradation.

j

[danny]

Yes, getting stuff sounding right going in is important, but summing is a separate issue, today's discussion You don't need to sacrifice ALL plugins and ALL automation to go the hybrid route. If you've got racks of high end comps and eq's, of course, then go ahead and trash the plugs, heh.

If you do most of the level, panning, eq, comps, in DP you'll end up using the mixer in a very simple way, so you'll end up with having like 85% recall, and the 15% you do have to recreate is basic fader levels and any outboard stuff you patched in. The improvement derived from summing analog will make this well worth the bother, imo. People don't need $4000 summing boxes when so many full on mixers can be had reasonably today, and they open up a world of possibilities for integrating stuff into your system, beyond the obvious sonic benefit.

Apparently G. Harrison said "Why would you want to recall a mix if you didn't like it in the first place?" :)

[beubbo]

Hi there,

May be a dumb question... in an all ITB approach could anyone know if it makes sense to make stems in DP like those who have a summing box do BUT to sum the said stems in a DAW that has a 64-bit mix engine (like Tracktion2...) ? any gain or something ?

Could a 64-bit mix engine make worthless the -6dBFS cautious way ?

thanks a lot in advance for any enlightment