Using tones to calibrate buss compression...?

[absrec]

Ok. I'm confused at why you would get so scientific about this. I use test tones to get my faders to read 0dB on my VU meters so that when I recall a mix, it'll sound the same as the last time I pulled it up. I use my console as a big summing box, really.

The only time I've ever done what you're saying is if I'm compressing a stereo buss with a pair of mono compressors and I need to get the levels exact and then control the threshold with a fader in ProTools. I've done that.

Now, when using an 1176, you have to keep in mind that the meters are approximate. I don't think they are very consistent from unit to unit. I hate to say it, but you just gotta listen first, look at your meters second and then trust what you're hearing.

I document a good deal of outboard settings in ProTools, but never would I calibrate a compressor with test tones without first hearing the approximate range that I want the compressor to reside in. You gotta play with the compressor to find the sweet spot. Every compressor has one and it's smaller on some and larger on others. I try to get all my faders in ProTools between -10 and -4 (usually) so I have some play for rides, et cetera. I takes a while to figure out how hard to hit your console so that you'll be in the sweet spot of your buss comp, give or take a couple dB. I spent a good amount of time adjusting my line trims so that everything would jive with PT as well as my mix chain. Makes life a lot easier when you need to just get things done.

Good luck,
-Aaron

[Worlez]

Thanks for the reply.

I'm setting up the compressors before I begin the mixes, so that I'm mixing into them. Just like mixing into a 2buss, only I'm doing it with 4, for different parts of the mix.

I've been calibrating them so that all the outputs are returning the same in Pro Tools, but I'm just wondering how other people approach it.

When you do send tones, are you sending Peak or RMS signals? What level are you sending them at? (I'm looking for ITB strategies here btw)

Cheers, take it easy.

Al

[robot gigante]

I think that even if you have basic settings that work ok to start out with with, you may still probably want to change your compressor settings as you go even when you start out mixing through them. I know I do. I usually check a few times to see what's working and what isn't as the mix progresses and what's feeding my busses changes.

I don't use test tones, I use my (gulp) ears, except for making sure two mono comps are the same level like absrec does. I'm sorry but I don't see how any other way makes sense. I just make sure nothing's clipping.

Maybe I'm misunderstanding what you're doing, but how can you predict your what your attacks, releases, compression ratios (and therefore amount of compression and the envelope) and so on are going to react to dynamic and different material and how this is going to sound with test tones???? Ears and checking/tweaking as you go is the only way that works for me, which not to say that I'm absolutely right and know everything about this...

Plus, you might want to push those bus faders up and down a little as you go as well, right? Automate them and so on. So what's the big deal if the compressor's output is a little hotter or quieter than what's going in so long as you're not clipping?

[Worlez]

Originally Posted by robot gigante

Plus, you might want to push those bus faders up and down a little as you go as well, right? Automate them and so on. So what's the big deal if the compressor's output is a little hotter or quieter than what's going in so long as you're not clipping?

I've been doing it so that I can balance my 4 (now 5) 'sub-masters' - 4 with compressors, and a 5th dry for FX busses etc.

Nobody else does this?

Cheers

Al

[Jim vanBergen]

I do the same thing with 0dB tones (+4, or 1VRMS) input & output. Set it up 1:1, check it at 2:1, and then mix to taste.

[Jules]

Are you sure you arent getting any wacky phase issues with all the different plug ins?

cause I did when I tried it, I had to abandon the idea it came out so bad..

And I had the plugins all time aligned to the nearest sample too (Mix + rig)

The artifacts / differenced caused by each compressor processing cause phase mayhem..

I'm not stiking the knife in - just baffled how I didnt get it to fly and others have... (and good luck to you! )

What is your plug in set up if you dond mind shareing?

Thanks in advance...

[recall]

michael brauer does this according to the interview he did in tapeop a while back, so you 're in good company, i don't myself but then i tend to mix itb by necessity.

[robot gigante]

Latency compensation (such as with PTHD) does't seem to completely fix the issues that Jules brings up, either... FWIW.

[MHB850]

Originally Posted by robot gigante

Latency compensation (such as with PTHD) does't seem to completely fix the issues that Jules brings up, either... FWIW.

My MB approach, when trying to combine both multi buss compression and the use of compression auxes, doesn't have very good results in the box. mike Caffery came by the studio the other day and told us how he gets it to sound good but he didn't take into account the compression auxes and some of the combinations that I use. When you add that to the session template it creates delay/phasing issues that Latency compensation can't handle.... YET.

In the analog world the latency issue is simply...not an issue, it's all fairly musical. In the digital world with life being a 1 or 0 latency translates to errors which comes off as unmusical. At worst, the side effect is comb filtering. As you guys already know, most of the time it's an unappealing change in the mid and low tones of an instrument, Bass and guitars usually the ones suffering the most.

Because I incorporate so many parallel and multi combinations, it's a sonic nightmare for the computer....sorrrrry.

michael

[Worlez]

Hmm, maybe my ears aren't as hot as yours. So far, I've never noticed any major phasing issues or artefacts...running PTHD 7.1 in a 5 buss approach (low, mid, high, warmth and dry - eaching having a compressor plugin, and occasionally an eq on there), along with paralell compression of various groups (kit, kick, snare, vox etc). Using McDSP, Bombfactory, Waves SSL, URS and Smack! mostly.

I used to do it on LE as well, but would put dupe plugins on all the tracks that were causing delay. My drum mixes in HD definately sound tighter...

Al

[studjo]

Worlez do you even use aux-sends as compressor-feeds?
Because I think what Michael adresses is a discrepancy from the ordinary group buss to a aus send buss - if that makes any sense to anyone

ok back to German


Jo

[MHB850]

Originally Posted by Worlez

Hmm, maybe my ears aren't as hot as yours. So far, I've never noticed any major phasing issues or artefacts...running PTHD 7.1 in a 5 buss approach (low, mid, high, warmth and dry - eaching having a compressor plugin, and occasionally an eq on there), along with paralell compression of various groups (kit, kick, snare, vox etc). Using McDSP, Bombfactory, Waves SSL, URS and Smack! mostly.

I used to do it on LE as well, but would put dupe plugins on all the tracks that were causing delay. My drum mixes in HD definately sound tighter...

Al

Phasing is phasing, i doubt your ears aren't as hot as mine. Maybe you have it figured out, I don't know. I'd have to see and hear how your doing it to be sure it incorporates all my permutations. I assume you are also using aux sends as compressor feeds.

My assistant and I are in the process of formatting our own P.T. template that exactly duplicates everything I do on the SSL 9KJ. I'll know for sure if there are any issues once i get into it. We're testing it with a simple metronome plug in to start. Once the click track passes all my tests, we'll go to the next step of mixing a song. Since I don't mix in the box, it's gonna take me longer to find the time to work out the kinks.

michael

[Mike Caffrey]

I think you can set up the auxes using the same technique as the busses to avoid latency.


As far as test tones, because of some of the issues raised in a few post, like how do you know what you want in advance - I think the best way to start is without tones and then use them at the end to see what your ear likes. Then eventually, you'll know how to set up in advance.

Or, you could argue that you know what you want in advance - otherwise, why would you be sending it to one particular buss? You have a sound or a fucntion that you want, so you set up a buss to work that way and then you send to it.

As far as attack and releast times, I do know at the beginning. If I'm using a Distressor on the drums buss, I know the attack, release and output level. I could also set as input level and vary everythin with how hard I send from the faders, but I choose not to.

I know the settings for LTF2 on the guitar buss. I know one or two options for the 6386 or Phoenix, depending on where they're used.

You can always set up with tones, then make changes and re-balance.

[Worlez]

Originally Posted by MHB850

Phasing is phasing, i doubt your ears aren't as hot as mine. Maybe you have it figured out, I don't know. I'd have to see and hear how your doing it to be sure it incorporates all my permutations. I assume you are also using aux sends as compressor feeds.

My assistant and I are in the process of formatting our own P.T. template that exactly duplicates everything I do on the SSL 9KJ. I'll know for sure if there are any issues once i get into it. We're testing it with a simple metronome plug in to start. Once the click track passes all my tests, we'll go to the next step of mixing a song. Since I don't mix in the box, it's gonna take me longer to find the time to work out the kinks.

michael

Thats a very good point actually Micheal - thinking about it, I'm just multi-bussing, I'm not using aux sends to feed compressors. My parallel compression returns all share the same returns as the groups, so for example I'll have both a dry and a slammed stereo drum buss, both being fed from buss 3-4. I only really use auxes for verbs/delays etc.

I'm going to try using auxes when I get down to the studio today and see what happens. Thanks for bringing this up

Now regarding the test tones - I'd really like to hear specifically how people do this...my way is to send a -6db tone in PT, adjust for the amount of compression and then set the output of PT to return at -9db, so that I have enough headroom to not clip the busses' output. Not sure if this is the most efficient way really though. I do know ballparks of what I like in advance, such as attack/release times, compression ratios etc. but I'll tweak as the mix goes on. But its the initial gain staging that I want to make sure I'm doing well.

Thanks for everyones advice so far.

Al

[MHB850]

Originally Posted by Mike Caffrey

I could also set as input level and vary everythin with how hard I send from the faders, but I choose not to.

Aah, now we finally get to the center of why some choose to set tones first and some don't. The whole way this works for me depends on finding the sweet spot where the desk and the compressors are sounding their best. All my stereo buss compressors are calibrated to read "0". (how i set up the compression for the calibration is described on my site www.mbrauer.com in the Q&A section.) I ride the faders to hit the sweetspot of the compressors, not the other way around. I already know what each stereo comp has to offer since i've set up different attack and release times for the 4 different busses to address everything from fast to slow transients.

I am always looking at my four buss compressors during the time that I get the mix up and rock'in. If I happen to be hitting the compressors to hard with the drums but I like where they're sitting in the mix, then i'll go over to the comps and back off on the attack time.

I would go crazy if I changed, for every song, the inputs, outputs, attack, release, and not be sure if one side of the stereo is working harder than the other(since I don't link) when I'm sending a mono signal. For me, that's way too complicated and it makes recalls a drag. Not to mention that I don't think that way.

I've spent years mixing this way so I know how to live in the land of compressors all being post fader. The best way I can describe the image is, it's like pushing into a thin rubber sheet. If you push too hard the sheet will resist and there won't appear to be anymore level available but if you ride it just right, it becomes very animated and alive.

michael

[MHB850]

Originally Posted by Mike Caffrey

I think you can set up the auxes using the same technique as the busses to avoid latency.

What if I want to return an aux compressor to both a processed and unprocessed stereo bus?

michael

[ronzie]

I seem to be "internet stalking" MHB & Mike Caffrey and the threads they're contributing to; sorry guys !

It took me a minute to desypher the issues and the replys, and what MHB & Mike said are true.

Here's the deal as I see it: You want "proper" gain structure for your mixes. Not a bad idea. You have plugs you dig/use. Cool. Your setting your buss comps to levels you think "might sum" to a useable stereo mix via a tone. Fine. BUT, it seems to me, you have no point of reference both level and power wise as a starting point to find out what works for you.

In ANALOG land 0vu = 1.23vac=+4db (simplified, I know, I know).This is THE reference for Pro audio.

In DIGI land you make the call depending on YOUR experiences/style/phobias/or WHATEVER the box gives you. this could be -20,-18,-16,-14 DBFS (among others). Unfortunetly the standard is NO standard. You want 0 in/0 out

Pick a point (-18's fine for starters) and go from there. That is 1.23vac will reflect -18dbfs on the input and output of your rig....IF YOU ARE ALIGNED... Obviously, you need to get in & out of the box at some juncture and you might as well marry your digi levels to the analog standard so you have a basis to determine what works for you wether it's -20 or -14dbfs

What is your rig aligned/defaulted to (-20, -18 dbfs etc...)?
Do you have an outside oscilator and multi meter to double check whats happening in the box?
What does this have to do with setting up buss comp? EVERYTHING....if a TONE is the road your taking in trying to figure out gain structure. It absolutely helps. And its REPEATABLE.

Ron Allaire, Skyline

[robot gigante]

Originally Posted by Worlez

Thats a very good point actually Micheal - thinking about it, I'm just multi-bussing, I'm not using aux sends to feed compressors. My parallel compression returns all share the same returns as the groups, so for example I'll have both a dry and a slammed stereo drum buss, both being fed from buss 3-4. I only really use auxes for verbs/delays etc.

I'm going to try using auxes when I get down to the studio today and see what happens. Thanks for bringing this up

Aha- when I tried it it was by using auxes, and there were indeed issues when done ITB (w/ latency compensation).

But that's OK, I like hardware compression much better anyway.

[MHB850]

Originally Posted by robot gigante

But that's OK, I like hardware compression much better anyway.

Amen on that!

[Mike Caffrey]

Originally Posted by MHB850

What if I want to return an aux compressor to both a processed and unprocessed stereo bus?

michael

It really depends on what you're sending and where else it's going.

The return of the aux compressor is easy, set it up just like when you send to a aprocessed and unprocessed buss.

Where it gets complicated is if your sending to the processed and unprocessed buss in addition to the aux.

Let's say snare on a send/return to an 1176. The snare is also going to B and the stereo and you want the 1176 to B and the stereo.

Duplicate the snare track. Put an 1176 (bypassed) on the original track and assign the duplicated track only to the aux send. Assign the aux return to the B and stereo simultaneously.

You might be able to do it with two returns fed by the same send - one processed and on unprocessed (meaning processed and bypassed).

That will be the way to do your vocals. You'll probably want to identically process the send faders, but with opposite processors bypassed incase you want to blend them rather than cut between them.

Each of their sends will feed two returns if you want processesed and unprocessed.

Finding AWA and Federal plugins is going to be the hard part.

[Mike Caffrey]

Originally Posted by robot gigante

Aha- when I tried it it was by using auxes, and there were indeed issues when done ITB (w/ latency compensation).

But that's OK, I like hardware compression much better anyway.

How are you avoiding comb filtering caused by latency when you have a dry and slammed buss?

You just have to do the same thing on the aux.

[Mike Caffrey]

Originally Posted by MHB850

The best way I can describe the image is, it's like pushing into a thin rubber sheet. If you push too hard the sheet will resist and there won't appear to be anymore level available but if you ride it just right, it becomes very animated and alive.

michael

Like jumping upwards into a trampoline above your head. That makes a lot of sense.

With that anology, how is a first timer going to set the height of their rubber sheets with tones? Maybe the sweet spot where you get the drums pumping and the sweet spot where you get the guitars pumping leads to one of them being too loud in the mix? Then when you pull it back, you loose your sweet spot for the pumping, so you have to change your outputs if your compression is exactly right (or send lower volume traks the the 1176 pair set up as outs).

I do't see how someone could know what the want to see with tones if they haven't done a few mixes that worked and then ran tones through them to see what they did and what they need to do again next time.

[Worlez]

This thread is getting very interesting, thanks a lot lads.

RE the phasing/latency - I am yet to hear this in my approach to multi-bussing. It occured to me that I've not actually been using aux sends to feed compressors - I've been sending things to multiple busses instead, and using delay compensation, I haven't noticed a problem yet. Can anyone see any potential flaws in this?

Maybe part of that phasing is something I/we compensate for when we are mixing? I always like how MB describes mixing the drums to Yellow, where blending some of that phasing back in with the dry drum track gave it that fantastic sound!

Now regarding what Mike Caffrey says about the height of the trampoline - very interesting point, and I like the analogy. What I'm now wondering about, is lets say I'm constantly adjusting my compressors across my 4 processed busses, would that not then jack up all my FX returns that are feeding the dry buss? If I was adjusting the volume of my middle buss for example, to lower it, wouldn't that mean that my ratios of direct to wet (reverb etc) be screwed? Or am I thinking about this too much?

Al

[Jules]

Originally Posted by Worlez

Maybe part of that phasing is something I/we compensate for when we are mixing? I always like how MB describes mixing the drums to Yellow, where blending some of that phasing back in with the dry drum track gave it that fantastic sound!

No, that is definitely a 'special effect' sound - a cardboard tubular type of sound (comb filtering)

[MHB850]

Originally Posted by Mike Caffrey

It really depends on what you're sending and where else it's going.

The return of the aux compressor is easy, set it up just like when you send to a aprocessed and unprocessed buss.

Where it gets complicated is if your sending to the processed and unprocessed buss in addition to the aux.

Let's say snare on a send/return to an 1176. The snare is also going to B and the stereo and you want the 1176 to B and the stereo.

Duplicate the snare track. Put an 1176 (bypassed) on the original track and assign the duplicated track only to the aux send. Assign the aux return to the B and stereo simultaneously.

You might be able to do it with two returns fed by the same send - one processed and on unprocessed (meaning processed and bypassed).

That will be the way to do your vocals. You'll probably want to identically process the send faders, but with opposite processors bypassed incase you want to blend them rather than cut between them.

Each of their sends will feed two returns if you want processesed and unprocessed.

Finding AWA and Federal plugins is going to be the hard part.

That is just way too much work and it get's too complicated. It opens up a can of worms. You do what you want. if it works for you, fine. For me, it's just not that simple a thing in the context of mixing a song, it seems like very much of a distraction from concentrating on getting a mix to feel great. I think everyone understands at this point what the conditions are. I don't think there is any need for me to drag this one out.

[MHB850]

Originally Posted by Mike Caffrey

I do't see how someone could know what the want to see with tones if they haven't done a few mixes that worked and then ran tones through them to see what they did and what they need to do again next time.

I can, I have, I do. Maybe it's because i've been setting up gear with tones since 1977 and i have a good imagination?

Michael

[Mike Caffrey]

I think the years of experience is the key, which is why I think people need to start without tones first. Once they're in the ball park, then run them for stereo balance and note taking for ultimatine finding their patterns and taste.

[robot gigante]

Originally Posted by Mike Caffrey

How are you avoiding comb filtering caused by latency when you have a dry and slammed buss?

You just have to do the same thing on the aux.

I think you're right. I just hadn't tried it the other way, and the poster had mentioned that he wasn't hearing issues. I just was willing to entertain the thought that latency compensation might work better in PTHD if you're using busses and not aux sends (since I always use auxes). But- somehow I doubt it now that I think about it.

[Worlez]

Heh, now I'm confused - does you get latency/phasing issues ITB when multi-bussing, or using two returns (dry and processed) from the same buss? Or is it just when your using auxes to send to your busses? I've never noticed anything before, but maybe its something so obvious that I haven't been able to see past my nose and spot it! That said, I just use busses to feed my returns when it comes to compression, very rarely do I do it via auxes (if ever).

Cheers, thanks for the input so far.

Al