good fader positions if levels are slammed?

[stealthbalance]

hi all !

just wondering what's the best way to deal with this. we all like to have our faders sitting physically in a handy position to do our level rides and work those dynamics right? but what do you do if the printed levels are so hot that you are forced to keep the fader positions so low that it is a problem to mix easily. you could adjust the first plug in's output level down 20 db but that seems demo league to me. or you could audiosuite the gain down on the files themselves but that doesn't seem right either. on ssl's i would always put my faders in a straight line at fader position -10 or at 0 or whatever and it was great. but in the box, the only way to do that is putting the fader in trim/write which i find a bit glitchy and not reliable (often after a fader write in that mode the fader plays back at a much lower level or just wrong -or id keep all my faders in trim write). i wish i could keep all faders at a set position and after doing a ride it would snap back to that delta position ( like it would in touch mode ). any suggestions fellow mix slutz??
thanks so much !!
s

[littledog]

Are you mixing in Pro Tools? According to a number of people in a previous post, just lower your master fader instead of the individual channel faders. Evidently the math is exactly the same.

If you are sending stems to an analog board, you can group each stem to a stereo aux return, and lower the aux gain. Or else insert the TRIM plug-in (which I believe adds no extra latency) on each individual track.

[stealthbalance]

thanks littledog

i am mixing hd4 all in the box. the thing is that by the time that you
beef up lifeless tracks with eq and bottom end and compression, the
faders are so far low phisically that it is imposible to do any rides. i wish there
was some way to set the value of the levels other than the output calibration
after PT. plus i might be wrong but i don't feel that its always important to
record as hot as possible in PT. im aware that the hotter you record the more
bits you take up and stay 24bit and if you record with low levels its more like
12bit or whatever - but it makes everything for me so unmanagable. i guess
what i really would love is an ssl vca mode so after your first pass of auto
you could put a delta value anywhere - with automation punch in and to be
able to stop and wind back a touch and punch into the same pass - and not
be forced to to keep it or undo. i guess that's for another thread.

[wolfhound]

Why not just adjust the input or ouput on the plugs? I believe they're calculating at a higher bit-depth than the faders anyway.

I think?

[Jack Ruston]

Originally posted by stealthbalance


1. the thing is that by the time that you
beef up lifeless tracks with eq and bottom end and compression, the
faders are so far low phisically that it is imposible to do any rides.

2. i wish there
was some way to set the value of the levels other than the output calibration
after PT.

3. plus i might be wrong but i don't feel that its always important to
record as hot as possible in PT. im aware that the hotter you record the more
bits you take up and stay 24bit and if you record with low levels its more like
12bit or whatever - but it makes everything for me so unmanagable.

Hi Stealthbalance. Hope all is well. I have numbered your points so that I can reply without having to copy loads of quote marks. Hope you dont mind!

1. It is important to match the plug ins output level to the input so that you can hear in bypass exactly what the plug is doing. If you let the volume creep up which it will do, everything just sounds better when you switch the plug in in. I would seriously recommend matching levels to give yourself the best chance of a totally objective decision about the benefit of the plugin and its settings. If the intention is to raise the level, then fine but that doesnt sound like your problem just now!

2. I think you could hapily just attenuate at the plug in ouputs. After all these plug ins are mostly designed to accept a -18dbfs average level (so perhaps peaking -10). Some plug ins will overload badly with a hotter signal than that at the input, and you need to pull the input level down.

3. You are not wrong. In fact the quickest way to a dodgy sounding 'in the box' mix is to let the levels start to creep up. The system is designed to operate at around -18dbfs calibration level. Assuming for a moment that you might see peaks some 8dbs above 0vu, that should leave about 10dbs of clear air above the peaks. This is crucial to getting a natural sound out of software. If you dont have this headroom things start to get cloudy and messy. And besides its nice to sometimes push a fader UP instead of always pulling em down. Plugins start to overload, and the mix bus gets it's knickers in a twist. Some people believe that you should leave equally large amounts of headroom on the master fader too. Obviously this is not a great way to get a loud cd but thats the mastering house's job. So why might things sound bad...well Nika Aldrich's recenty posted white paper discusses the fact that the metering in most DAW's (including PT) meters the samples themselves. Once the signal is sent to the DA converter (either in PT or on a cd player) a reconstruction filter is applied to rebuild the wave that should exist between the amplitude points of these samples. By definition the wave will always be higher in some places and lower in others. Studies have shown that in extreme cases the wave can end up more than 6db over the highest sample, and so the meter. Compressed material is particularly prone to this. Nika is not saying that you MUST leave 6db of headroom on a cd, but a generous margin must be considered or it's digital harshness time. The way to avoid this problem is to insert a reconstruction filter into the path of the meter (a la new Trillium Lane Labs meter plug in). Mastering houses tend to have this type of meter in use as far as I can discover. Working at 96k will somewhat reduce the likelihood of this occurring due to the reduced energy in the higher frequencies but it is still totally possible.

Regarding using up all the bits, the ONLY consideration in this regard is your noise floor. If the signal is above it, you are in the clear...The audio will sound perfect. Dont go recording on any runways! Leave bags of headroom. Always.

Hope this is of some use.

Jack

[Jack Ruston]

Regarding point one, I understand if you might not have wanted to take my word for it, but Bob Olhsson said this....

"I've gotten good results using the PSP however it's really easy to go to far with it and you need to always check everything at perfectly matched levels to make sure you aren't choosing worse sound that's a half dB louder rather than simply turning it up a half dB"

And you cant argue with Bob!

J

[faeflora]

Good question! It bothers me too and I usually end up doing semi cheesy workarounds like using gain reduction plug ins.

Personally I try to get my peaks in around -10 on the master buss. It sounds better like that than if I get closer to digital 0. I pull it closer to 0 during "mastering" with analog compression.

[alphajerk]

Originally posted by stealthbalance
im aware that the hotter you record the more
bits you take up and stay 24bit and if you record with low levels its more like
12bit or whatever - but it makes everything for me so unmanagable.

i think that statement is wrong. you might be only USING 12 of the 24 bits but the resolution is still 24 bits. it doesnt matter how hot you record as long as you hit your peak to floor window.

[littledog]

alpha is correct

[Frost]

Originally posted by stealthbalance
im aware that the hotter you record the more
bits you take up and stay 24bit and if you record with low levels its more like
12bit or whatever - but it makes everything for me so unmanagable.
------------------------------------------------------------------------


i think that statement is wrong. you might be only USING 12 of the 24 bits but the resolution is still 24 bits. it doesnt matter how hot you record as long as you hit your peak to floor window.
------------------------------------------------------------------------

Its just a matter of semantics. Recording low enough IS only using some of the bits. Therefore the recording is equal to a 12 bit recording. However, in 24 bit, -48 on the digital meter is equal to 16 bit. Are you really recording at -65? If your recording closer to -4 or 5 your using more than enough bits to make a good recording. Heck, -10 or -20 is probably enough to be just fine.
Frost

[stealthbalance]

thanks so much gearslutz !
great tips. i am always careful to match levels so that i can bypass the
plug-in and a/b clinically - always. i wish mastering engineers would
set that level matched volume for their client instead of the smoke and
mirrors playing their mastering 5 db louder than my mix flat
maybe the best thing for me is to use that trim plug-in on the first position
insert. or i wonder if id hear a change in quality if i audiosuite the tracks
down to a lower level. whatever happens this should be an easy ajustment-
hopefully digi will add some cool automation features in future software
versions. maybe i could try keeping my whole stereo buss level down -10 db
and then on the very last stereo buss plug i use ( like the inflator or L2 )
then crank the output on that for best sound ( or is that the same as
slamming the stereo buss anyway??
i think its time for a drink and a nice lie down.
thanks again all !!
s

[Jules]

I would lower it with the plug in out put level.

I frequently do this to put over hot or over quiet levels on my Pro Control faders to a convinient level for:

a) Visual reference
b) the best 'fader action' zone (ie Zero) for ease of use / fine tweaking etc

I put the fader on zero then listen with my hand on the plug in output level, and mix it in to taste....

[davemc]

Gain staging with Pro Tools is something to watch out for
A lot of people slam into PT, then through the plugins and sends.
Then think pulling down the fader afterwards will fix this.
I always check meters pre and post fader..And double check that one plugin is not clipping the next...

If I remember someone said (maybe a myth) PT does some soft limiting on the faders etc so it does not digital clip until extremes?
I could be a wrong with this please correct me if you want

I often think bad gain staging through busses and plugins has a bit do do with how PT has a bad rap for sounding shit.

So if you have to use the trim on the last instert, one of the plugins beforehand is getting hit hard or has a high output.

Just turn your speakers up if the level is too low
Or mix with a stereo comp or limiter on if it annoys you

[jazzius]

Originally posted by davemc
Just turn your speakers up if the level is too low

man, then you better hope your 'puter doesn't decide to drop full-scale digital grot over you D/A........

[Wiggy Neve Slut]

Im new to the whole PT/ digital recording thing but in my limited experience recording to digital (PT) it is certainly different than recording to Sony Dash 3324.... man what a piece of shit! Thos machine should be called bit grinders as they sound crap and have AD/DA's like abrasive sandpaper! You really had to peg them to get any decent levels to tape... now with my recent PT sessions i was slamming WAY to hard... and it sounded Fuct.. what ever you like any medium proper gain staging is important and even more so in PT and digital in general as there is no inherent 'safety net' of tape compression etc.

PEACE
Wiggy

[Jack Ruston]

It sounds so simple but its always easy to forget the calibration. You wouldnt send a plus 18db signal to a tape machine so why send it to pt?

[stealthbalance]

Originally posted by Jules
I would lower it with the plug in out put level.

I frequently do this to put over hot or over quiet levels on my Pro Control faders to a convinient level for:

a) Visual reference
b) the best 'fader action' zone (ie Zero) for ease of use / fine tweaking etc

I put the fader on zero then listen with my hand on the plug in output level, and mix it in to taste....

jules - that's exactly what i do and it works great. i was just trying to figure out
a more technically correct way of doing it as this does take away my ability to
a/b without level change - that is unless i just use an additional trim plug or any plug on insert #1 to cut or boost the level.
thanks again slutz
s

[wolfhound]

Curious to know what you guys think about this:

It seems I hear a lot of people talking about running the levels on individual channels in PT really hot, ignoring the clip indicators, citing that it's almost impossible to clip the output of the PT master fader cause its calculated at a higher bit-depth (at 56-bit?), but what I don't understand is how they're not clipping the INPUT to the master fader. If the sum of all your channels equals really hot, it doesn't matter how low you set the master fader it still shows over because the input's too hot. So how do people mix this way? Just ignore it? Opinions?

I wonder the same thing about the analog summing boxes for sale? Do they just attenuate the signal coming in on each channel?

[Jack Ruston]

Thats a scary post wolfhound )

OK. Recording hot into PT with a default -18dbfs calibration (which is about all you can get on the new 192) has a plethora of associated problems and about zero benefits.

1. Basic gain structure tells us to observe zero vu. To achieve a +18db peak from our pre suggests that either we are heavily overloading on the way in or boosting a ludicrous amount on the way out. Either way...why? These things are capable of operating like that, but they dont generally sound their best.

2. So you come at that level into pro tools. You hit the converter really hard, then the plug ins really hard, the channel really hard and the mix bus really hard...all that stuff about not being able to overload the mix bus may be true, but it doesnt sound good. Again it might be able to cope, but I would suggest that the damage is already done.

3. You then hit the DA hard on the way out. Even if you are watching the peak lights on the meters, it is possible to have overloads occurring which are invisible to the meter. If you go straight to cd you are going to have it even worse when the cd player converts (if its anything like most domestic cd players)

There are no good reasons to do this. No advantages.

Regarding analog summing its slightly different. Many analogue devices can cope with seriously hot signals in a way that digital cant. Its not a good idea though. You can obviously attenuate with the trim on each channel, but if the signals are that hot, you can drop ten dbs off them and still find that the monitor level is rather loud considering the main pot is only at 9 oclock. Again observing zero vu is the key.

Whatever you hear, digital overloads just dont sound good. They are cumulative too. One tiny element of a track isnt so bad but when a few tracks are clipping, then the mix bus and then ultimately the master, the whole thing turns harsh. No wonder vinyl purists think cd sounds bad. Its not the cd its the level!

In a world where people are so keen to reach for the 'level three mega bass' button on their miserable hifi, they seem to have completely forgotten the function of the VOLUME control, and expect this to be provided by the cd itself.

I'm now in a fury of puritanical zeal and must certainly pop up the pub for a medicinal pint.

Have a good one everybody

Jack

[wolfhound]

I agree with everything you said, yet I keep hearing good people say that's how they get a good mix in PT. Now I don't think they RECORDED with digital clipping, just mix with faders real hot.

Part of their point, maybe, is that the red indicators in PT don't actually mean that clipping has occured (or is sonically apparent anyway).

Just thought maybe somebody good give an explanation of why that would be a desireable way to work....I'll stick with observing normal levels for now

[Jack Ruston]

Originally posted by wolfhound
I keep hearing good people say that's how they get a good mix in PT. Now I don't think they RECORDED with digital clipping, just mix with faders real hot.

Well if they've recorded at the 'correct' level they should be seeing peaks of -10 which would very probably be ok if boosted the 6db PT will allow.

Originally posted by wolfhound
[/B]Part of their point, maybe, is that the red indicators in PT don't actually mean that clipping has occured (or is sonically apparent anyway).
[/B]

No. Much the opposite. Most meters accidentally allow peaks through. This is not a new discovery. It is the reason why oversampling meters were developed. Also the practice of putting reconstruction filters infront of meters...which while not common in the main DAW's is still a feature of some of the industry standard mastering apps.

J

[faeflora]

Originally posted by jazzius
man, then you better hope your 'puter doesn't decide to drop full-scale digital grot over you D/A........

I hate the grot as well. It hurts.

[Jay Kahrs]

Did anyone read the interview with Chris Verrea (is that his last name? the NIN guy) in this months issue of EM? That whole first page should be included with every DAW that's sold. It took me about two years to figure out what he was talking about.

And FWIW, I still prefer tape.

[wolfhound]

Yeah, his view made sense to me, and that's the way I try to do things (in terms of recording, I'm not mixing through a board like he is). His view was consistent with what Sly and others were saying: keep the analog gear in its sweet spot level-wise even though "low" in PT.

What I didn't get (cause I don't know any of the technicalities of it all) was why his O2R would have any more headroom on the mix bus than PT. He was sending each channel of PT out digital to an O2R to mix, which he said allowed him the headroom to leave his faders in PT at unity, when otherwise they were overloading the mix bus in PT.

[Jack Ruston]

I'm wondering if the red clip lights on the interfaces themselves indicate the level after the reconstruction filter. I think maybe they do because I've had situations with mixes where the level looked OK in PT but showed overloads on the hardware. I ignored it. Maybe that was not so clever in retrospect. Still maybe this is a little safety net for that metering problem.

J