Gain Staging

Page: < 12 Showing page 2 of 2
Author
Post
Essentials Only Full Version
Bristol_Jonesey
Max Output Level: 0 dBFS
  • Total Posts : 16775
  • Joined: 2007/10/08 15:41:17
  • Location: Bristol, UK
  • Status: offline
Re:Gain Staging 2011/06/17 06:01:15 (permalink)
Thanks for that Jeff, much appreciated. I use the BlueCat plug and have it built into my master buss on all of my templates.

If we concentrate for a moment solely on the Sonitus compressor (not the multiband), my concern was those rare occasions when an incoming signal might be a bit on the hot side thereby making it difficult to decide where the threshold should be set. (It's analogous to mixing with your track faders near the bottom of the range where there's not enough resolution for fine control)

I was asking if using trim to lower the signal before it hits the compressor would be standard practice to enable the threshold to be set much more accurately.

Seems I've stimulated some debate here - all of it of the good kind of course!

CbB, Platinum, 64 bit throughout
Custom built i7 3930, 32Gb RAM, 2 x 1Tb Internal HDD, 1 x 1TB system SSD (Win 7), 1 x 500Gb system SSD (Win 10), 2 x 1Tb External HDD's, Dual boot Win 7 & Win 10 64 Bit, Saffire Pro 26, ISA One, Adam P11A,
#31
Jeff Evans
Max Output Level: -24 dBFS
  • Total Posts : 5139
  • Joined: 2009/04/13 18:20:16
  • Location: Ballarat, Australia
  • Status: offline
Re:Gain Staging 2011/06/17 06:52:48 (permalink)
Hi Bristol I would not hesitate to do as you suggest. If I felt the incoming signal was so hot that it made setting the threshold difficult then you could trim it down to make setting the threshold easier and then makeup the gain (if required) with the makeup gain.

But I must say I have not encountered this situation too often. I would be concerned as to why a signal was so loud entering the compressor. If there is a plugin before then perhaps the output of that could be trimmed back some. Actually it is when we might have several plugins in series to form a chain that we need to be mindful of gain staging in and out of each of these stages.

I think that is why some plugins have an input and output control. More often than not we would not be adjusting the input control but there can be occasions when it may be required.
post edited by Jeff Evans - 2011/06/17 06:55:07

Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface 
 
Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
#32
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re:Gain Staging 2011/06/17 07:49:19 (permalink)
Bristol, IMO you should feel free to use any of the controls that have been provided.

Sometimes making an adjustment creates implications in other places that need to be adjusted in response... the more you use any and all adjustments the more adept you will become at predicting what effects any changes you are making may have throughout your routing.

all the best,
mike



#33
Jonbouy
Max Output Level: 0 dBFS
  • Total Posts : 22562
  • Joined: 2008/04/14 13:47:39
  • Location: England's Sunshine South Coast
  • Status: offline
Re:Gain Staging 2011/06/17 08:36:51 (permalink)
Good, because there's nothing that I love more than twiddling all the knobs I've been blessed with.

"We can't do anything to change the world until capitalism crumbles.
In the meantime we should all go shopping to console ourselves" - Banksy
#34
Karyn
Ma-Ma
  • Total Posts : 9200
  • Joined: 2009/01/30 08:03:10
  • Location: Lincoln, England.
  • Status: offline
Re:Gain Staging 2011/06/17 08:50:05 (permalink)
Jonbouy


Good, because there's nothing that I love more than twiddling all the knobs I've been blessed with.


Anything to fill 30 seconds of boredom

Mekashi Futo. 
Get 10% off all Waves plugins.
Current DAW.  i7-950, Gigabyte EX58-UD5, 12Gb RAM, 1Tb SSD, 2x2Tb HDD, nVidia GTX 260, Antec 1000W psu, Win7 64bit, Studio 192, Digimax FS, KRK RP8G2, Sonar Platinum

#35
Jonbouy
Max Output Level: 0 dBFS
  • Total Posts : 22562
  • Joined: 2008/04/14 13:47:39
  • Location: England's Sunshine South Coast
  • Status: offline
Re:Gain Staging 2011/06/17 09:03:28 (permalink)
Karyn


Jonbouy


Good, because there's nothing that I love more than twiddling all the knobs I've been blessed with.


Anything to fill 30 seconds of boredom


It wouldn't take me that long because I'm an expert...

(sorry wrong thread forum)

post edited by Jonbouy - 2011/06/17 09:04:56

"We can't do anything to change the world until capitalism crumbles.
In the meantime we should all go shopping to console ourselves" - Banksy
#36
jm24
Max Output Level: -54 dBFS
  • Total Posts : 2127
  • Joined: 2003/11/12 10:41:12
  • Status: offline
Re:Gain Staging 2011/06/17 10:41:27 (permalink)
Bristol_Jonesey

my concern was those rare occasions when an incoming signal might be a bit on the hot side thereby making it difficult to decide where the threshold should be set.

In the analog world this would be "attacked" by placing a limiter before the compressor.
 
J
#37
Karyn
Ma-Ma
  • Total Posts : 9200
  • Joined: 2009/01/30 08:03:10
  • Location: Lincoln, England.
  • Status: offline
Re:Gain Staging 2011/06/17 11:16:15 (permalink)
jm24


Bristol_Jonesey

my concern was those rare occasions when an incoming signal might be a bit on the hot side thereby making it difficult to decide where the threshold should be set.

In the analog world this would be "attacked" by placing a limiter before the compressor.
 
J

In the analog world you'd turn down the channel input gain (trim) or even press the -20db pad button.
 
In the Sonar world you never get this scenario with live instruments (they're already been limited by the AD conversion) or from recorded tracks (as with live input) but you CAN get it from soft synths, as demonstrated multiple times.
 
 
The trim control is used to set the signal level going into the FX bin.
 
Simples.

Mekashi Futo. 
Get 10% off all Waves plugins.
Current DAW.  i7-950, Gigabyte EX58-UD5, 12Gb RAM, 1Tb SSD, 2x2Tb HDD, nVidia GTX 260, Antec 1000W psu, Win7 64bit, Studio 192, Digimax FS, KRK RP8G2, Sonar Platinum

#38
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re:Gain Staging 2011/06/17 11:32:59 (permalink)
I think jm may be referring to the practice of placing an 1176 limiter before a LA2A in a vocal tracking chain.

best regards,
mike


#39
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re:Gain Staging 2011/06/17 11:48:10 (permalink)
Karyn


jm24


Bristol_Jonesey

my concern was those rare occasions when an incoming signal might be a bit on the hot side thereby making it difficult to decide where the threshold should be set.

In the analog world this would be "attacked" by placing a limiter before the compressor.
 
J

In the analog world you'd turn down the channel input gain (trim) or even press the -20db pad button.
 
In the Sonar world you never get this scenario with live instruments (they're already been limited by the AD conversion) or from recorded tracks (as with live input) but you CAN get it from soft synths, as demonstrated multiple times.
 
 
The trim control is used to set the signal level going into the FX bin.
 
Simples.


This.

You just use the trim (AKA Input Level) control to set the level.

The point about Sonar's engine is because it uses floating point math, it doesn't really care about levels:

Too hot signal (even it's way, way, way over 0dB) - Sonar doesn't care, as there's ridiculous amounts of headroom above 0dB.

Too low signal (even it's way, way, way below 0dB) - Sonar doesn't care, as digital quantization errors do not increase at lower levels under floating point.  

Anywhere in between - Sonar doesn't care.

So you won't clip Sonar if you signal's too hot and you won't hurt signal to noise ratio if it's too low. And Sonar itself won't sound any different if you mix things at any one level vs. at any other. There is no "ideal" level here - because the math used is identical across more than 1,500dB of dynamic range.

Therefore, and this is the key point, since Sonar itself doesn't care, you can just set its levels wherever best accommodates the things that absolutely do care - like certain plugins or your output - and not worry about it.

 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#40
Sonico
Max Output Level: -85 dBFS
  • Total Posts : 266
  • Joined: 2003/11/25 12:53:58
  • Status: offline
Re:Gain Staging 2011/06/17 13:30:45 (permalink)
drewfx1



Therefore, and this is the key point, since Sonar itself doesn't care, you can just set its levels wherever best accommodates the things that absolutely do care - like certain plugins or your output - and not worry about it.



I totally agree with the above statement and would like to add something that may be a little bit off-topic but still I find very important to improve the clarity and punch of my mixes:

If we pay attention to the gain stage in the analog part of the chain (before hitting the converters) especially with prosumer gear, the preamps will work near the sweet spot and will sound it's best, and that way levels will not clip converters, plug-ins, channels, busses and master buss (given a good gain staging practice in the digital domain).


I use a VU meter plug-in and set it to -18dBFS = 0VU (+4dBu) and record with peaks of -18 to -12 dBFS.
Waves SSL plugins and classic compressors are my main mixing plug-ins and are calibrated that way (-18dBFS = 0VU) so I feel comfortable with that calibration and think I'am "setting levels where best accomodate the things that absolutely do care" my preamps and plug-ins.

Just an opinion!!! 


Desktop: Intel Core i7 3.60Ghz, 16 GB Ram Windows 10 64bit
Laptop: Intel i5 2.3Ghz, 6GB Ram Windows 10 64bit
Focusrite Scarlett 18i20
Presonus Faderport
My Music
#41
Karyn
Ma-Ma
  • Total Posts : 9200
  • Joined: 2009/01/30 08:03:10
  • Location: Lincoln, England.
  • Status: offline
Re:Gain Staging 2011/06/17 13:43:40 (permalink)
drewfx1

Therefore, and this is the key point, since Sonar itself doesn't care, you can just set its levels wherever best accommodates the things that absolutely do care - like certain plugins or your output - and not worry about it.
Like the first plug in the FX bin,  for which you use the trim control to set the level.


Mekashi Futo. 
Get 10% off all Waves plugins.
Current DAW.  i7-950, Gigabyte EX58-UD5, 12Gb RAM, 1Tb SSD, 2x2Tb HDD, nVidia GTX 260, Antec 1000W psu, Win7 64bit, Studio 192, Digimax FS, KRK RP8G2, Sonar Platinum

#42
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re:Gain Staging 2011/06/17 14:26:28 (permalink)
I went and tried to clip SONAR today and dawg gonnit... I had trouble doing it.

I've under stood the 64 bit engine part for many years but I guess I was superstitiously clinging to the idea that the plugins had windows that could be clipped.

And so when I saw a red on a track or effect and I heard it as well I just made the adjustment locally with out fully acknowledging that the distortion was actually on the Main Outs. That still seems like a useful way to work... by that I mean observing the meters and making local adjustments.

I did many tests with Sonitus EQ where the meters were just pegged solid red, but as suggested if I compensated the levels in a complimentary manner on the master bus everything would resolve.

I did several tests where I boosted the tracks levels and reduced the bus... and visa versa and then did null tests and everything seemed almost exact each time.


What I did notice is that when I had the EQ meters pegged and the bus levels reduced so that it was clean on the output (and would bounce to track cleanly as described above) that I still heard an occasional faint audible nastiness, clear digital distortion, sputtering away even though the master bus meters would indicate that there was nothing clipping. I couldn't find proof of this noise in the bounces... but it was clearly audible.

I was suspecting that perhaps some fast spikes were getting past the meters because of ballistic characteristics and making it to the ADA... but I couldn't see anything in the Bounces.

I thought maybe it was just happening in the speaker/amps but the meters on my MOTU output were indicating that nothing should be stressing the speakers. So, once again it could be something getting by the meter ballistics. I don't know what to make of it.



Then I decided to fire up the Sonitus Compressor.

I just went for it:



And I got this horrific grungy clipping noise... but the master output levels are -5dBFS+/-

Maybe that's just the sound I dialed up? It sure sounds like clipping to me.

Anyways, I apologize for making the incorrect statements in the previous posts in this thread.

I'm curious about the compressor noises... but I don't think it proves or disproves anything... so I am just offering it as a curiosity.


all the best,
mike




post edited by mike_mccue - 2011/06/17 15:02:30


#43
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re:Gain Staging 2011/06/17 14:50:54 (permalink)
If you hit a compressor hard enough the results are very similar to clipping - the signal (not just the peaks) is so far above the threshold that the gain reduction is massive (and not polite), causing the distortion you hear.

 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#44
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re:Gain Staging 2011/06/17 15:04:16 (permalink)
I suppose I should add that, technically, compressors always distort the signal when they're working. It's just that if they work gracefully enough we don't hear it as distortion.

But try running a perfect sin wave through a spectrum analyzer both before and after your favorite compressor and see what you get (make sure the comp's set so you see some gain reduction).

 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#45
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re:Gain Staging 2011/06/17 15:10:49 (permalink)
Yes, some would say that any change between the input and output is defined as distortion.

I've been thinking in terms of a simple digital over rather than analogous distortion.

best regards,
mike


#46
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re:Gain Staging 2011/06/17 15:17:22 (permalink)
Re: Sine Wave testing.

This type of testing can sometimes overlook characteristics that occur at specific frequencies.

You can use a constant tone or a pulse to simulate transient reactions.

A compressor will alter a sine wave in different ways as the relationship between the sine wave's amplitude and frequency and the compressor settings are altered.

So use of pure tones may be better for creating a demonstration while not so good for investigating issues.


best regards,
mike



#47
SCorey
Max Output Level: -80 dBFS
  • Total Posts : 538
  • Joined: 2011/04/26 15:13:14
  • Location: Salt Lake City, UT
  • Status: offline
Re:Gain Staging 2011/06/17 15:55:37 (permalink)
Then I decided to fire up the Sonitus Compressor. I just went for it: And I got this horrific grungy clipping noise... but the master output levels are -5dBFS+/- Maybe that's just the sound I dialed up? It sure sounds like clipping to me.

You've got the limiter engaged in the Compressor. I've found that to be a terrible sounding limiter that sounds remarkably like horrific grungy clipping noise.  Plus, the compressor itself will not respond very gracefully to a slammin input. IMO, of course.  But the sonitus compressor is very good at handling 'sane' inputs.  The sonitus eq will happily lend its voice to signals at +300 dBFS, and provided you turn it down again somewhere below 0dBFS will sound just fine.

-Steve Corey
#48
craigb
Max Output Level: 0 dBFS
  • Total Posts : 41704
  • Joined: 2009/01/28 23:13:04
  • Location: The Pacific Northwestshire
  • Status: offline
Re:Gain Staging 2011/06/17 15:57:46 (permalink)
Good stuff!


 
Time for all of you to head over to Beyond My DAW!
#49
The Maillard Reaction
Max Output Level: 0 dBFS
  • Total Posts : 31918
  • Joined: 2004/07/09 20:02:20
  • Status: offline
Re:Gain Staging 2011/06/17 16:02:38 (permalink)
SCorey



Then I decided to fire up the Sonitus Compressor. I just went for it: And I got this horrific grungy clipping noise... but the master output levels are -5dBFS+/- Maybe that's just the sound I dialed up? It sure sounds like clipping to me.

You've got the limiter engaged in the Compressor. I've found that to be a terrible sounding limiter that sounds remarkably like horrific grungy clipping noise.  Plus, the compressor itself will not respond very gracefully to a slammin input. IMO, of course.  But the sonitus compressor is very good at handling 'sane' inputs.  The sonitus eq will happily lend its voice to signals at +300 dBFS, and provided you turn it down again somewhere below 0dBFS will sound just fine.


For the record... I am a big fan of the Sonitus Compressor...  I like it a lot.


#50
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re:Gain Staging 2011/06/17 16:04:37 (permalink)
mike_mccue


Re: Sine Wave testing.

This type of testing can sometimes overlook characteristics that occur at specific frequencies.

You can use a constant tone or a pulse to simulate transient reactions.

A compressor will alter a sine wave in different ways as the relationship between the sine wave's amplitude and frequency and the compressor settings are altered.

So use of pure tones may be better for creating a demonstration while not so good for investigating issues.


best regards,
mike
True. The point was just that compressors do cause distortion. But since it might not actually sound like distortion to us, it's not necessarily something we need to care about.

I also should have said that you will only see distortion when the compressor is actually reacting to gain changes in the signal - once the amount of gain reduction is constant, it's just a gain change. And gentler compression (low ratio, RMS detection, soft knee, slower attack and release) won't distort nearly as much as hard limiting (high ratio, peak detection, hard knee, very fast attack and release).

 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#51
jm24
Max Output Level: -54 dBFS
  • Total Posts : 2127
  • Joined: 2003/11/12 10:41:12
  • Status: offline
Re:Gain Staging 2011/06/17 16:11:54 (permalink)
drewfx1

 And gentler compression (low ratio, RMS detection, soft knee, slower attack and release) won't distort nearly as much as hard limiting (high ratio, peak detection, hard knee, very fast attack and release).
'Tis why it is reasonable to use 2+ compressors in series to not have any one of them sound angry.

#52
Jeff Evans
Max Output Level: -24 dBFS
  • Total Posts : 5139
  • Joined: 2009/04/13 18:20:16
  • Location: Ballarat, Australia
  • Status: offline
Re:Gain Staging 2011/06/17 17:06:56 (permalink)
I agree with SCorey that limiter in the Sonitus compressor should never be used. It is just plain bad.  I can hear the difference quite clearly when its ON or OFF. The Sonitus compressor sounds better without it. If you have been using it all the time then you have not been extracting the best sound out of that compressor.

It is also silly the attack defaults to 160 ms as well. Those two parameters are the first thing I reset everytime.

Specs i5-2500K 3.5 Ghz - 8 Gb RAM - Win 7 64 bit - ATI Radeon HD6900 Series - RME PCI HDSP9632 - Steinberg Midex 8 Midi interface - Faderport 8- Studio One V4 - iMac 2.5Ghz Core i5 - Sierra 10.12.6 - Focusrite Clarett thunderbolt interface 
 
Poor minds talk about people, average minds talk about events, great minds talk about ideas -Eleanor Roosevelt
#53
jm24
Max Output Level: -54 dBFS
  • Total Posts : 2127
  • Joined: 2003/11/12 10:41:12
  • Status: offline
Re:Gain Staging 2011/06/17 20:53:51 (permalink)
drewfx1


Karyn


jm24


Bristol_Jonesey

my concern was those rare occasions when an incoming signal might be a bit on the hot side thereby making it difficult to decide where the threshold should be set.

In the analog world this would be "attacked" by placing a limiter before the compressor.

J
In the analog world you'd turn down the channel input gain (trim) or even press the -20db pad button.
Doing this reduces the average signal processed by the compressor. The point here is to address the intensity of the spurious transient so the average signal can be properly processed. This can be done by adding a limiter/compressor to "smooth" the average signal.
 
 
 
#54
gracypetro
Max Output Level: -90 dBFS
  • Total Posts : 5
  • Joined: 2011/06/20 02:21:57
  • Status: offline
Re:Gain Staging 2011/06/20 05:04:58 (permalink)
Now that is fascinating Drew, thanks for sharing it. I don't get much in to the scientific stuff or how things work or method, I use my ears & adjust until it sounds right. Compressors are not going to be useful if the signal is low to get to the top.

usa online casino
#55
Beagle
Max Output Level: 0 dBFS
  • Total Posts : 50621
  • Joined: 2006/03/29 11:03:12
  • Location: Fort Worth, TX
  • Status: offline
Re:Gain Staging 2011/06/20 07:46:15 (permalink)
wow - personalized spam.  drew you should feel privilged! 

http://soundcloud.com/beaglesound/sets/featured-songs-1
i7, 16G DDR3, Win10x64, MOTU Ultralite Hybrid MK3
Yamaha MOXF6, Hammond XK3c, other stuff.
#56
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re:Gain Staging 2011/06/20 10:53:14 (permalink)
Beagle


wow - personalized spam.  drew you should feel privilged! 

Wow, you're right!

But actually, in that case maybe I should be entitled to a piece of whatever they're getting from the evil spam lords?

 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#57
Beagle
Max Output Level: 0 dBFS
  • Total Posts : 50621
  • Joined: 2006/03/29 11:03:12
  • Location: Fort Worth, TX
  • Status: offline
Re:Gain Staging 2011/06/20 10:58:24 (permalink)
drewfx1


Beagle


wow - personalized spam.  drew you should feel privilged! 

Wow, you're right!

But actually, in that case maybe I should be entitled to a piece of whatever they're getting from the evil spam lords?


what does it matter?  it won't be BECAN! 

http://soundcloud.com/beaglesound/sets/featured-songs-1
i7, 16G DDR3, Win10x64, MOTU Ultralite Hybrid MK3
Yamaha MOXF6, Hammond XK3c, other stuff.
#58
drewfx1
Max Output Level: -9.5 dBFS
  • Total Posts : 6585
  • Joined: 2008/08/04 16:19:11
  • Status: offline
Re:Gain Staging 2011/06/20 11:14:10 (permalink)
Beagle


drewfx1


Beagle


wow - personalized spam.  drew you should feel privilged! 

Wow, you're right!

But actually, in that case maybe I should be entitled to a piece of whatever they're getting from the evil spam lords?


what does it matter?  it won't be BECAN! 



 In order, then, to discover the limit of deepest tones, it is necessary not only to produce very violent agitations in the air but to give these the form of simple pendular vibrations. - Hermann von Helmholtz, predicting the role of the electric bassist in 1877.
#59
Page: < 12 Showing page 2 of 2
Jump to:
© 2024 APG vNext Commercial Version 5.1