R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 [2]  All   Go Down

Author Topic: Digital summing saturation = myth?  (Read 10391 times)

Jay Kadis

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2165
Re: Digital summing saturation = myth?
« Reply #15 on: April 02, 2009, 11:04:31 am »

Jon Hodgson wrote on Wed, 01 April 2009 16:10


All of the above basically adds up to the following...

in modern DAWs, so long as you don't see any clipping lights, you shouldn't have anything to worry about.
I agree in theory, but I have noticed mixing in Logic that relatively higher levels sometimes results in audible distortion even though the level indicators remain below the red.  Other mixes sound OK at the same level.  This may have to do with plug-ins employed and/or their interface to the main DAW software or some other issue, but distortion does indeed seem to occur.  Lowering all the channel levels eliminates the problem.

Jon Hodgson

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1854
Re: Digital summing saturation = myth?
« Reply #16 on: April 02, 2009, 11:38:39 am »

Jay Kadis wrote on Thu, 02 April 2009 16:04

Jon Hodgson wrote on Wed, 01 April 2009 16:10


All of the above basically adds up to the following...

in modern DAWs, so long as you don't see any clipping lights, you shouldn't have anything to worry about.
I agree in theory, but I have noticed mixing in Logic that relatively higher levels sometimes results in audible distortion even though the level indicators remain below the red.  Other mixes sound OK at the same level.  This may have to do with plug-ins employed and/or their interface to the main DAW software or some other issue, but distortion does indeed seem to occur.  Lowering all the channel levels eliminates the problem.


You would need to look into where the distortion is actually happening, also what the level indicators actually show (they might not show instantaneous peaks).

Where the distortion is almost certainly not happening however is in the mix bus itself, since in logic that's using floats and has obscene headroom. It's either something clipping it before it gets there, or after.
Logged

Ashermusic

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 684
Re: Digital summing saturation = myth?
« Reply #17 on: April 02, 2009, 11:39:21 am »

Jay Kadis wrote on Thu, 02 April 2009 16:04

Jon Hodgson wrote on Wed, 01 April 2009 16:10


All of the above basically adds up to the following...

in modern DAWs, so long as you don't see any clipping lights, you shouldn't have anything to worry about.
I agree in theory, but I have noticed mixing in Logic that relatively higher levels sometimes results in audible distortion even though the level indicators remain below the red.  Other mixes sound OK at the same level.  This may have to do with plug-ins employed and/or their interface to the main DAW software or some other issue, but distortion does indeed seem to occur.  Lowering all the channel levels eliminates the problem.


In Logic, the signal passes through the channel strip from the top insert to the bottom, so if you have a fixed point plug-in in that chain that is being fed too hot a signal, it certainly can distort. And certainly you can distort the output.

However, assuming that is not going on, if you have pre-fader metering engaged and nothing is going into the red, it is not possible to be creating "distortion" on the channels. For that matter even if it IS going into the red, except for the output, of course.

Still, in my book "Going Pro With Logic Pro 8" and in my private training, I encourage students to observe the good standard mixing practice of controlling the signal at its source so that it is not going into the red.
Logged
Composer, Logic Pro Certified Trainer, Level 2
Author of "Going Pro with Logic Pro 8"

www.jayasher.com

cerberus

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2651
Re: Digital summing saturation = myth?
« Reply #18 on: May 08, 2009, 01:32:14 pm »

compasspnt wrote on Wed, 01 April 2009 20:19

Jon, what would you say the trade off is in using floating point?  Or is it a Free Lunch?


although jon has answered. i find there is a separate, but related issue which
is a consequence of floating point's variable resolution. (32 float
values can have -up to- 23 bits of resolution, but unless the
material is a test tone, not every sample could have the
maximum resolution.) but let's assume that we have
more than enough resolution for our purposes; i
believe there is still another important issue:

with floating point, the noise floor of the system also "floats". i find that
adding some steady low level noise to summing busses in my daw
has a benefit. i do not understand the phenemonon perfectly
well, but after many years of this practice: i am sure that
whatever minor effect i am exploiting is real; and
although it's small,  it's always significant.

(since we are already splitting hairs here,
or are we?  imo,  such subtleties
are why some people with
discerning ears claim
to prefer analogue.)

jeff dinces

Kassonica

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: Digital summing saturation = myth?
« Reply #19 on: May 16, 2009, 06:08:12 am »

cerberus wrote on Sat, 09 May 2009 03:32

compasspnt wrote on Wed, 01 April 2009 20:19

Jon, what would you say the trade off is in using floating point?  Or is it a Free Lunch?


although jon has answered. i find there is a separate, but related issue which
is a consequence of floating point's variable resolution. (32 float
values can have -up to- 23 bits of resolution, but unless the
material is a test tone, not every sample could have the
maximum resolution.) but let's assume that we have
more than enough resolution for our purposes; i
believe there is still another important issue:

with floating point, the noise floor of the system also "floats". i find that
adding some steady low level noise to summing busses in my daw
has a benefit. i do not understand the phenemonon perfectly
well, but after many years of this practice: i am sure that
whatever minor effect i am exploiting is real; and
although it's small,  it's always significant.

(since we are already splitting hairs here,
or are we?  imo,  such subtleties
are why some people with
discerning ears claim
to prefer analogue.)

jeff dinces



Very interesting. What kind of noise and how do you introduce it ?

Thanks in advance if you answer.

Logged
Mark Ludwig

Should have been a Drummer


cerberus

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2651
Re: Digital summing saturation = myth?
« Reply #20 on: May 16, 2009, 11:13:03 am »

hi mark;

it's from one of the waves vintage plug-ins, seems to be mostly 60hz line noise.
there is a switch on these plug-ins which can turn off the noise; so it is easy
for anyone to experiment and perhaps discover the effect.

i recorded the noise to a 32 bit file, which i run as a daw track. i can send it
to any summing busses. in order to prevent noise  from building up in
complex chains: i run a polarity-inverted  copy, sent to alternate
busses; i tweak the exact levels carefully by ear.

jeff dinces

Kassonica

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: Digital summing saturation = myth?
« Reply #21 on: May 16, 2009, 07:54:47 pm »

cerberus wrote on Sun, 17 May 2009 01:13

hi mark;

it's from one of the waves vintage plug-ins, seems to be mostly 60hz line noise.
there is a switch on these plug-ins which can turn off the noise; so it is easy
for anyone to experiment and perhaps discover the effect.

i recorded the noise to a 32 bit file, which i run as a daw track. i can send it
to any summing busses. in order to prevent noise  from building up in
complex chains: i run a polarity-inverted  copy, sent to alternate
busses; i tweak the exact levels carefully by ear.

jeff dinces



Much thanks for your answer Jeff.

I'm still trying to get my head around this so bare with me if you can.

Your saying that you run this to your internal (DAW) busses and it helps with the summing process say like (and please correct me if I'm wrong) a kind of dither?

I am truly fascinated by this and any other info would be graciously accepted.

I am a ITB guy and anything I can learn about making the digital process better for me and my clients I'm very interested in.

I don't have the waves plugins, but I'm sure I will have something that will do the trick.

 
Logged
Mark Ludwig

Should have been a Drummer


compasspnt

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 16266
Re: Digital summing saturation = myth?
« Reply #22 on: May 16, 2009, 09:04:42 pm »

Just for fun, try adding a hum track and a noise track to your mix.

Hum from perhaps an old radio or other device that has a steady low (60-ish) hum.

Noise from perhaps a TV set on a blank channel.

Keep them very, very low, just below where you can/can't hear them.

Then see how your 'regular' mix compares to this one with the added artifacts.
Logged

cerberus

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2651
Re: Digital summing saturation = myth?
« Reply #23 on: May 16, 2009, 11:30:50 pm »

Kassonica wrote on Sat, 16 May 2009 19:54

Your saying that you run this to your internal (DAW) busses and it helps with the summing process say like (and please correct me if I'm wrong) a kind of dither?

that is my anecdotal observation, but it cannot be literally dither.

Kassonica wrote on Sat, 16 May 2009 19:54

I am truly fascinated by this and any other info would be graciously accepted.


when i refer to floating point noise floor, i mean the quantisation noise,
which is unfortunately correlated to the sample value (amplitude).

i will go out on a limb and say that "activity occurring below -150db"
could have an effect on the sound... daw companies may not
agree, but whatever is going on there is quite insidious,
and were it to be amplified...

so at the very least, i am masking that with noise at the minimum 24 bit level,
(somewhere near -143db).note: i always monitor through dither, which must
be placed just after the output fader in the daw, imo, one will not learn
anything about what we are discussing if they monitor without dither.

well, maybe that is all that is going on? or maybe there is more?

Quote:

I am a ITB guy and anything I can learn about making the digital process better for me and my clients I'm very interested in.

I don't have the waves plugins, but I'm sure I will have something that will do the trick.

compasspnt wrote on Sat, 16 May 2009 21:04

Just for fun, try adding a hum track and a noise track to your mix.

Hum from perhaps an old radio or other device that has a steady low (60-ish) hum.

Noise from perhaps a TV set on a blank channel.

Keep them very, very low, just below where you can/can't hear them.

Then see how your 'regular' mix compares to this one with the added artifacts.



60hz is near a "b",  (50hz is not a note).
so perhaps best done in n.a. or japan?

jeff dinces

Kassonica

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 280
Re: Digital summing saturation = myth?
« Reply #24 on: May 17, 2009, 05:48:20 am »

WOW

Thank you both for this.

I remember reading years ago about mixers putting white noise very very quietly underneath their mixes (in the analogue days) for A, to mask tape hiss and as a psychoacoustic effect.

This seems like a digital version in a way.

Can't wait to try it and I will let you know the outcome (monitoring with dither of course)

CHeers, you've made a long hard day much better Very Happy      
Logged
Mark Ludwig

Should have been a Drummer


Tomas Danko

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 4733
Re: Digital summing saturation = myth?
« Reply #25 on: May 18, 2009, 05:49:28 am »

compasspnt wrote on Sun, 17 May 2009 02:04

Just for fun, try adding a hum track and a noise track to your mix.

Hum from perhaps an old radio or other device that has a steady low (60-ish) hum.

Noise from perhaps a TV set on a blank channel.

Keep them very, very low, just below where you can/can't hear them.

Then see how your 'regular' mix compares to this one with the added artifacts.



On a game I worked on that was released last year (Mirror's Edge), we put "ambient noise" very very low into the background of the audio mix and it helped making everything sound more connected/glued together.
It masked individual sounds start and endings so that they sounded more like the belonged to the soundscape instead of having an obvious "out of nowhere" transient beginning as well as abrupt decaying end that can sound unnatural.

I made a lot of 5-10 second loops in stereo, that sounded like the ambient background noise you have in halls, small rooms, large areas, air vent drums etc

Then we switched between them whenever you entered a hall, or stepped out from the hall into a small office room etc

It also gave a subtle sense of 'air' and awareness of your position in 3D space, you know like when you exit a hall and enter a large room and the noise/air changes. It was a cool little trick, with several benefits.

Noise is good. Smile
Logged
http://www.danko.se/site-design/dankologo4s.gif
"T(Z)= (n1+n2*Z^-1+n2*Z^-2)/(1+d1*z^-1+d2*z^-2)" - Mr. Dan Lavry
"Shaw baa laa raaw, sidle' yaa doot in dee splaa" . Mr Shooby Taylor

Ian Visible

  • Full Member
  • ***
  • Offline Offline
  • Posts: 179
Re: Digital summing saturation = myth?
« Reply #26 on: May 29, 2009, 10:30:05 am »

The atmosphere in Mirror's Edge is amazing!

Your hard work definitely paid off - in my humble opinion, the game would not be the same without it!

Tomas Danko

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 4733
Re: Digital summing saturation = myth?
« Reply #27 on: May 29, 2009, 11:09:56 am »

Ian Visible wrote on Fri, 29 May 2009 15:30

The atmosphere in Mirror's Edge is amazing!

Your hard work definitely paid off - in my humble opinion, the game would not be the same without it!


Thanks, Ian. We were five people working on the sound, not counting outsourcing (such as music composers etc).

It was a bit tricky to get the mix into the sweet spot, because we only had one mix setting to tweak for all three platforms (PC, Xbox 360 and Playstation 3) due to the way the Unreal 3 engine works.

And to make it worse, Xbox 360 will automatically downmix to the subwoofer while Playstation 3 will not. So in practice, PS3 runs a 5.0 setting. And in order to get enough low end thump/feel on that machine, we had to crank up the low end on certain sounds which then made the Xbox 360 version very bass heavy. Also, the fold down to stereo summing system and pan law was a tad bit different between PC, Xbox 360 and Playstation. So you can really only tweak it to be as good as possible across all systems, but it's not going to be perfect anywhere.

That's technology for you. Smile
Logged
http://www.danko.se/site-design/dankologo4s.gif
"T(Z)= (n1+n2*Z^-1+n2*Z^-2)/(1+d1*z^-1+d2*z^-2)" - Mr. Dan Lavry
"Shaw baa laa raaw, sidle' yaa doot in dee splaa" . Mr Shooby Taylor

Nick Sevilla

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 853
Re: Digital summing saturation = myth?
« Reply #28 on: December 14, 2009, 02:35:00 pm »

cerberus wrote on Sat, 16 May 2009 08:13

hi mark;

it's from one of the waves vintage plug-ins, seems to be mostly 60hz line noise.
there is a switch on these plug-ins which can turn off the noise; so it is easy
for anyone to experiment and perhaps discover the effect.

i recorded the noise to a 32 bit file, which i run as a daw track. i can send it
to any summing busses. in order to prevent noise  from building up in
complex chains: i run a polarity-inverted  copy, sent to alternate
busses; i tweak the exact levels carefully by ear.

jeff dinces


I hear this on the new "Eddie Kramer" collection. Quite a lot of it actually. I ended up bypassing the plugin except when it was being used. the 50 or 60 cycle hum definitely added an interesting texture to the lead vocal, and the artist was happy with the results. I now have to see if you can turn that feature off, and if not, email Waves so they can put a switch on there to turn it off when not desired.

Cheers
Logged
-------------------------------------------------
It is quite possible, captain, that they find us grotesque and ugly and many people fear beings different from themselves.

www.nicksevilla.com
Pages: 1 [2]  All   Go Up