volume level: soundcard vs monitors for mixing

Ed Kishel

Ed Kishel

Veteran
I am upgrading my PC with some better audio hardware: switching out my SoundBlaster soundcard for an M-Audio Audiophile 192 and will connect those to a pair of BX5 D2 monitors. It's not Skywalker Sound- but it's better than what came with my computer. My question is- the pc/soundcard has a volume control, as does the monitors... but what do you set each at? And how do their volume levels relate to each other? Do I set them both at 50%? My guess is below:

1) play back -20db ref tone from NLE and make sure your soundcard is chosen for audio playback
2) watch the levels in your soundcard's mixer
3) adjust gain volume from NLE's mixer so that you see -20db in your soundcard's level meter (if it doesn't by default)
4) place a sound level meter where you sit
5) adjust your monitor speakers volume until the meter shows (?)
5) begin mixing

With video- you have scopes in your NLE for adjusting exposure and color. With sound, well I'm not sure where to go to get the most accurate volume that my modest gear can output.

Thanks,
 
Last edited:
Good working volume is subjective to a certain point with room acoustics being part of it. some people mix to a level that is comfortable to them rather than a pre-set level as everyone has different thresholds.
When I mix Im more comfortable with levels being in the low 80's A weighed. I work with other engineer/producers that work at much higher levels and Ihave to step out of the room sometimes because its too loud for me. I also will do several passes with it barely audible which points out to me things that are sticking out too much much easier than at louder levels as your ears tend to have built in compressors when exposed to louder sound levels.

Find a level that is comfortable for you as you are the person that will be sitting in front of your monitors for hours on end. but I vote towards the lower comfortable level which will give you more time before listener's fatigue.

good luck
 
Analogue gain structure tends to be more audible in terms of signal to noise. The sound card meters inside the computer and the computers volume control should be set so that any computer generated warning sounds are lower than your sound card audio. Sometimes, you set what appears to be a great level only to discover that when the warning bong goes - it's ear shattering.

On my audio PC in the studio, I've got those sounds turned off anyway, but what I do is look at the output meters and check the point at which distortion is evident. On my system, I can't go past 0 - because if I do, the audio device can't manage it. I have a Lexicon Omega I use occasionally on my macbookpro, and this does seem to be able to go into the red slightly with no problems, so I suspect it's metering accuracy is a bit off. If I take a test tone of 0dB 1KHz and put it into premiere, sound forge or cubase, which are my day to day bits of software, they all read correctly. This means any excursion into the red will be more than digital 1, and will be distorted. So I then connect the output from this computer into an analogue audio mixer which I use for summing the computer with other sources, and as it has loads of headroom, being analogue, I adjust the input gain so that with the channel fader at o - I get the change into the red area of the LED output meters at 0. The channel indicators are a bit out of step, but I can run the mixer like this and then I adjust the monitor speaker level to give me a comfortable working level. This means for my system, the amp attentuators are quite low. For my system this works ok. In the edit suite next door, the computer sound card goes straight to the active speakers, so I set it up by sticking in the same test tone, and then making sure that it is just at the 0dB mark, I run the windows volume at about 75%. This I've discovered is the best compromise, much above 80% and the audio starts to distort slightly, looked at as a waveform. I assume the internal matching is not exact again. Once I established the optimum level like this, I bring up the monitors on their own volume control. The 75% setting on the windows fader also makes the onboard sound effects to be not quite so loud - and on the edit machine I leave the warnings on. In Premiere, I treat the audio levels as absolute, and if I burn a DVD of the test tone, and then strip out the audio and bring it back in the levels are pretty similar. A little less than the original. I don't want to set Premiere lower and then boost gain on the windows control and speakers - this produces more noise. This gain structure works for me. Once set, I don't change it if I can, so levels remain similar from sound studio to edit suite - but can, if necessary go UP considerably in the sound studio - but as I'm getting older, I have got used to mixing at lower levels.
 
Depending on what you are doing the setup can change.
If your working in film the specs are pretty straight forward. A good explanation of how to get there is HERE.

That is based on editing for the big screen and feature video.

Personally I calibrate for my small room and mark the position on the amp. Then I work most of the time at a lower level and only go up to full for checking and final mixing.

You need to calibrate to good pink noise. Dolby Pink is the standard and there is a link (in the linked thread) where you can download some. Some pink noise generators in software are OK some are not and I'm not sure which are good but I know ProTools is not, so use a file that has been checked out.

You have to keep digital VS analog levels straight in your head, zero is not zero.
As a rule you don't ever want to get close to digital zero. At over -3dB higher frequencies may be clipping. You won't "hear" it but it will muddy up the sound.

Films are mixed by ear in calibrated spaces. TV shows are usually required to hit certain delivery specs so you will also have to pay attention to the numbers.

Calibration is at -20 pink and you are aiming for around 80 dB (varies with room size and your distance).

As a general rule anything involving level change is probably quietest at about 75%. You will notice that is usually the sweet spot on mixers. If you have a "unity" setting then as much as possible that is where you want things set.

Make the final level adjustment as close to the speaker end of things as possible.

Mark something so you know you can get back to the calibration spot, so you don't have to work at that level if you don't want.

The BIGGEST thing is to mix at a consistent level. On mix stages the room is calibrated at least once a day and often again after lunch. If your speaker levels change from session to session your mix will be a mess.
 
Hi Noiz, thanks so much for the detailed answer!

A quick question, when mixing 5.1 surround for digital cinema delivery, is the peak volume of the sound track usually around -20dBFS?
 
BlueMoon Audio has calibration files and such on their site for the -20/85 cal. though 85dBSPL is hot for a small room, I use 75dB for TV stuff in a smaller room w/ nearfields with the 'new' ATSC meterering.
 
marvinhello said:
A quick question, when mixing 5.1 surround for digital cinema delivery, is the peak volume of the sound track usually around -20dBFS?
Click to expand...

That's a negative. I've seen films hit the clipping indicator plenty of times in Pro Tools for digital cinema deliveries.
 
There are a number of "digital Cinema" formats so I don't think that would be a deciding factor.

In theory when mixing for the big screen in a calibrated room dialog will sit around -20, or used to. Unless there is a specific spec for a deliverable films are mixed by ear, so levels vary a lot. An average over all is especially variable. I don't know what the average of "300" was but it was close to the pain threshold. On the flip side that film of questionable ethics (murder is fine as long as you think the guy is guilty of something and well it makes you feel better) "In the Bedroom" was extremely quiet. Other than the concept the thing that disturbed me most about the film is it didn't sound like they spent any time in post.

So as Ryan said you aren't going to find a lot of films with an average level of -20.
 
Noiz2 said:
There are a number of "digital Cinema" formats so I don't think that would be a deciding factor.

In theory when mixing for the big screen in a calibrated room dialog will sit around -20, or used to. Unless there is a specific spec for a deliverable films are mixed by ear, so levels vary a lot. An average over all is especially variable. I don't know what the average of "300" was but it was close to the pain threshold. On the flip side that film of questionable ethics (murder is fine as long as you think the guy is guilty of something and well it makes you feel better) "In the Bedroom" was extremely quiet. Other than the concept the thing that disturbed me most about the film is it didn't sound like they spent any time in post.

So as Ryan said you aren't going to find a lot of films with an average level of -20.
Click to expand...

Thanks for the reply, for Digital Cinema I mean those cinemas with DCI compliant servers, projectors and speakers.

I'm a bit confused now to be honest, for web and TV delievery, dialogue usually peaks around -12dBFS, can I do the same for my Digital Cinema Package? so I just need to have one master mixdown, or do I need to decrease the volume a few dBs specificly for theatrical release so dialogue peaks around -20dBFS? Are speakers in a standard digital cinema adjustable by projectionist? so they don't have a fixed volume to avoid films that are too loud?
 
ryanjf said:
That's a negative. I've seen films hit the clipping indicator plenty of times in Pro Tools for digital cinema deliveries.
Click to expand...

Wow really? if the speakers in the cinema are calibrated as -20dBFS equals 85dBSPL, won't they instantly make the audience deaf? or the projectionist will adjust the calibration prior to the first public screening to a acceptable level? like we use our remote control to "+" and "-" when we watch TV....
 
marvinhello said:
Thanks for the reply, for Digital Cinema I mean those cinemas with DCI compliant servers, projectors and speakers.

I'm a bit confused now to be honest, for web and TV delievery, dialogue usually peaks around -12dBFS, can I do the same for my Digital Cinema Package? so I just need to have one master mixdown, or do I need to decrease the volume a few dBs specificly for theatrical release so dialogue peaks around -20dBFS? Are speakers in a standard digital cinema adjustable by projectionist? so they don't have a fixed volume to avoid films that are too loud?
Click to expand...

Dialogue can certainly peak higher than -12dBFS for theatrical. The AVERAGE (not peak) can/might be around -20dBFS for theatrical. We're not saying it has to, but it can. Remember that mixers mix by ear if they are in a calibrated room. If it sounds good, it is good; if it's too loud, then it's too loud. For TV delivery, each broadcaster has their own specs so that varies. There is a great sticky on Gearslutz (post production forum) about this that is a great read. For web, there really isn't any standard.

Printmasters are sent directly to the DCP "authorers" so there is no change in level that I know of. Yes speakers can be adjusted by the projectionist and typically are at regular theaters. The projectionist can choose to play the film "at level" but more often than not, they turn it down a few dB.

EDIT: DCPs are 48kHz @ 24bit @ 24FPS so whether you'll be creating that version (or the client) is a different issue.
 
Last edited:
marvinhello said:
Wow really? if the speakers in the cinema are calibrated as -20dBFS equals 85dBSPL, won't they instantly make the audience deaf? or the projectionist will adjust the calibration prior to the first public screening to a acceptable level? like we use our remote control to "+" and "-" when we watch TV....
Click to expand...

Yes. That means in your DAW, you have 20dB of headroom (to digital 0) which means that if you have an explosion that hits 0dBFS, then that sound is played at 105 SPL or higher. Hopefully that explosion (or whatever it may be) is not sustained at that level because it does hurt. Have you watched the train wreck sequence in Super 8 in theaters??! You'll notice that that sequence is LOUD! I watched Super 8 at two different theaters and the second theater definitely turned it down. It was almost a no brainer what the projectionist did.
 
ryanjf said:
Yes. That means in your DAW, you have 20dB of headroom (to digital 0) which means that if you have an explosion that hits 0dBFS, then that sound is played at 105 SPL or higher. Hopefully that explosion (or whatever it may be) is not sustained at that level because it does hurt. Have you watched the train wreck sequence in Super 8 in theaters??! You'll notice that that sequence is LOUD! I watched Super 8 at two different theaters and the second theater definitely turned it down. It was almost a no brainer what the projectionist did.
Click to expand...

Thanks, I think I'll just mix the 5.1 surround at the same level as web/TV version... and tell the projectionist to adjust the volume as required... it's a student film which is going to be screened in a small digital cinema (takes DCP), so I don't have access to a professionally calibrated room or funds to hire a mixer...
 
I think that is the problem - the end users, be it home or cinemas set their own levels. The big players in the superbly fitted studio complexes can set their own levels but sometimes the output sounds like a brick wall limiter to keep the level below the prescribed point - while others are really expanded - so although the peaks are high, far too much happens at a level below the noise floor in the listening space. One small multi space cinema near me has amazingly noisy air handling - so the solution is to really push the levels, and peaks are very unpleasant. Explosions and that kind of sound effects are very rough! It's a habit that even the audio CD people are now doing - mixing too low and expanding the dynamic range to a point where the listener in the imperfect space cannot cope.
 
How are you going to get the tracks in a format compatible with the projector? I haven't done a Digital Cinema mix so personally don't know but if you haven't checked with them you should.

You can not for instance encode a Dolby Digital (big screen version) on your own. I could be wrong but I suspect that there may be issues getting your tracks encoded in a way that the projector can read.

I would also not recommend trying a 5.1 mix in a non calibrated mix stage let alone a non calibrated room, for playback on the big screen. There is a HUGE difference in how your room sounds compared to a theatre and you can't do a decent mix, especially a 5.1 one, that way. At least I would stay away from 5.1. The surrounds and sub could make watching the film Very distracting and or painful.


Cinemas are "supposed" to set the Dolby decoder at "7", but since trailers, ads and sometimes films come in at blistering levels they often turn them down. It's a hot debate with mixers, most of whom think Dolby ought to lock the level.

Personally I don't think you can lock out the theatre owner as long as you don't have a working standard for what kind of loudness is allowed to show up at the theatre. If it's too loud people walk out and want their money back. The studio sure isn't going to compensate them for that so...

It would be nice if there were something like the Dialog Normal number that gets encoded into DVDs, IF it were implemented as it was supposed to be. Networks require mixers to mix to a DiaN number a lot of times now. It was supposed to be a number that the network used to adjust the level on their end. Thus the mixing would be the same but at the playback end the level could be intelligently altered to fit the networks standards, time of day, etc.
 
Hi Noiz, Digital Cinema Initiative (DCI) is built entirely on open standards, audio to be specific, is just 24bit 48Khz uncompressed LPCM mono tracks, two mono tracks for 2.0, six mono tracks for 5.1, 8 tracks for 7.1

Picture wise, it's 12bit X'Y'Z colour space, encoded in JPEG2000 sequence and wrapped in MXF.

BTW, your idea on Dialogue Normal number is really interesting :)
 
marvinhello said:
Hi Noiz, Digital Cinema Initiative (DCI) is built entirely on open standards, audio to be specific, is just 24bit 48Khz uncompressed LPCM mono tracks, two mono tracks for 2.0, six mono tracks for 5.1, 8 tracks for 7.1

Picture wise, it's 12bit X'Y'Z colour space, encoded in JPEG2000 sequence and wrapped in MXF.

BTW, your idea on Dialogue Normal number is really interesting :)
Click to expand...

Also, audio has to be 24FPS!
 
ryanjf said:
Also, audio has to be 24FPS!
Click to expand...

Ryan, sound doesn't have a frame rate. The only way the frame rate would affect the sound is IF you were working with pulled down version of the film, OR your supplying sound with a TC stripe for sync.

There is no reason you would be working at anything but the correct frame rate these days and I don't imagine your sending sound on a separate tape (actually any tape at all but...).
 
ryanjf said:
Also, audio has to be 24FPS!
Click to expand...

And actually no. It supports a lot of frame rates.

Technical specifications

The DCP root folder (in the storage medium) contains a number of files, some used to store the image and audio contents, and some other used to organize and manage the whole playlist.[SUP][4][/SUP]
[edit]Picture MXF files

Picture contents may be stored in one or more reels corresponding to one or more MXF files. Each reel contains pictures asMPEG-2 or JPEG 2000 essence, depending on the adopted codec. MPEG-2 is no longer compliant with the DCI specification. JPEG 2000 is the only accepted compression format.
  • Supported frame rates are:
    • SMPTE (JPEG 2000)
      • 24, 25, 30, 48, 50, and 60 fps @ 2K
      • 24, 25, and 30 fps @ 4K
      • 24 and 48 fps @ 2K stereoscopic
    • MXF Interop (JPEG 2000) – Deprecated
      • 24 and 48 fps @ 2K (MXF Interop can be encoded at 25 frame/s but support is not guaranteed)
      • 24 fps @ 4K
      • 24 fps @ 2K stereoscopic
    • MXF Interop (MPEG-2) – Deprecated
      • 23.976 and 24 fps @ 1920×1080
  • Maximum frame sizes are 2048×1080 for 2K DC, and 4096×2160 for 4K DC. Common formats are:
    • SMPTE (JPEG 2000)
      • Flat (1998×1080 or 3996×2160), ~1.85:1 aspect ratio
      • Scope (2048×858 or 4096×1716), ~2.39:1 aspect ratio
      • HDTV (1920×1080 or 3840×2160), 16:9 aspect ratio (although not specifically defined in the DCI specification, this resolution is DCI compliant per section 3.2.1.2).
      • Full (2048×1080 or 4096×2160) (Official name by DCI is Full Container)
    • MXF Interop (MPEG-2) – Deprecated
      • Full Frame (1920×1080)
  • 12 bits per pixel precision (36 bits total)
  • XYZ colorspace
  • Maximum bit rate is 250 Mbit/s (1.3 MBytes per frame at 24 frame/s)
[edit]Sound MXF files

Sound contents are stored in reels, too, corresponding to picture reels in number and duration. In case of multilingual features, separate reels are required to convey different languages. Each file contains linear PCM essence.
  • Sampling rate is 48,000 or 96,000 samples per second
  • Sample precision of 24 bits
  • Linear mapping (no companding)
  • Up to 12 independent channels.
  • WAV container
[edit]Asset map file
Click to expand...
 
marvinhello said:
Wow really? if the speakers in the cinema are calibrated as -20dBFS equals 85dBSPL, won't they instantly make the audience deaf? or the projectionist will adjust the calibration prior to the first public screening to a acceptable level? like we use our remote control to "+" and "-" when we watch TV....
Click to expand...

If the peaks were at -20 dBFS you would be wasting over 3 bits of resolution. Dialog should sit somewhere around that figure, but certainly the peaks should go to about -6 dBFS to -3 dBFS, but not higher for clipping safety. If the dialog is kept at around -20 dBFS it is the job of the sound mixer to make the audio listenable with systems which are calibrated at 85 dB SLP for -20 dBFS signals, which means 102 dB SPL peaks with -3 dBFS peak level. With clean sound and good system our ears can take it, after all cinema audio is not like pop rock leaning on a limiter. Thank god there still are some standards, rules and best practices which ensures that, like keeping the dialog levels naturally low. Cinema sound can be dynamic as there is no competition in the theater, the problem is the DVD etc listened at home, which might need a slight compression compared to a theatre release.
 
You must log in or register to reply here.
Back
Top