LOW Audio levels from Pro Tools

[Jerry Hofmann]

Um… the audio post houses are likely just being conservative. Have you asked them about this? There are no settings I know of that could be causing this problem at your end IMHO…

Jerry

Apple Certified Trainer, Producer, Writer, Director Editor, Gun for Hire and other things. I ski.

8-Core 3.0 Intel Mac Pro, Dual 2 gig G5, AJA Kona SD, AJA Kona 2, Huge Systems Array UL3D, AJA Io HD, 17″ MBP, Matrox MXO2 with MAX Cinema Displays

[Richard Martz]

I have a call into the audio house that did this latest one. But the former house was Crawford Communications here in Atlanta. Those guys are one of the largest post houses on the east coast. So I know they know their stuff. You can see the different waveforms here:

https://idisk.mac.com/richardmartz//TECH STUFF/Montville Audio Waveform

Or tryL
files.me.com/richardmartz/aopa1d

One waveform is the low on e and one is a higher level and what I’m used to getting.

FCP 7
MAC 8 core
BlackMagic Studio Pro
After Effects
PhotoShop
Illustrator
Lots of other Fun Stuff!

[Adam Taylor]

you may want to check with them what reference level was used.

You may find they have recorded the files with a -20dbfs (or -18dbfs in europe) reference.

Ideally, they would have either included a sine wave tone signal ( although these are becoming less common in the digital file age.

If you are working with the full audio range in FCP then anything coming from a post audio facility that has been recorded with lower reference will sound quieter.

You need to know that reference level so that you can accurately compensate for it.

adam

Adam Taylor
Video Editor/Audio Mixer/ Compositor/Motion GFX/Barista
Character Options Ltd
Oldham, UK

http://www.sculptedbliss.co.uk

[Jerry Hofmann]

Have to agree with Adam… it’s not uncommon for post houses to choose -20db as the reference level. Just as common as using -12… so it’s not that they don’t know their stuff, it’s that they didn’t use the same level for reference most likely.

It’s also not impossible to use the file and raise it up to your ref level, right?

I’d do so and move on.

Jerry

Apple Certified Trainer, Producer, Writer, Director Editor, Gun for Hire and other things. I ski.

8-Core 3.0 Intel Mac Pro, Dual 2 gig G5, AJA Kona SD, AJA Kona 2, Huge Systems Array UL3D, AJA Io HD, 17″ MBP, Matrox MXO2 with MAX Cinema Displays

[Richard Martz]

Thanks. It is a house here in the states. I will repost details when I hear from them. I sort of expected a Zero dB file. I agree that tone would certainly be beneficial. And it is about as rare as a surviving 1980 Yugo. Even with tone the levels would still be the same.

FCP 7
MAC 8 core
BlackMagic Studio Pro
After Effects
PhotoShop
Illustrator
Lots of other Fun Stuff!

[Michael Gissing]

Record for 0dbfs and you are asking for trouble. Anything with peaks will inevitably clip. Most pro studios would not have a VO peaking above -10dbfs and some may be even safer.

[Erik Olsson]

The theory behind the setting of these levels is to ensure that there is plenty of headroom so as to prevent and digital distortion. This headroom is usually 10db.

As Adam has pointed out reference tone should be recorded at the beginning so as to ensure that the next link in the chain can set correct levels. Failing this you must assume that you can lift the signal level by 10db and this will bring the level into line with the other material on the timeline.

This can be achieved by raising the level by 10db either in the timeline or in Soundtrack Pro where you have better control of levels.

[Michael Gissing]

None of the audio should be peaking to zero. Anything delivered to a broadcaster should not have peaks above -10dbfs.

Don’t just lift the VO levels by 10db as that may cause clipping. Look at your monitoring levels and the overall levels of your clips and try to leave some headroom.

[Bret Williams]

What are you guys doing about DVDs? DVDs shouldn’t have any headroom. At least purchased DVDs don’t. Just like music CDs. After making plenty of DVDs for clients that they complained were too quiet, I’ve personally quit mixing for analog tape or broadcast since nothing I ever do will ever see the either. Generally, everything I do goes out to a qt file for DVD production or web production, both of which should essentially have no or minimal headroom. The mix of course is the same as always, but the overall level is more like averaging at -6 with peaks just under 0. it’s not hard to do of course. FCP will show any peaks over 0 on the fly, and it can also mark any peaks over 0 in the timeline for you, so it’s easy to do a quick check before outputting to DVD. In fact, since there’s no -10 peak reader built in, it’s actually easier, for me at least, to keep things in line. The meters are more sensitive and display a broader range as well toward the top end. I guess it’s my own sort of normalizing. Anyway, if I had to output a tape for broadcast, I’d basically drop the levels across the board 10-20db and be done. Either via the master or select all and opt+cmd+L. Reference tone for my work, if I were to output as is to tape (which I wouldn’t) would probably be around -10 or even -8, ensuring that if it was handed to a broadcaster as is, with the peaks nearing 0, they would sync up with the tone and lower the volume of my tone to -20, thus ensuring my peaks are also lowered to -10.

Richard, looks like you’ve abandoned the thread but in your tiff, looks like the one audio level is pretty normal and what I would expect. Peaks in the -10 or less range. The other track looks like it barely reaches -20 if you’re lucky. Peaks that low are just way to conservative imo. At what point are we losing resolution? I know it’s digital so there’s no tape hiss or anything, but without being an audio engineer, it just seems like the more the sound is squashed down into the lower volume range, the less range there would be. You can’t just go around shooting video with the whites at 40 ire to be safe, then crank them up in FCP to 90 ire and expect it to look the same as if it were recorded at 90 ire in the first place. The data would get stretched out and there’d be a lack of luminance resolution. I wonder at what point does this start to take place with audio?

[Richard Martz]

Thanks for all the comments:

Yes Bret. You are right. I’ve been mastering DVD audio levels consistently higher than recommended for some time with great results. We absolutely avoid peaks greater than zero dB but try to get as much saturation as possible for greater Signal to Noise ratio. Even with Digital there is still a noise floor that is induced in analog portions of the original recording. That includes microphones, mixers, and other signal processing. If noise is constant then the only way to improve this ratio is through the application of more signal in the mix. I personally try to apply a small amount of compression to make my levels consistent. But not enough that it is apparent to the listener.

If you lower your reference tone to attempt to make the audio “street legal” then, although you will achieve that goal, your audio will have a lowered signal to noise ratio. Translation: you have far less dynamic range in your program and your audio will be generally less interesting. The audio may sound somewhat flat.

Although there are FCP video plugins aplenty on the web I find that there are few reliable sources of Audio plugins for FCP so if anyone knows of sources for those “I’m all ears”.

I started off my career doing audio as a studio musician and then moved to video to earn a living wage. Somewhere in there I discovered you can be famous without being rich. But I’ve always retained a keen interest in having quality audio be apart of every video effort.

Onward!

FCP 7
MAC 8 core
BlackMagic Studio Pro
After Effects
PhotoShop
Illustrator
Lots of other Fun Stuff!