Audio Mixing Question

[Ben Holmes]

Eli

The short answer is that audio levels for broadcast, and audio levels for music playback/web playback are very different.

Over in the UK, we use PPM audio meters to judge audio levels. It’s a scale that was developed by the BBC originally to give an absolutely consistant and legal audio level for broadcast. Because it’s logorithmic in nature, it also flattens out the levels on the needles so that you can judge the dynamic range of a programme. Basically (and the reason I am telling you all this nonsense – as I don’t think PPMs are used everywhere else in the world, more’s the shame) the PPM scale goes from 1-6, with PPM4 being what 0db tone reads. Now, in an NLE like FCP, that tone recorded at unity via a Kona card comes into the system at -18db. My maximum ceiling for audio level for broadcast is PPM6, which reads as -12db on FCP. Getting all my dialogue and music levels between PPM4-5, with peaks at PPM6 makes my broadcasters happy.

Now this sort of translates to audio levels between -18 and -12db. It’s not exact, since PPMs are not linear like the audio meters in FCP, but it’s always got me through in situations where I can’t have PPMs in the system.

The reasons for the ceiling are many and varied – and not required for all applications – especially web delivery, and DVD transfers (where I usually find broadcast levels a bit low) but these are the requirements for broadcast equipment all over the world. Eventually your digital audio will meet the analogue world, and this 12db ‘buffer’ will be required to ensure nothing comes out distorted, and the programme doesn’t make you reach for the volume control on your TV. Like it does every time I watch a US movie on the TV.

Maybe this answers your query – probably not, but I thought I’d bang on about it anyway!

Ben

Editec Broadcast Editing Ltd

EVS & FCP specialists for live broadcast.

OB Server 1 HD – Mobile FCP editing done right.
https://www.editecuk.com/OBServer2.html

[Eli Mavros]

Thanks for the response Ben. I guess since this is actually going to be for the web I have a little more leniency with the levels. I would love to learn more about audio, but it always seems to be the last thing on a very long list of things that I want to learn more about.

Best,
Eli

[David Roth weiss]

[Eli Mavros] “I would love to learn more about audio, but it always seems to be the last thing on a very long list of things that I want to learn more about.”

Eli,

As someone in DV magazine wrote recently, “there is a reason viewers are called the audience, not the vidience.” Believe it or not, the audience watching your videos will not be nearly as forgiving with anomalies in the audio tracks as they are with anomalies in the picture.

So, move audio to the top of your list.

David

David Roth Weiss
Director/Editor
David Weiss Productions, Inc.
Los Angeles

POST-PRODUCTION WITHOUT THE USUAL INSANITY

A forum host of Creative COW’s Business & Marketing, and Indie Film & Documentary forums.

[Michael Gissing]

Eli,

Don’t be lead astray by PPM speak. Broadcast levels are speced in dbfs which is “decibel full scale” and refers to digital peak meters, which are what FCP uses. Also digital record decks use this scale too.

Broadcast specs vary slightly from country to country but the golden rule is to make sure peak levels do not exceed -10dbfs. This will work for practically all PAL and NTSC countries. Reference tone on a broadcast tape should be -20dbfs for NTSC and -18dbfs for PAL.

For the web, you can use full scale if you wish, peaking up to but not going over 0dbfs. You don’t need to do this though. Always give some headroom in digital in case your meters are not accurately showing short peak excursions.

Follow the peak rules, not how loud it sounds on your speakers. Next make sure your speaker/ amplifier combination gives you plenty of volume to monitor with so that you are not flat out. Again you need headroom in your monitoring so that distortion you hear is not monitoring distortion but program distortion. Once you have a comfortable monitoring level mark it off with a texta or chinagraph if you want to be retro. Finally, don’t mix too loud. It is fatiguing and gives a false impression.

Remember the wise words of Stanley Kubrick – Films are 60% audio and 40% picture. That from an image obsessed old deaf bugger.

[David Roth weiss]

Michael,

I happened upon a funny quote earlier today that I thought you would like…

“No one likes audiophiles–even other audiophiles.”

– from Esquire’s the Rules – A Man’s Guide to Life

David Roth Weiss
Director/Editor
David Weiss Productions, Inc.
Los Angeles

POST-PRODUCTION WITHOUT THE USUAL INSANITY

A forum host of Creative COW’s Business & Marketing, and Indie Film & Documentary forums.

[Michael Gissing]

“No one likes audiophiles–even other audiophiles.”

David, I agree. Audiophiles hear what doesn’t exist, don’t hear what actually exists, fail double blind listening tests consistently and offer expert opinion without knowledge. To pros like me they are a pain.

Let them buy monster cable at ten times the price.

[Ben Holmes]

[David Roth Weiss] “there is a reason viewers are called the audience, not the vidience.””

Absolutely. I used to QC movies for a broadcaster, and we were always taught that many more people complain about audio than video. Making audio levels consistent, maintaining a tight dynamic range and getting a good balance between different elements (music, speech, fx) should be top of your list when finishing a project. I have a harder time getting things past the audio supervisors I work with than the video guys – and given the audience’s views, that’s how it should be.

Although peak levels (and I forgot to mention dbfs, apologies) are higher for the web, people are still listening via similar means. The low levels cause just as much problems as the high ones…

Ben

Editec Broadcast Editing Ltd

EVS & FCP specialists for live broadcast.

OB Server 1 HD – Mobile FCP editing done right.
https://www.editecuk.com/OBServer2.html