Understanding peak audio levels
by Wesley Dysart
on
Aug 12, 2009 at 3:53:18 pm
Hi,
I'm trying to wrap my head around audio levels in regards to post production, things like distribution masters, broadcast masters, etc.
I have an audio background (music) and we always peak everything to 0dBFS, or -0.2 or something.
I've been told for DigiBeta & and all broadcast/distribution master tapes tone should be laid down at -20dBFS. However, I've recently learned that -10 should be the actual peak of the audio. I've read about -18dBFS also being used in some cases. Can anybody link me to a good article or perhaps just break this down for me real quick? Why lay tone down at -20 if the audio will go up to -10. If the original creator of the program material had his project go all the way up to zero, then wouldn't lowering his volume by 10dB cause some really quiet segments to potentially be cut off? (because of the -98dB limit of 16bit).
And if -10dB is to be treated as 0dBFS then when does -18dBFS come into play?
Any info or links to info on this would be greatly appreciated. I was taught one thing at school, ran into other things at work and I think I get it, but the -18 thing is definitely an unknown right now.
So far this is what I think I know. 0dBFS for CD audio, -20dBFS for bars & tone (unless it's DV then -12dBFS) and -10dBFS for program material on master digital video tapes.
Re: Understanding peak audio levels by Alan Smith on Aug 13, 2009 at 3:09:08 pm
This is generally defined by the network broadcasting the show, not the producer who produced the show. We produce shows that air on two different networks and have different requirements for peak levels. One network requires -20db tone and allows peak levels to hit -6db. The other network requires -20db tone but limits peak levels to -10db with normal voice levels bouncing at/around -20db. So, levels for broadcast are largely defined by the network and they have production guidelines that you can obtain.
Re: Understanding peak audio levels by Wesley Dysart on Aug 13, 2009 at 3:59:30 pm
So there would potentially be different master tapes for different networks depending on their required specs? That makes sense I suppose. So if a client were planning on sending out many copies of a master tape the burden would be on them to provide the right specs to us. If we knew the networks they were using we could double check the specs with each network.
Thanks for the help guys, I get a lot of on-the-job training but I try to supplement that with my own research. So far this forum has been very helpful.
Re: Understanding peak audio levels by Terry Mikkelsen on Aug 13, 2009 at 7:14:49 pm
Don't confuse tone with peak levels.
Tone was set at 0VU in the analog days. You can have program audio louder than 0VU and it is natural to do so. Analog also had an inherent compression attribute with tape saturation.
When we moved to digital, there is no compression to the format when you exceed its maximum abilities. So, it was determined to lower the average level to allow some headroom in the program levels. This became a highly subjective issue. There is no absolute (except as those determined by each house) and so that is why you see different specs.
So it breaks down like this:
1. Find out what your final delivery format is and what levels they require (both average or tone levels and the maximum).
2. Lay your tones at the appropriate level. Lets say -16dBfs for this example. Now you know that if you need to goto analog for some reason, you will have -16dBfs on your digital machine and your analog needs to get set to 0VU.
3. Adjust your speaker volume to a comfortable level while listening to your -16dBfs tones.
4. Without adjusting your speakers, adjust your program volumes to be roughly the same comfortable volume that your tones were.
5. Now scan through your program and make sure you do not exceed the maximum level specified by your final house. (If needed, add some gentle compression and/or limiting)
