The No.1 Website for Pro Audio
 All  This Thread  Reviews  Gear Database  Gear for sale     Latest  Trending
Does True Peak represent a single peak or an entire song?
Old 1st February 2019
Gear Nut

Does True Peak represent a single peak or an entire song?

I'm mastering a song that I hope can be submitted to streaming services who recommend -2.0 dBTP.

According to Logic Pro's Level Meter plugin, the vast majority of my song is steady at -2.0 dBTP — but in a few different places, it jumps to -0.1 dBTP.

Is the goal to let the plugin analyze the entire song and make sure there are absolutely no individual volume jumps past -2.0? Or should I be more concerned whether cumulatively the whole song is around -2.0 average?

Thank you!

Last edited by deern; 1st February 2019 at 08:08 PM..
Old 1st February 2019
Does True Peak represent a single peak or an entire song?
The True Peak meter measures and tells you the absolute peak of an audio's waveform as its heard in the analogue realm through speakers.
Old 1st February 2019
Gear Nut

Originally Posted by CJ
the absolute peak of an audio's waveform
Thanks. So in my case, when I analyze the song, the absolute highest peaks are what should never go above -2.0? The part that makes it tricky is that it's mostly a pretty consistent song at around -2.0 and I'm trying to keep it loud without turning the whole song's volume down just to avoid that one -1.0 peak....
Old 1st February 2019
Lives for gear
Lady Gaia's Avatar
This is a decent, if slightly technical, discussion of how True Peak differs from what we normally think of as the peak level in a sampled signal.
Old 3rd February 2019
Lives for gear
SmoothTone's Avatar

You can use a brickwall limiter to handle the 1 or 2 peaks that are over your desired ceiling. The one in TDR Limiter6GE is great for this.

Bear in mind that the Spotify recommendation re -2dBTP is a recommendation, not a requirement. You will find that the vast majority of material on Spotify is much closer to 0 than that. Your songs won't be rejected if they're over -2dBTP. Spotify employs a limiter at -1dBTP when it is raising the level UP to a particular target loudness. Most of the others don't.

Also, many listeners won't be listening to loudness normalised streams. And premium users can choose from 3 different normalisation targets. So mastering to a particular loudness standard is a bit like shooting in the dark.

I would encourage you to master to the loudness that best suits your material. 'Loud enough' without compromising anything. Keep it sensible and set your ceiling to around -1dB and you'll be fine.

What it sounds like is much more important than what the numbers say.
Old 3rd February 2019
Originally Posted by deern View Post
Is the goal to let the plugin analyze the entire song and make sure there are absolutely no individual volume jumps past -2.0?
No, just the section you are playing. Some meters show small indicators holding the maximum peak throughout playback, though.

But it seems that many measures are thrown together here. An overview:
  • dB FS - Level of PCM values relative to full scale (0dB).
  • dB TP (true peak) - Level of the continuous waveform represented by the PCM values. This is the actual data, the thing to expect to come our of your DA (the music).
  • dB RMS - A root mean square average
  • LU m(omentary) - short term loudness
  • LU s(hort term) - mid term loudness measure
  • LU i(ntegrated) - long term loudness measure

The last three can also be described relative to a certain loudness reference or full scale. The suffix "FS" then appears.

All measured above can also mention the suffix "max" or "hold" telling the operator that these values memorize the largest values until reset. For example, to track the maximum true peak of a song, or the maximum short term loudness, and so on.

dB RMS and LU are roughly equivalent. LU m(omentary) is k-filtered dB RMS. k-filter is a suggestion by the EBU, and describes combination of highpass and subtle high-shelf (or a bell boosting HF depending on you interpret the recommendation).

Peak values generally do not correlate well with loudness. To approximate the perceived loudness, one has to average peaks out, "memorize" some events and "forget" many others, much like our perception does.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.

Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump