Loudness war is a direct result of how we are consuming the music today - on the move, in cars, as background listening in noisy listening environments.
Dynamic range compression, limiting, EQ, normalization etc are the tools that are used to achieve higher perceived loudness. A musical recording has two key volume metrics - Peak volume which describes the loudest bit and RMS volume which is the average. Dynamic range is essentially the spread between these two metrics. Most recordings till the 90s left the RMS volume levels untouched.
Given that the way music consumption has changed, just higher amplification (ie overall higher listening levels) won't make the recording appealing. The entire recording is now being boosted to near peak volume levels thereby boosting the RMS levels. This results in significant compression of dynamic range where softer bits are seem loud enough in the master mix.
On top of this, most streaming services earlier used to apply an EQ boost (sometimes at select frequency ranges) to make the increase the loudness further.
However, there seems to be reversal of this loudness trend in the last few years. Mixing engineers now use a new measurement called Loudness Units relative to Full Scale (LUFS or its K-weighted equivalent LUKS) which takes in account both human psychoacoustics and actual audio signal characteristics. These are used to set relative targets for loudness both within a track and across tracks. This standard is being adopted by broadcasting industry including streaming services to normalise loudness without impacting dynamic range or listening experience. As more mixing engineers adopt these standards, the loudness war will hopefully come to an end.
Now MQA applying their own proprietary loudness, limiting algorithms while remastering so-called 'Master' tracks is another story all together. Maybe a topic for discussion in the MQA-bashing thread!
Dynamic range compression, limiting, EQ, normalization etc are the tools that are used to achieve higher perceived loudness. A musical recording has two key volume metrics - Peak volume which describes the loudest bit and RMS volume which is the average. Dynamic range is essentially the spread between these two metrics. Most recordings till the 90s left the RMS volume levels untouched.
Given that the way music consumption has changed, just higher amplification (ie overall higher listening levels) won't make the recording appealing. The entire recording is now being boosted to near peak volume levels thereby boosting the RMS levels. This results in significant compression of dynamic range where softer bits are seem loud enough in the master mix.
On top of this, most streaming services earlier used to apply an EQ boost (sometimes at select frequency ranges) to make the increase the loudness further.
However, there seems to be reversal of this loudness trend in the last few years. Mixing engineers now use a new measurement called Loudness Units relative to Full Scale (LUFS or its K-weighted equivalent LUKS) which takes in account both human psychoacoustics and actual audio signal characteristics. These are used to set relative targets for loudness both within a track and across tracks. This standard is being adopted by broadcasting industry including streaming services to normalise loudness without impacting dynamic range or listening experience. As more mixing engineers adopt these standards, the loudness war will hopefully come to an end.
Now MQA applying their own proprietary loudness, limiting algorithms while remastering so-called 'Master' tracks is another story all together. Maybe a topic for discussion in the MQA-bashing thread!
