DRC-Meter

DRC-Meter is a small software application that can be used to approximate the amount of dynamic range compression that has been applied to a digital audio file. It can hence be used to measure the effects of the loudness war. It can be used on any .mp3, .m4a, .wav, .aif, .flac or .ogg audio file. It will not affect the files in any way.

Note that dynamic range compression has nothing to do with data compression (such as mp3 encoding). If you do not know the difference between the two, you probably do not need this tool.

Only compression of the microdynamics are measured. Compression of the macrodynamics (if any) are ignored.

A Java Runtime Environment (JRE) version 5 or greater. the latest JRE can be downloaded at https://www.adoptium.net.

DRC-Meter.zip (version 1.0.3, 1.68 Mb)

Double-click on DRC-Meter.jar to launch the application.

Just choose one or more audio files in the file dialog that opens and click ok. You can also choose folders, in which case the application will recursively scan the folders for all supported audio files it can find. The results of the analysis are displayed in the main window. Each audio file may require some seconds to some minutes to be analyzed.

Lauch the application using

    java -jar DRC-Meter.jar <file> [<file> ...]
   

Note: lauching the application from the command line without specifying any file will launch the graphical user interface.

The amount of dynamic range compression is given for each audio file by a number. The greater the number, the higher the dynamic range compression (Note well: this measures the dynamic range compression and not the dynamic range). Here's how I interpret the results (your interpretation may will vary as dynamic range compression is highly subjective):

• -1.5 - 1.0: likely not compressed, or only slightly compressed
• 1.0 - 2.5: weak compression
• 2.5 - 4.0: medium compression
• 4.0 - 5.5: strong compression
• 5.5 - 7.5: aggressive compression

The applications uses heuristics and produces an approximation of the dynamic range compression. Never take any of the results firmly. Furthermore, DRC-Meter has a limited precision: when applied on several audio files with exactly the same amount of dynamic range compression, the results typically have a variation of +/- 1.0, and results with +/- 2.0 or more occasionally happen. Hence if for example the average result of an album is greater by 3 or 4 than the average result of another album, you can be pretty sure that the former has been more compressed than the later. However, if a single song scores for example 1.5 more than another one, it may not mean anything at all.

More generally, the results will almost never magically tell you the exact parameters of the compressor plugins that was used to compress the dynamics, and may even be quite far from the real values. The results are only comparable to what you can infer by carefully looking at the audio file's waveform. If you know how effective it is to look at an audio waveform, you know how effective DRC-Meter is. Note however that:

• DRC-Meter is not simply based on an estimation of the loundness. In fact it is (mostly) insensitive to the amplitude of the audio file (as long as the peaks are above -16dB!)
• DRC-Meter is not a measure of the amount of dynamics. It measures the amount of compression of the dynamics. This is important to understand. A file that scores high (highly compressed) can still have a lot of dynamics if it had even more dynamics before the compression. Conversely, a file that scores low (weakly compressed) may have few dyamics if it already had few dynamics before the compression. The two notions, dynamic range and dynamic range compression are independant, although in practice, they are frequently correlated
• DRC-Meter is not just measuring the amount of clipping. It works with several kinds of dynamic range compression, including clipping.

• DRC-meter only measures compression of the microdynamics, not of the macrodynamics
• DRC-meter uses heuristics and hence has a limited precision. While it constantly produces the same result for the same audio file, results for different audio files that have been compressed by the same amount typically exhibit variations of +/- 1, with occasional variations of +/- 2 or more.
• The precision gets quite low on short audio files (less than 1 minute).
• DRC-Meter works poorly with multi-band compression.
• Only .wav, .aif, .mp3, .m4a, .flac and .ogg files are supported.
• Error handling is poorly implemented.
• DRC-Meter only works with files sampled at 16bits, 44.1 kHz (CD quality). Other files may produce innacurate results due to the use of a poor automatic resampling algorithm.
• When an audio file has both uncompressed and compressed parts, DRC-Meter tends to base the result only on the part that is the most compressed, rather than on the average.
• DRC-meter only works when the dynamic range compression was the last processing step applied to the audio file (apart from dithering). This is the case for almost all files ripped from a CD. Any further processing, even a simple frequency equalizer, may drastically reduce the accuracy of the results (most of the time toward lower values). Data compression (mp3, ogg) is usually safe as it only results in minor inaccuracies. Hence DRC-Meter should still produce accurate results with files downloaded from iTunes or Amazon.
• A consequence of the previous limitation is that DRC-Meter is ineffective with recordings taken from analog mediums (such as tapes or vinyls) as there is almost always some processing involved during playback with such mediums.
• DRC-Meter measures the amount of dynamic range compression, not the amount of data compression such as .mp3 bitrate.