Dynamic range (DR, PLR, PSR)

Definition

The term dynamic range refers to different quantities depending on the context.

• Dynamic range of a system: the difference between the maximum undistorted level and the noise floor, DR_system = L_max − L_noise (dB). For a linear N-bit digital system with a full-scale sine wave it equals 6.02·N + 1.76 dB (98.1 dB at 16-bit).
• PLR (Peak to Loudness Ratio): PLR = TP_max − L_I, the difference between the maximum true peak (dBTP) and the integrated loudness (LUFS). It measures in a single number how far the peaks stick out above the average level of the whole song.
• PSR (Peak to Short-term loudness Ratio): PSR(t) = TP(t) − L_S(t), the difference between the true peak and the short-term loudness in the same 3 s window, measured continuously. It describes the microdynamics moment by moment; the minimum value indicates the most compressed point.
• DR (index of the TT Dynamic Range meter): for each channel the song is divided into 3 s blocks; for each block the peak and the RMS are calculated (with 3 dB added, so that a sine wave has an RMS equal to its peak); the 20% of blocks with the highest RMS are taken and DR = 20 · log10(P_2 / RMS_20) is calculated, where RMS_20 is the root of the mean of the squares of the RMS values of those blocks and P_2 is the second-highest block peak. The final value is the average of the channels rounded to the nearest integer (DR6, DR12).

PLR, PSR and DR measure the microdynamics (peak-to-average-level ratio); LRA measures the macrodynamics (loudness variation between sections).

Parameters and units

• Unit: dB for PLR, PSR and DR (DR is expressed as an integer).
• Typical PLR values: very loud EDM and trap 6–8 dB; pop and hip hop 8–10 dB; rock 9–12 dB; acoustic and jazz 12–18 dB.
• Typical PSR values in choruses: 6–8 dB in very loud masters; 9–12 dB in open masters.
• Typical DR values: DR4–DR7 in very loud modern masters; DR8–DR11 in dynamic pop and rock; DR12 and above in jazz and classical music.
• Dynamic range of formats: 16-bit 98 dB, 24-bit 146 dB (theoretical values).

Musical function

Microdynamics determines the impact of transients: kick, snare and the attacks of guitar and piano. With platform normalization the integrated loudness is equalized, so a master with a higher PLR reaches the listener with larger peaks at the same volume and sounds more open; a master with a low PLR sounds dense but less incisive. The right value depends on the genre: an EDM drop calls for density, a jazz trio does not. PSR and PLR let you measure this choice and compare it with references.

Practical example

Master of a pop song and its less limited version:

After normalization the two masters have the same loudness, but in the open master the transients rise 2 dB more above the average level.

Common mistakes

• Confusing DR and LRA: a song with an LRA of 10 LU can have a PLR of only 7 dB if every section is limited.
• Calculating PLR with the sample peak: the value comes out lower than it should; PLR uses the true peak.
• Comparing DR values with PLR values: different algorithms give different numbers.
• Chasing a high DR in a genre that calls for density.
• Judging dynamics by ear on references with different loudness.

How to check it

• App analysis: the PLR (true peak − integrated LUFS) row uses the ×4 true peak and the BS.1770 integrated loudness; the PSR (true peak − short-term max) row is the difference between the maximum true peak and the maximum short-term loudness, that is the PSR in the loudest part of the song. Crest factor (peak − RMS) and Loudness Range (EBU 3342) complete the picture.
• Dedicated meters: Youlean Loudness Meter shows PLR and PSR in real time; the TT Dynamic Range meter and its derivatives calculate the DR index.
• Comparison with the reference: analyze the master and the reference in the app; a PLR within ±1.5 dB indicates a similar density.