Help understand Burrrn ReplayGain WaveGain Option?
Reply #4 – 2006-12-22 01:43:21
Does ALBUM ADJUSTMENT strive to make every the AVERAGE VOLUME of every Album you burn with BURRRN about the same volume level. I assume Track Adjustment tries to make the average volume level of every track in the compilation about the same. Sort of level things off. I assume if you check this and dont check album adjustment the overall volume of the album might differ from a different CD you burn but at least the tracks on THIS CD will all sound about the same. To clarify, Track Adjustment adjusts each track to an estimated loudness perceived as equal to 89 dB SPL, so quiet tracks within an album become relatively louder, loud ones relatively quieter. Album Adjustment treats each album as if it were a single track so the relative loudness of quiet and loud tracks within the album remains the same, but the album still sounds as loud as 89 dB SPL overall. So a home-made compilation album using Track Adjustment will still sound as loud as an original album using Album Adjustment. If I've ripped from full albums, I'll usually apply Album Gain even when I'm making a compilation playlist or shuffle/random playback because tracks that are intentionally quiet stay that way. If I've purchased single tracks, I'll simply use Track Gain for those and possibly adjust it if manually I feel it ought to be quieter.What does Dither Do? Under Dither what do the Dither Without Noise Shaping With Light Noise Shaping With Heavy Noise Shaping do? Dither is relevant when you adjust the bit-depth, but by applying gain we're actually performing multiplication of 16-bit sample values by the scaling factor, which produces values with greater than 16-bit precision. If we simply round these values back to 16-bit before writing to CD (or worse truncate them) it's quite common for the rounding error to be correlated and generate new tonal distortion frequencies that weren't present in the music. To solve this, proper dither adds just enough very small random values to the unrounded values before they are rounded to ensure that the error introduced by round is spread over a wide range of frequencies. With CD audio, for flat dither (Without Noise Shaping), for each frequency the ear can perceive this noise is at around 120 dB less audible volume than a full-scale tone of that frequency. Without dither, tonal distortion frequencies which might be as much as around 96 dB below a full-scale tone (about 250 times more power per typical audible frequency bin than flat dither noise). It's always good practice to dither because when truncation distortion becomes audible it's ugly. It probably makes no audible difference which type of dither you use. Good heavy noise shaped dither is typically 15 to 18 dB less audible than flat dither even though it contains more noise power, because the noise power is concentrated in the frequencies where it is far less audible (e.g. high frequencies). However, at normal or even really uncomfortably loud playback volume, it's very unlikely that you'll notice the difference between flat and noise-shaped dither noise because they'll both be practically inaudible. If you ever lower the pitch (and quite possibly the tempo) of a track (e.g. when DJing), it's quite possible you could bring Noise Shaped dither into a more audible region.