hello! i have some underground tracks from a deceased artist that were never released retail. apparently the songs were transferred in studio using something called protools and were given to me in wav format. however, these wavs are not in proper cd format - a few are in a 32 bit depth as well as a few being in 24. i also have one in proper 16 bit depth, but it's in 48000hz sample rate.
now, using audacity i can downsample these to proper cd format (16/44000), however, am i supposed to apply something called dithering in order to avoid artifacts/distortion? i've read a bit on google about the subject, but i'm not sure how it applies to my situation... i don't know if these songs were already dithered, whether they've already been converted from even another bit depth/sample rate before now, etc. in other words i don't know what's already been done to them. they are however of course wav's so i'd like to preserve their quality, as well as convert them properly. should i dither in tjis situation, or just render them to cd format straight up?
i don't know if these songs were already dithered, whether they've already been converted from even another bit depth/sample rate before now, etc.
No... Dither is applied when you reduce the bit depth so even if it was dithered when going from 32-bit to 24-bit,
it can be dithered again.At 16-bits (or better) you are unlikely to hear the dither or the effects of dither so
as a practical matter it doesn't matter. If you are going down to 8-bits it would probably sound better with dither.
i don't know if these songs were already dithered, whether they've already been converted from even another bit depth/sample rate before now, etc.
No... Dither is applied when you reduce the bit depth so even if it was dithered when going from 32-bit to 24-bit, it can be dithered again.
At 16-bits (or better) you are unlikely to hear the dither or the effects of dither so as a practical matter it doesn't matter. If you are going down to 8-bits it would probably sound better with dither.
great, thx for the reply.
Even if they are dithered during previous conversions, this doesn't matter for the new conversion (even for the 16-bit one). If you don't know if there is dither, you may see it at the end of a fade-out if you zoom up vertically until you see the staircases.
If you want to do it "right", apply dither when going from 24-bit to 16-bit. The correct order is: Resample first, then dither and finally truncate to 16-bit.
Dithering is no more than a very very silent noise that randomize the rounding error happening from bit depth reduction.
If the dither is missing, a slow fade-out may end in a ugly fuzzy glitchy thing short before it finally hits the flatline. But that only gets audible at very high volume and in very rare cases. That is the distortion than can occur.
There are a lot of commercial CDs where the dither is missing and where multiple sample rate and bit depth conversions were done during production. There are CDs too (especially compilations), where you can see (but not really hear) a stack of multiple dithers (i.e. 16-bit->gain change->16-bit...) ;D
Take more care to the resampling. If that is done in a poor way (using a crappy resampler), the audio quality can be totally destroyed.