Since this topic has picked up again, I’d like to come in with some facts about digital audio which I hope will alleviate any concerns that people might have.
I’ll even put the TL:DR first
- Recordings should be normalized (have the volume increased so that the maximum is within a few db or so the digital max) as a matter of standard best practice before it is distributed. This can be done completely for free with open source audio programs (like Audicity).
- The recording quality is sufficiently high that any normalization in volume will still result in significantly higher than cd quality.
The longer explanation:
Digital recordings have a hard limit to their total loudness. If the input gets louder, the recording will clip and distort. The margin between this digital limit and the recording volume is called headroom.
Engine records with a comfortable amount of headroom, because in the wild world of DJing - sometimes things get loud. Someone may jump on the mic, or maybe there’s some wild scratching all of a sudden - all which can increase the peak volume of the recording. In addition, the recording volume is disconnected from the main volume knob. This is to prevent things like turning up (when the party really gets going) or turning down (when the police come) the mix from having an affect on the recording itself.
All digital recordings also have a limited dynamic range - that is, there is a limit volume range between the maximum and minimum volume that the recording could have. Anything below this is technically digital noise. This measurement is done in bit depth. CD Quality recordings in 16 bit have a dynamic range of 96dB. Engine records in 24 bits, giving it a 144dB dynamic range.
The point is this - normalizing the volume does lower the dynamic range of the recording as well. However, the volume adjustment would have to be increased 48 db (144-96) before it dips below CD quality. The average headroom - which probably runs anything from 16-12dB means that these recordings are still well above CD quality audio when normalized.
For the future - there are several ways to tackling this issue. One would be to simply put a normalization process into Engine. That way, the recordings would be more or less ready to go as soon as they are processed and normalized on the players. Another option would be to be a user controlled volume offset - which is what the majority of people are asking for. Keep in mind however, that this manual control means that if you set this too high, clipping may occur. Also -the track would still need to be normalized anyway.
So in summary - I hope that this helps explain why this isn’t such a dire issue in terms of audio quality. The conversation can continue as needed, but I’d like to flag this issue as the “solution” as the it answers the question “How to Adjust”. The answer is “To normalize it.”
Please feel free to refer other threads or questions on the topic to this post.