Individual CD audio measurements database

Intro

There are sonic differences between the four waves of digitally remastered Olivia CDs. This is because the data on the discs has been remastered differently, according to the prevailing practice. This collection of data summarises some measurable differences between the releases. For some listeners mastering does matter, and since you can end up spending a lot of money on secondhand CDs then this may help some of you target your purchases better.

I am not going to say which releases are better1 because that depends on you. Your ears, your money, your way of listening, your choice. Some listeners prefer early releases with more dynamic range, many don't care, some favour modern remasters. Primary Wave do a decent modern take IMO, and the lovely presentation and bonus tracks make these versions worth getting.

Olivia reached the peak of her commercial success with the album Physical in 1981. Music is mastered differently now that it was in the 1970s. It is therefore not unreasonable to remaster the releases using more modern styles. The data here are objective measurements, and you can use this to inform yourself which issues you may prefer.

This section is about differences in media, so it is somewhat detailed and geeky. If you want to read about the albums as a whole rather than technical detail, start with Albums

The albums

The images are linked to the details of the various copies I have measured.

The analysis

I use MasVis from the Sound Engineering Society in Stockholm. It is expressly designed for this sort of comparison, and has one graph explicitly showing dynamic compression, and also tonal balance. It is also free, so you can compare my results with your own CD rips. I have used Foobar2k to extract true peak and LUFS Integrated2. I also use the older DR tool (TT DR offline 1.4 Win), because people that care aboout dynamic range are familiar with the DR program because it was part of a campaign to raise awareness of the loudness wars issue, and the CD DR database uses that deprecated method. On most of Olivia's tracks the difference between these three methods of qualifying dynamic range is not huge. MasVis gives you some visual indication of differences in tonal balance, though this is not as clear visually as the difference in dynamic range compression. Follow the link to the individual album for the technical comparison. Although preserving the difference between loud and soft is important if the original work was intended to be heard that way, it is not the only sound quality parameter that is important.

Identifying CDs

The unambiguous way of identifying the disc3 will be my photograph of the CD. The two visually different Australian pressings of Totally Hot, one labeled remastered the other not, contain identical data. Not all sonically same CDs look identical, I am hoping that all the discs that look the same are identical ;)

Equipment used

I ripped the CDs to lossless FLAC individual tracks using CUETools. I did check to screen out duff rips, though it wasn't a general problem. I treat my Olivia CDs with the reverence they deserve.

 

back to the four waves of digitally remastered Olivia CDs

page top


  1. A layman's take on mastering differences is Imperfect Sound Forever. I have listed my own view on remastering here. You may find it opinionated, obnoxious and at variance with your experience. That's why it's only linked in a footnote. I don't have every Olivia Newton-John CD ever made, so this is not exhaustive. It focuses on the main albums before Soul Kiss (which was the first album to come out on CD on first release).
  2. Like this, I set replaygain to scan per track, because I wanted track by track info in isolation. I have the feeling 4x oversampling is only accurate to within 0.5dBTP so anything indicating True Peak of < +0.5dB may well be within spec. Life is too short to do that all over again. This isn't how you would use RG if you were normally listening to whole albums, but it's right for this experiment to separate the variables track by track.
  3. When I started this project I had grand ambitions of using Discogs and/or CDDB for disc identification. I was thwarted by inconsistencies in Discogs and some missing items, and the confusion about how/where to look up a CDDB id. In the end my ambition is limited to Olivia Newton-John CDs, and at the moment I only have data on ones that I physically own or have owned. A photo of the CD is unique enough IMO, though it is an assumption. I have not searched the entire universe to be able to know that there are no ONJ CDs that look the same but have different audio data!