Hi All,
Great to see you guys care about these issues ! Just to clear up a few points raised earlier in this thread:
- There's no question that the Quobuz master was the source for all the released versions of "Random Access Memories", including the vinyl
- The CD isn't a poor transfer, it's identical in terms of dynamics to the Quobuz release. The higher bit-depth and sample rate simply mean the Quobuz version has lower noise floor and wider frequency response than the CD
- My video demonstrating increased DR readings on vinyl isn't a specially chosen extreme example, it's just a case where it was easy for me to get the files make the comparison. What it shows is typical, in my experience - most vinyl rips of moderately limited material (e.g. DR 8) will show an increase in peak level (and therefore DR) of at least 4 dB, because of phase rotation. More heavily limited releases (DR6 and lower) may show even bigger increases. It's easy to do this test yourself, just apply an all-pass filter (non phase linear) to a digital source
- That's not to say that there aren't vinyl releases that used more dynamic masters and sound better as a result - there certainly are. But it does mean you can't figure out which ones they are by measuring with the TT meter, unfortunately
- The TT meter doesn't use the ITU standard, it pre-dates it by several years. It uses raw RMS and peak level measurements, to measure the so-called "crest factor". I recently developed a new plugin, called Dynameter, which does use the ITU standard, to measure a similar property called the short-term peak to loudness ratio (PSR). Unfortunately it's every bit as vulnerable to being fooled by phase rotation distortion as the TT meter. (Fwiw for material with balanced EQ content they both give very similar readings)
- Phase rotation is pretty benign sonically - most listeners are unlikely to notice it. In my opinion it doesn't give the impression of more dynamics, if anything it reduces them because of transient smearing
- In general, vinyl doesn't "need" a different master - heavily crushed digital masters can be cut as-is - for example Metallica's "Death Magnetic". The loudness of the cut is not determined by peak level, but rather by factors like loudness and running time. Since really "loud" digital sources will be turned down as a result, there's no point in crushing it that much in the first place, so you might as well take advantage of the extra available peak headroom, and use a less limited master
Summary: Many (probably the majority of) current vinyl releases are still made from the same digital source as the CD. In my opinion the apparently more "dynamic" sound of vinyl in these cases is often caused by a combination of: EQ differences; valve gear etc. in the playback path; additional processing during the cut including RIAA filters, de-essing, elliptical EQ etc; minor pitch variations; added distortion; crosstalk; surface noise; clicks etc etc.
And sometimes, by genuinely more dynamic mastering - meaning simply less peak limiting, usually.
Hope that helps !
Ian
