i want to update my perception of the cd master of ormandy's 1975 recording of holsts's the planets. the spars code is add, but it doesn't explicitly state anywhere it's remastered from analog (and you'd expect that as a marketing decision if it was). it got a lot of bad reviews. i was wondering if it's because they grafted the analog mix to digital and didn't benefit from the technology upgrade but after listening to it carefully on a system that is just barely good enough to tell, i'm convinced that it's the opposite and that the classical fans, here, are making the same dumb analytical errors that the folk fans were making.

the planets is a good workout for any sound system. it has a huge volume and dynamic range and makes full use of the orchestra, so it has a full spectrum. it's consequently actually very easy to tell that this was remastered when it was digitized. a vinyl master could not produce this dynamic or frequency range. it's actually a prime example of the kind of composition you need digital for, and you shouldn't bother with vinyl with.

take the track saturn, for instance. the track has this long build that features a prominent plucked stand up bass and peaks in a very heavy metal type horn blare. this clearly showcases both the superior dynamic and superior frequency range of digital audio technology. if you tried to mix this on vinyl, you'd have to high and low pass it and you'd lose the insane ramp up, in terms of volume difference. you can tell it's digitally mastered because it manages to get very quiet. the solution is to turn it up, and enjoy the loud parts as the loud parts they are, when they get loud.

but it's easy to imagine somebody listening to it and reacting by complaining that they can't hear it. they may have a vinyl master that they can hear better at low volumes, so they may compare it, and decide this is bad and worse. if they know little about music or music production, and they read something on reddit, they might decide they can't hear it because it's too compressed. in fact, they can't hear it because it's not compressed, it's been allowed to exist in a larger volume spectrum. compression reduces the dynamic range, it doesn't increase it. if you don't really know what dynamic compression is, but you know compressing your pillow makes it smaller, you might find the language counter-intuitive and get confused.

this recording has almost no compression, so you have to really turn up the quiet parts to hear them. if you don't, and your speakers are competing with something, the result might sound muffled or distant and the ignorant consumer might decide it's compressed. no. it's just quiet. that's dynamic range you're hearing - that's the absence of compression, not the presence of it.

if i were to do a cover of this and port the instruments to rock instrumentation, it would be with bass and electric guitar stealing the spectrum and just background keyboard parts. i might sequence some sample horns. and guess what? i'd compress the fuck out of it to get the bass part up and in your face without requiring you to turn it up. but i wouldn't have 50 other musicians making ambient sounds to listen to. i'd have a heavily gated effects section run through an eventide emulator, or something.

that's not what this, and the symphony needs that space to breathe to get a proper reproduction. i have the budget rca classics version and i even bought it in the bargain bin, new. i'd say the mix is fully digital and actually very well done, but you have to be able to turn your system up to appreciate it and your speakers have to be able to reproduce the signal. otherwise, you might get some problems in enjoyment, because your stereo can't reproduce the cd well.

i'm going to consequently shift my note of a dsd existing in the best quality section from "yes" to "maybe". dsd doesn't use conventional sampling. an sacd of a dsd digitization of the original recording might sound better if the tape is still usable, but the product for sale is a dsd to 16-bit/44.1 khz conversion (i suppose this is something like a downsample). this would still probably be better than a vinyl master, but i would guess that you can't tell the difference between the two cds, except in places where it's been manipulated by the mastering engineer. i would not consider it legitimate to run a 40 year old recording of a hundred year old symphony through digital reverb and market it as an improved master. i'd rather listen to the mastered-for-cd mix from 40 years ago.

i want record labels to clarify these decisions. i acknowledge my ears are not a spectrum analyzer, and this isn't a mathematical analysis. i could easily just rip the cd to wave and run it through some basic stats and that would tell me, but i'm confident that's not necessary. there's no way a vinyl master would have this level of dynamic range. it's actually extremely obvious. but i shouldn't have to do either of these things. the label should tell me if the master was new from the tape for the cd or built on the existing for-vinyl master. it should tell me what bit rate it was transferred at. the existing multitude of competing standards and the reality that there's no objective consumer preference means consumers should have the specs available at the purchase stage so they can pick what they want, even if they make objectively wrong decisions from the information.