• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Audio Science review of Music ... any options or standards?

JeremyFife

Addicted to Fun and Learning
Joined
Jan 8, 2022
Messages
704
Likes
792
Location
Scotland
Hi All,
I'm new here - so let me know if this is wildly off topic or has already been dealt with.
This site is messing with my head (re-educating me) about equipment; gone are my dreams of beautiful class-A valve monoblocks costing more than my car and instead I'm thinking about teeny little 'clean' boxes - it's exciting stuff.
Getting to the point: a huge part of the fun of choosing music is in researching and finding "good" recordings. For Jazz, Classical and live music in general (Folk too I guess) this is partly about the performance itself and partly about how well it was recorded. There is also a huge amount of material, and mystique, about the mastering / re-mastering and, for vinyl, individual pressings.

However interesting this research is ... it's hardly "scientific": it's all based on opinion and consensus from a self-selecting bunch of music fans (including me).
If I'm going to choose my playback equipment based on objective measurement, is it in any way possible to apply objective criteria to the music that I choose to play on it?

Ironically, there is much more information on music recorded on vinyl i.e. bigger sample size, so I have a better chance of identifying listener bias and general BS. The Classical world especially is good for discussion and comparison of individual recordings. This is still subjective though.

In the area where there could be more objectivity - downloaded or streamed digital files in various resolutions - is it possible to have meaningful measurements or objective metrics? Is this even a thing?

Curious
 

Purité Audio

Master Contributor
Industry Insider
Barrowmaster
Forum Donor
Joined
Feb 29, 2016
Messages
9,049
Likes
12,147
Location
London
Recordings are the subjective bit, I believe some companies detail their particular technique,venue etc.
Keith
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,440
Likes
9,100
Location
Suffolk UK
Recording quality is totally subjective, as it's a Work of Art, and as such, is evaluated in the same way as books, paintings, theatre or cinema.
Of course one can report the bit rates, microphones used, recording equipment used, studios or recording venue, names of the recording engineers/producers as those are factual, but as to the music itself and the recorded quality of the music, that's opinion.

As far as recorded quality goes, there are some indicators, which labels who take these things seriously, like Nimbus, Chandos, Linn, Naim will point out in their publicity. With these labels, so you can make certain assumptions about the recorded quality, but not about the music, whether composition or performance.

To make an analogy, you can do a nutritional analysis of a meal, but it won't tell you what it tastes like.
S.
 

thulle

Active Member
Joined
Aug 31, 2021
Messages
100
Likes
134
Svante Granqvist in the Swedish Audiotechnical Society (Ljudtekniska sällskapet) made a program called Masvis to visualize the differences between different masters and give some objective basis for discussion.

Can be found with explanations here: https://www.lts.a.se/lts/masvis
Or as an in-browser version here (files not uploaded, just runs in the browser): https://andreasarvidsson.github.io/MasvisOnline/

Featured in the wikipedia article about loudness war, where they have this comparison about ZZ-tops Shar-dressed man where they end up mastering it into clipping: https://upload.wikimedia.org/wikipedia/commons/f/f7/Masvis-zz_top-sharp_dressed_man.gif

Comparing different recordings compared to masters is much more complex, but it gives some data at least.
 

kschmit2

Active Member
Joined
Oct 8, 2018
Messages
159
Likes
208
There are several aspects to consider when labeling something as a "good recording":
  • quality of the performance (which by itself can he highly subjective)
  • the venue and its influence on acoustics (mainly for live recordings)
  • technical quality of the recording
  • microphone placement (multi-mic, single-mic/one-point)
  • type of recording (studio vs. live venue)
  • in case of live venue: sometimes you want the audience's reactions captured, sometimes you try to suppress them as much as possible (again: each way can be done better or worse)
  • mixing of the recording in multi-mic setups
  • processing/Mastering of the recording (not everything you can do in processing/mastering is bad)
In the end it will most likely be a mix of some or all of the above. And it will always boil down to your own preferences. And of course to what is actually available. Performances of yesteryears cannot simply be re-recorded. So a Maria Callas recording from her earlier years can only be re-mastered but not be re-recorded. Whether that is beneficial or detrimental compared to previously available releases of quite often a personal choice.
 
  • Like
Reactions: ADU

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
4,499
Likes
5,417
Location
UK
To make an analogy, you can do a nutritional analysis of a meal, but it won't tell you what it tastes like.
That's great.

I'd rather listen to a poor recording of music I think is great than a great recording of music I think is poor. There is some value in discussing different masters of frequently remastered records, but all to often it's simply the oldest master is best, and your time is better spent finding music that is new to you that excites you.
 

ADU

Major Contributor
Joined
Jul 29, 2021
Messages
1,584
Likes
1,086
As some others have mentioned, there are some objective things you can look at in digital content, like the bit and sample rates. And also the amount of dynamic range compression used to boost the loudness in a recording, via its waveform (the Wikipedia article on loudness wars that thulle mentioned, and also the article on dynamic range compression might be one place to start for a bit more info on this). And there are also potentially some tweaks or adjustments that you can make to your gear to take a little better advantage of some of these things.

Re bit depth and sample rate, I always use a 24-bit depth and 48 kHz sample rate on my gear, because those are the standard rates for video content, and my video gear. And also the rates used for YouTube video content encoded with the Opus codec, which is what my browser always defaults to (codecs used on YouTube content can be found in their "stats for nerds"). Some people prefer higher or lower rates though depending on the type of content they are listening to, and what their gear can handle.

I haven't delved into the remastering issues with older vinyl (and digital) recordings as much. But I believe the net result of that can often be an increase in the recording's loudness. Which can also result in a loss of dynamic range.

Most of the content I listen to on YouTube is compressed for greater loudness btw.
 
Last edited:

anmpr1

Major Contributor
Forum Donor
Joined
Oct 11, 2018
Messages
3,721
Likes
6,405
... is it in any way possible to apply objective criteria to the music that I choose to play on it?
The objective way to evaluate any musical composition as it relates to hi-fi gear has been established by the latest Harman research, but few practitioners practice it. Even here on ASR, you won't see it... mostly because Amir doesn't do record reviews. It is known as the Total Harmanic Distorbution, and can be plotted on a graph.

First, count the notes in the score, then examine the runs of notes that make up repeating arpeggios. Next, note (no pun) the frequency (ie. number) of repeats. Subtract the number of bars/measures from the total number of arpeggios, minus the repeats. If this results is a negative number, substitute its additive inverse. Then, plot each score's score on a continuum. Once you do that, it will all be pretty clear, and you can concentrate on important stuff, like whether your directional cables are oriented correctly.

But seriously, there was once a music/gear critic (whom I won't name, but whose initials were David Ranada), who would take his records and time each and every section, and then compare his results for... well, for reasons that no one was quite sure of. His Excel worksheet was no doubt impressive (I guess), but what was his point? To me it was a misapplication of the idea of quantity. Simply, a wrongheaded substitution of a unified aesthetic experience for a scientistic quantitative exposition; in fine, something that didn't tell you anything important about anything at all (unless the goal was to fit the most music on to your C-90 cassette tape--then it was something useful).

For those who use their gear to actually listen to music, all one needs to do is put the record on, listen a short while, and then intuitively realize that Toscanini conducts it faster than Furtwangler. Once you know that, you can make a personal judgement about what is most appropriate to the score.

That said, there are certain 'objective rules' audiophiles should probably live by, when choosing recorded music for their gear. One rule is that if you own original Quad ESL, you probably want to avoid Blue Cheer, yet most chamber music (and even opera) will be OK.

ranada1.jpg
 

ADU

Major Contributor
Joined
Jul 29, 2021
Messages
1,584
Likes
1,086
It's been a long time since I purchased content on a physical disc. But I believe they also often used to list some info on the type of equipment that was used for the recording, production/mastering, and the media? Where A=analog and D=digital.

I think a recording that was recorded and mastered on analog gear, and distributed on a CD might be designated as "AAD", for example. But I'm not sure about this. (?)
 

dkinric

Addicted to Fun and Learning
Joined
Nov 21, 2018
Messages
666
Likes
1,446
Location
Virginia, USA

I think this is an interesting site. Compares vinyl vs cd vs download editions of a particular album, with DR, waveform and spectrum analysis of the various releases, including remasters vs original.
 
OP
JeremyFife

JeremyFife

Addicted to Fun and Learning
Joined
Jan 8, 2022
Messages
704
Likes
792
Location
Scotland

I think this is an interesting site. Compares vinyl vs cd vs download editions of a particular album, with DR, waveform and spectrum analysis of the various releases, including remasters vs original.
Some interesting stuff here, many thanks for all contributions
Just to clarify - I'm not looking for any objective measure of "good" music - totally agree that music is ART and that what matters is how it stirs us emotionally and intellectually.
That tag line for the above site "... what is the best edition of an album ..." gets to the heart of my question really well.
Ultimately, I'll listen and feel and enjoy (or not) but I was curious to know if there were tools to add to what I do when I'm digging in the virtual crates.
And, yes, this does mostly apply to older music where there are many different versions available - often on vinyl. For new music there are fewer options and I'll often just stream. Buying new vinyl is a nostalgia thing, and a financial 'thank you' to the artist (as is a digital purchase)
 
  • Like
Reactions: RHO

RHO

Major Contributor
Joined
Nov 20, 2020
Messages
1,182
Likes
1,087
Location
Belgium
Svante Granqvist in the Swedish Audiotechnical Society (Ljudtekniska sällskapet) made a program called Masvis to visualize the differences between different masters and give some objective basis for discussion.

Can be found with explanations here: https://www.lts.a.se/lts/masvis
Or as an in-browser version here (files not uploaded, just runs in the browser): https://andreasarvidsson.github.io/MasvisOnline/

Featured in the wikipedia article about loudness war, where they have this comparison about ZZ-tops Shar-dressed man where they end up mastering it into clipping: https://upload.wikimedia.org/wikipedia/commons/f/f7/Masvis-zz_top-sharp_dressed_man.gif

Comparing different recordings compared to masters is much more complex, but it gives some data at least.
I tried the tool, but it doesn't work with flac on my side. Anything I need to change in the settings?
 

ZolaIII

Major Contributor
Joined
Jul 28, 2019
Messages
4,069
Likes
2,409
Great gig in the sky. It's a good tone master nightmare how will it all play out in a recording process and if it's no good for what ever reason no amount of post production will really help. For example you can have a pounding drummer who you would like to kick... or large chorus with full blown orchestra to cover... Sometime people simply won't be at their best and won't perform as they can, sometimes they won't perform at all. And everything is possible to happen from the equipment falior to the earthquake.
Sometime even crazy things happen that end sounding good. When it's right you know it and as performer (artist) has a bigger discography there is a better chance that there will be a better sounding track with same numera.
 

thulle

Active Member
Joined
Aug 31, 2021
Messages
100
Likes
134
I tried the tool, but it doesn't work with flac on my side. Anything I need to change in the settings?
You need to download the flac codec from https://www.xiph.org/flac/download.html and after install change the path to it in settings -> codec. The tool doesn't have native flac support, but decodes to .wav before doing analyses. It was written pretty long ago. I'd kinda recommend the inbrowser version where it works out of the box.
 

RHO

Major Contributor
Joined
Nov 20, 2020
Messages
1,182
Likes
1,087
Location
Belgium
You need to download the flac codec from https://www.xiph.org/flac/download.html and after install change the path to it in settings -> codec. The tool doesn't have native flac support, but decodes to .wav before doing analyses. It was written pretty long ago. I'd kinda recommend the inbrowser version where it works out of the box.
I do have the flac codec. The tool just doesn't seem to accept it.
 
OP
JeremyFife

JeremyFife

Addicted to Fun and Learning
Joined
Jan 8, 2022
Messages
704
Likes
792
Location
Scotland
Here's the kind of thing I'm wrestling with: Daft Punk's Random Access Memories (chosen at random, widely available, feted as good to own on vinyl and clearly well suited to digital).

First off, I listen to it because I like it (also like rock, blues, folk, jazz, string quartets and full-blown baroque orchestras ... I just like music). I don't own this on vinyl, I had it on CD (gone now) and I now stream it on Amazon HD - sounds great.

#1 https://dr.loudness-war.info/album/list?artist=&album=random+access+memories
Shows a huge variation between 24 different versions

#2 https://magicvinyldigital.net/2022/...access-memories-review-lp-qobuz-tidal-amazon/
Compares 5 versions and seems to say that the 4 digital sources compared have been compressed to the detriment of the music. Vinyl has the expected higher noise and attenuated high and bass frequencies, but not the compression. (I do *not* read this as saying vinyl is better)

I'm interested in how my equipment measures. I'm interested in how my room measures. I am learning more about measuring so that I can start correcting / adjusting / EQ ing ... all this in the name of my hobby of enjoying music.

Seems like there is a gap in the measurement of the audio file that I choose to play. Why go to the effort of choosing good gear and then play any old mince (Scottish term) on it!

I understand the process of choosing vinyl; researching the mastering engineer, the pressing plant, the (subjective) opinions of others. I have no clue how do do this for digital releases, or even how to tell what version is available for streaming or download. It's a new thing to learn, and that's exciting.

Is there any appetite for a dedicated forum on this site? I'm not qualified to run it but some of you guys may be?

Disclaimer: I make no comment on the "music", that's subjective and personal and I like it that way
 

tvrgeek

Major Contributor
Joined
Aug 8, 2020
Messages
1,017
Likes
566
Location
North Carolinia
Totally subjective so maybe this forum is not the best place for it.

All that matters is how it sounds to you and if you enjoy it. No one can tell you what you hear or like. There is no measurement. There is no science. Yes, the loudness wars screwed up a lot of good music. So did the first generation 14 bit Sony mastering, so did a a lot of poor mixing engineers.

Source is what it is. A lot of the high bit rate downloads are just resamples and "fake" They sound no better. Some are well done from the original master tapes. Some are modern all digital. Read the label.
 

LuvTheMusic

Active Member
Joined
Jun 7, 2020
Messages
111
Likes
137
Although we may not have objective standards to evaluate recordings (dang, wouldn't that nice?), you *can* use a bit of the knowledge that ASR provides when doing your own evaluation.

One thing that comes to mind is level matching. This would particularly apply to different masterings of the same performance, but also has some application to different recordings of the same composition (classical music). Just knowing that louder sounds better (all other things being equal) can put you on the alert so that you aren't unduly swayed by the volume level.

If you are listening to a remastering of the same performance, you can try to level match before deciding whether the spiffy new version actually sounds better than the dusty old original. If you are listening to two different recordings of the same composition, you can't exactly level match, but you can try to get close, or at least play around with the volume control. Just being conscious of the fact that louder sounds better can help you avoid being fooled.
 
Top Bottom