• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Listening to Hi-rez & CD versions of the same music

John Atkinson

Active Member
Industry Insider
Joined
Mar 20, 2020
Messages
107
Likes
548
In the “MQA- Deep Dive” thread - https://www.audiosciencereview.com/...i-published-music-on-tidal-to-test-mqa.22549/ - there was some discussion of whether bit depths >16 and sample rates >1Fs were necessary.

I thought it would be useful to allow ASR readers to test this idea for themselves. In 2020 Naxos released an album I co-engineered and mastered of the Portland State Chamber Choir performing works by Latvian composer Eriks Esenvalds. In particular I request you audition the track “In Paradisum.”

RoonShareImage-637552159331651080.png

I can’t share the master files, as this would infringe on the copyright. However, you can stream the 24/96 version from Qobuz and the Red Book (non-MQA CD) version from Tidal. There is also a YouTube version (not sure of the provenance) at

When I mastered the CD version from the 24/96 file, I tried several different decimation filters and types of dither and noiseshaping. The CD was prepared using a combination that the music director, producer, and I independently felt sounded closest to the hi-rez original. However, there was one 16/44.1k version I prepared early on that all three of us decided didn’t sound close enough.

You can download an extract from this rejected version from DropBox:
https://www.dropbox.com/s/y2ipkrinaqgbkwm/In Paradisum Version A.wav?dl=0

I will be interested to learn if ASR readers find audible differences between the 4 versions.

(You can find more information on the album at
https://www.stereophile.com/content/recording-june-2020-ešenvalds-translations )

John Atkinson

Technical Editor, Stereophile
 

Mnyb

Addicted to Fun and Learning
Forum Donor
Joined
Aug 14, 2019
Messages
975
Likes
1,248
Location
Sweden, Västerås
We are then testing both variables at the same time ? is it >16 bits or FS >44.1 that makes the difference if any ?
 
OP
J

John Atkinson

Active Member
Industry Insider
Joined
Mar 20, 2020
Messages
107
Likes
548
We are then testing both variables at the same time ? is it >16 bits or FS >44.1 that makes the difference if any ?

If you don't hear any differences, as has been suggested by posters in the other thread, the question is moot.

John Atkinson
Technical Editor, Stereophile
 

Mnyb

Addicted to Fun and Learning
Forum Donor
Joined
Aug 14, 2019
Messages
975
Likes
1,248
Location
Sweden, Västerås
If you don't hear any differences, as has been suggested by posters in the other thread, the question is moot.

John Atkinson
Technical Editor, Stereophile

But if you hear a difference ? Then it becomes interesting again . I’ll bet on birth depth :)
 

iwantobelieve

Member
Joined
Apr 9, 2021
Messages
32
Likes
18
I’d like to have a listen, although I don’t have Qobuz, only Tidal. Is there somewhere I can purchase the hi-res album?

I would also have to look into whether it’s feasible to try and temporarily defeat the default upsampling in my Meridian DSP based system. It might be, but I’m very doubtful it can be done ‘on the fly’, in which case any listening tests, blind or otherwise would be difficult due to the time break and inevitable difference in listening position before and afterwards, etc.

Or would we be also interested to hear whether it’s possible to discern differences in upsampled redbook vs native 24/96?
 

dmac6419

Major Contributor
Joined
Feb 16, 2019
Messages
1,235
Likes
729
Location
USofA
I haven't listened but some redbook sounds better than hires and vice versa.
 
OP
J

John Atkinson

Active Member
Industry Insider
Joined
Mar 20, 2020
Messages
107
Likes
548

dmac6419

Major Contributor
Joined
Feb 16, 2019
Messages
1,235
Likes
729
Location
USofA
Ok I listened to the Tidal 16/44 and the Qobuz 24/96 version via Roon and I can't tell a difference ,also downloaded the wav file and played that through Jriver with Ozone 9 loaded it was louder than the Tidal and Qobuz version enjoyed all three by the way.
 
OP
J

John Atkinson

Active Member
Industry Insider
Joined
Mar 20, 2020
Messages
107
Likes
548
Ok I listened to the Tidal 16/44 and the Qobuz 24/96 version via Roon and I can't tell a difference ,also downloaded the wav file and played that through Jriver with Ozone 9 loaded it was louder than the Tidal and Qobuz version

I just checked the amplitude statistics of the original files with Adobe Audition:

24/96 version, perceived loudness: -19.21 LUFS
CD version, perceived loudness: -19.23 LUFS
Rejected 16/44.1k version, perceived loudness: -19.22 LUFS

So, unless I inadvertently changed the gain when I prepared the excerpt for the download - I will check - if the last one sounds louder, that might be due to the different downsampling/decimation process.

enjoyed all three by the way.

That's good to hear. It's a wonderful composition, superbly well-performed.

John Atkinson
Technical Editor, Stereophile
 
Last edited:
OP
J

John Atkinson

Active Member
Industry Insider
Joined
Mar 20, 2020
Messages
107
Likes
548
So, unless I inadvertently changed the gain when I prepared the excerpt for the download - I will check - if the last one sounds louder, that might be due to the different downsampling/decimation process.

I checked the amplitude statistics of the excerpt I prepared for the download from DropBox and compared them with those of the original file:

Perceived loudness of excerpt on DropBox: -21.54 LUFS
Perceived loudness of same region in original rejected 16/44.1k version: -21.51 LUFS

So if you are hearing this file as louder than the Tidal or Qobuz versions, there must be something else happening.

John Atkinson
Technical Editor, Stereophile
 
Last edited:

dmac6419

Major Contributor
Joined
Feb 16, 2019
Messages
1,235
Likes
729
Location
USofA
I checked the amplitude statistics of the excerpt I prepared for the download from DropBox and compared them with those of the original file:

Perceived loudness of excerpt on DropBox: -21.54 LUFS
Perceived loudness of same region in orginal rejected 16/44.1k version: -21.51 LUFS

So if you are hearing this file as louder than the Tidal or Qobuz versions, there must be something else happening.

John Atkinson
Technical Editor, Stereophile
I had Ozone 9 loaded in Jrivers,that could be the reason, I get -21.5 LUFS in iZotope RX 8 for 16/44 version so it the same as yours.
 

Jimbob54

Master Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
7,283
Likes
8,638
I checked the amplitude statistics of the excerpt I prepared for the download from DropBox and compared them with those of the original file:

Perceived loudness of excerpt on DropBox: -21.54 LUFS
Perceived loudness of same region in orginal rejected 16/44.1k version: -21.51 LUFS

So if you are hearing this file as louder than the Tidal or Qobuz versions, there must be something else happening.

John Atkinson
Technical Editor, Stereophile
Probably volume normalisation on the streaming apps.
 

Head_Unit

Senior Member
Forum Donor
Joined
Aug 27, 2018
Messages
489
Likes
257
I can’t share the master files, as this would infringe on the copyright.
Ah great thread! Though I am not set up to do this right now...maybe at another house.
- Is it not possible to post an excerpt without infringing copyright?
- Are you aware of anything recorded both in DSD and PCM in parallel? That would also be an interesting comparison.
I like the album having listened due to Stereophile mentions. Any thought to mix a surround version?
 
OP
J

John Atkinson

Active Member
Industry Insider
Joined
Mar 20, 2020
Messages
107
Likes
548
- Is it not possible to post an excerpt without infringing copyright?

I was a contractor for the sessions and the hi-rez master is owned by Naxos. Giving the Qobuz link protects their interest (and generates revenue for the choir).

- Are you aware of anything recorded both in DSD and PCM in parallel? That would also be an interesting comparison.

Acoustic Sounds has many albums available in DSD that are also downloadable in PCM from, for example, HDTracks. But provenance is always a concern.

I like the album having listened due to Stereophile mentions. Any thought to mix a surround version?

The miking was for stereo only, I am afraid.

John Atkinson
Technical Editor, Stereophile
 
Last edited:

iwantobelieve

Member
Joined
Apr 9, 2021
Messages
32
Likes
18
Acoustic Sounds has many albums available in DSD that are also downlaodble in PCM from, for example, HDTracks. But provenance is always a concern.

I’ve experienced “provenance concerns” myself when purchasing hi-res downloads from apparently reputable sources in the past. This is one of the reasons I am personally a fan of MQA, over and above any sound quality arguments for or against the format.
 

Kal Rubinson

Major Contributor
Industry Insider
Joined
Mar 23, 2016
Messages
3,528
Likes
5,497
Location
NYC/CT
I’ve experienced “provenance concerns” myself when purchasing hi-res downloads from apparently reputable sources in the past. This is one of the reasons I am personally a fan of MQA, over and above any sound quality arguments for or against the format.
This does not make sense. Can you explain why?
 

iwantobelieve

Member
Joined
Apr 9, 2021
Messages
32
Likes
18
This does not make sense. Can you explain why?

I have purchased hi res music in the past from a well known (and therefore I assume reputable) online store. It turned out not to even be the recording it was claimed to be (I noticed specifically because I was comparing the hi-res download to a CD copy I had and it was clearly an entirely different performance). When this was pointed out, they refunded, but no explanation was given as to what the other recording was, or why it was incorrect - it wasn’t just a case of mis-identifying the album on the website; they actually didn’t know what the recording was, even after investigating. My impression is that they had not attempted to be dishonest themselves but rather they, too, had been unaware that the recording they were selling was not what it claimed to be.

Aside from the claims for sound quality, MQA “authenticates” the genuine provenance of the recording you are listening to/purchasing for both buyers and sellers. In the above case I only knew the recording wasn’t as it claimed to be as I had a CD copy and happened to compare it, or I would never have known. MQA gives instant visual confirmation of whether the authenticity of the recording is in doubt.
 

Tks

Major Contributor
Joined
Apr 1, 2019
Messages
2,900
Likes
4,566
I have purchased hi res music in the past from a well known (and therefore I assume reputable) online store. It turned out not to even be the recording it was claimed to be (I noticed specifically because I was comparing the hi-res download to a CD copy I had and it was clearly an entirely different performance). When this was pointed out, they refunded, but no explanation was given as to what the other recording was, or why it was incorrect - it wasn’t just a case of mis-identifying the album on the website; they actually didn’t know what the recording was, even after investigating. My impression is that they had not attempted to be dishonest themselves but rather they, too, had been unaware that the recording they were selling was not what it claimed to be.

Aside from the claims for sound quality, MQA “authenticates” the genuine provenance of the recording you are listening to/purchasing for both buyers and sellers. In the above case I only knew the recording wasn’t as it claimed to be as I had a CD copy and happened to compare it, or I would never have known. MQA gives instant visual confirmation of whether the authenticity of the recording is in doubt.

It doesn't actually follow that MQA immunizes you from provenance perversion. I recently bought a new Hi Res file (48Khz 24bit), and it had no content above 22.1kHz to speak of even though this is a classical release, and not some electronic music that I perhaps misidentified. In fact at least on Tidal they were caught red-handed offering upsampled "master files" which weren't provided to them (GoldenSound's experiment). So I can sympathize with your plight about Hi-Res offerings.

Provenance can't be guaranteed by anyone other than the engineers who can't speak (as they're under NDA by employers and publishers), the best you can get is spectral analysis and approximating if your content was up-sampled. It gets worse when you start digging old stuff up, as many lossless offerings were produced with lossy format bits sometimes (rare early 90's and 2000's stuff). While today you have a somewhat different issue of multiple sample rates being mixed (like recording one part of a song using some other hardware that only does a certain sample rate, and taking these sections and merging them all into one master edit). And then you have electronic music that emulates older quality analogue sounds and then you can't tell what the heck is going on even by looking at the spectrum plot.
 
Top Bottom