• 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). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

[POLL] Hi-res music

Hi-res music?

  • I do pay for hi-res music files and I can discern an improvement over lower rate files

    Votes: 45 16.6%
  • I don't pay for hi-res music files and I can discern an improvement over lower rate files

    Votes: 20 7.4%
  • I do pay for hi-res music files but I can't discern an improvement over lower rate files

    Votes: 88 32.5%
  • I don't pay for hi-res music files but I can't discern an improvement over lower rate files

    Votes: 105 38.7%
  • Dont think I have ever heard a hi-res music file.

    Votes: 13 4.8%

  • Total voters
    271
OP
Jimbob54

Jimbob54

Master Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
8,295
Likes
10,035
"2. Let us also park the whole MQA debate, its lossiness or not, legitimacy of advertised benefits etc and just go with if you consume MQA that decodes to above 16/44.1 it is "Hi-res" for the purposes of this poll. So Tidal Masters go in the Hi-res category. "

The issue here being you can't pay for Tidal high-fi only without paying for the MQA BS. So I had to pick the "I pay for high res but can't tell difference" one to comply with the rule of this poll while it does not reflect my choice on high res at all.
Yup. Tidal, qobuz, amazon HD all have some form of hi res content accessible via the same price tier as the "lossless" (Read redbook). It would get too messy to try and unpick whether people use or want the hi res content (or mqa). So thanks for complying, I would say you picked the best answer per the definition and rules.
 

DSJR

Major Contributor
Joined
Jan 27, 2020
Messages
1,366
Likes
1,672
Location
Suffolk Coastal, UK
Can we be sure that the files we *can* hear the difference on aren't from totally different masters? This happened a lot with SA-CD releases back in the day I remember, where the SA-CD issue had been 'tweaked' a bit sonically.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,844
Likes
13,174
Location
Riverview FL
OP
Jimbob54

Jimbob54

Master Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
8,295
Likes
10,035
OP
Jimbob54

Jimbob54

Master Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
8,295
Likes
10,035
Can we be sure that the files we *can* hear the difference on aren't from totally different masters? This happened a lot with SA-CD releases back in the day I remember, where the SA-CD issue had been 'tweaked' a bit sonically.

A good question. Without significant testing and or analysis on an album by album basis. I'm not sure you ever can.

One of my pet peeves with streaming is do we get served the 'best' mix /master? No, we get what they are given.
 

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
1,315
Likes
1,880
Location
San Diego
Can we be sure that the files we *can* hear the difference on aren't from totally different masters? This happened a lot with SA-CD releases back in the day I remember, where the SA-CD issue had been 'tweaked' a bit sonically.

The easiest way I have found to answer this question is to download a 192/24 file that you know is "real hi-res" through testing or trusting the source. Then dither it down to 44.1/16 (lots of free programs will do this, Google is your friend or dB Poweramp for a few dollars) and use the Foobar2000 ABX plugin (free) to see if you can tell them apart. This way you can be assured you have the same master and you are comparing Hi-res to CD quality. I can not tell them apart blind testing. I do subscribe to Qobuz though.
 

Tks

Major Contributor
Joined
Apr 1, 2019
Messages
3,049
Likes
4,878
I just try to get anything I can in lossless. If there is a recording offered in the native sample rate and bit depth as the recording itself on offer, I'll buy it (though I'm not going to waste my time downloading Gigabytes long DSD or stuff like that). I can't tell the difference unless there's a glaring flaw of some kind.

Now the problem is. I can't tell if I am getting the native recording sample rate. And a few times (just recently) I have a suspicion some of the hi-res files are simply upsamples due to the cuttoff.

I needa ask around here how people can tell if it's some ADC anomaly or type of recording vs straight up being sold a bullshit up-sample. I've seen people periodically talk about it when looking at some spectral plots. Come to think of it, with all the smart people here, maybe we can have a thread that compiles a visual guide on how to pick out perhaps some fake upsamples versus ADC or post processing anomalies (like mixed sample rate content before the final export is made).
 

JEntwistle

Active Member
Forum Donor
Joined
Oct 25, 2019
Messages
103
Likes
128
Can we be sure that the files we *can* hear the difference on aren't from totally different masters? This happened a lot with SA-CD releases back in the day I remember, where the SA-CD issue had been 'tweaked' a bit sonically.
A good question. Without significant testing and or analysis on an album by album basis. I'm not sure you ever can.

One of my pet peeves with streaming is do we get served the 'best' mix /master? No, we get what they are given.

The problem with remasters is not just whether they are higher fidelity, but also that the mixing is often redone to suit the new engineer's taste. I have a few albums where I own several different reissues, and there are noticeable differences in the mixing; e.g.: bass levels lowered. I don't know if there really is one definitive "best"; the original is usually the definitive best mix, but the fidelity is often the worst.
EDIT: "original" being older stuff when LPs were the medium
 

Bullwinkle J Moose

Active Member
Joined
Mar 31, 2021
Messages
217
Likes
87
That is simply not true. Non-integer ratio resampling is a little more complicated, but the result is no less accurate than with an integer ratio. If you believe otherwise, you'd better steer clear of any ESS based devices.

You are incorrect

Converting an uncompressed 44.1 Khz sample to an uncompressed 96Khz file and then back to 44.1 confirms my assertion

Simply INVERT the converted file and then add it to the original in an audio editor (Both remaining at the exact same volume)

If they are identical, they will entirely cancel each other out

THEY DON'T!

They are not identical!

Only identical files completely cancel each other out when one is inverted
 

devopsprodude

Senior Member
Joined
Mar 9, 2021
Messages
316
Likes
300
Location
Beaverton, OR
You are incorrect

Converting an uncompressed 44.1 Khz sample to an uncompressed 96Khz file and then back to 44.1 confirms my assertion

Simply INVERT the converted file and then add it to the original in an audio editor (Both remaining at the exact same volume)

If they are identical, they will entirely cancel each other out

THEY DON'T!

They are not identical!

Only identical files completely cancel each other out when one is inverted
There are a lot easier ways to find out if two digital files are identical, just sayin'.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
11,844
Likes
13,174
Location
Riverview FL
Converting an uncompressed 44.1 Khz sample to an uncompressed 96Khz file and then back to 44.1 confirms my assertion

Simply INVERT the converted file and then add it to the original in an audio editor (Both remaining at the exact same volume)

If they are identical, they will entirely cancel each other out

Bit-identical, no.

I agree.

That doesn't necessarily matter.

Compare the resultant decoded analog waveform, and report back.

---

Consider a test tone that is not a sub-multiple of the sample rate.

The sample values are all over the map, depending where you look, but the properly decoded analog tone is the same throughout.

Therefore, by my extrapolation, two bit-different digital files can contain the same musical analog waveform.
 
Last edited:
D

Deleted member 30699

Guest
For the sake of this poll, let us define a few things:

1. "paying for" - if you subscribe to Tidal, Amazon HD, Qobuz lets assume one way or another you are "paying" for hi res files. Even if, like Tidal, the MQA bit comes as part of the lossless tier.

2. Let us also park the whole MQA debate, its lossiness or not, legitimacy of advertised benefits etc and just go with if you consume MQA that decodes to above 16/44.1 it is "Hi-res" for the purposes of this poll. So Tidal Masters go in the Hi-res category.

This will all go horribly wrong I am sure as we all love criticising polls- but, humour me.


Edit, let's go with the Wikipedia definition for "hi res". So CD is not hi-res for this poll

"High-resolution audio (High-definition audio or HD audio) is a term for audio files with greater than 44.1 kHz sample rate or higher than 16-bit audio bit depth. It commonly refers to 96 or 192 kHz sample rates. However, there also exist 44.1 kHz/24-bit, 48 kHz/24-bit and 88.2 kHz/24-bit recordings that are labeled HD Audio."

I believe this is an important and interesting topic. However, may I make the suggestion to define 'high-res' as music that was _mastered_ at at least 24bit (and optionally at least 88.2kHz)?

Technically it is difficult to justify 'high-res' albums/music that is up-sampled from standard definition masters or analog sources. These files usually don't contain any frequencies/bits beyond 16/44.1k and therefore we don't hear a difference. With high-res mastered files we 'may' hear a difference.

I know this excludes most 'high-res' music that should probably be sold/streamed as 'digitally remastered/restored' in standard def.
 

danadam

Addicted to Fun and Learning
Joined
Jan 20, 2017
Messages
582
Likes
826
If they are identical, they will entirely cancel each other out

THEY DON'T!

They are not identical!
And with integer ratios they are? Hint: no, they are not.

Here are diffs of a pink noise resampled to 88'200 and 96'000 and then back to 44'100, using SoX:
diff88k.png
diff96k.png
 

taner

Active Member
Joined
Jun 13, 2020
Messages
153
Likes
103
I think the poll is not about lossy and lossless formats. So, I can easily say that the term of "Hi-res" utterly wrong from right at the start. A 16/44.1 redbook file is already high resolution because it contains all information that you need, just noise floor is at -100dB region.
 

tonyo123

Member
Joined
Nov 22, 2019
Messages
44
Likes
49
I no longer worry about HiRes as I look for well mastered high dynamic range recordings. Generally, I prefer pre-loudness war recordings as originally released in CD format, not, remastered more compressed newer versions.
 
Top Bottom