How good is the science on people's ability to differentiate Hi-Res audio?

[N Fowleri]

First, I want to be clear that I am not trying to provoke heated argument, but appreciate that this is a controversial topic. Second, I am only interested in the results of properly-executed, methodologically-sound, statistically-valid double-blind experiments. Third, I am aware that some purported differences are worthwhile, but they are due to different mastering or artifacts of other differences, as opposed the actual Hi-Res vs. Redbook CD bit-depth and sample frequency. Fourth, I am also aware that just because some people can differentiate the two, it doesn't mean my over-50 ears can.

With that preamble, I saw a large meta-analysis from 2016 that showed that some people can beat random chance differentiating Hi-Res audio, but was it done well? Meta-analyses can be flawed, just like any other research. Also, I know that a novice is not in a good position to judge whether specialized research valid. (Though a novice can sometimes see when a study is clearly not valid.)

As a bonus follow-up, not necessarily related to Hi-Res audio, though I've seen it discussed relative to enhanced upsampling, is there currently any Artificial Intelligence or other advanced software that can improve audio files. Just as with computational photography, you can't guarantee that you're putting back something that was present in the file, but you can add something in based on knowledge of the world previously acquired.

I look forward to hearing what people have to say.

As full disclosure, I do own a small amount of hi-res audio, including DSD. I sometimes believe I can hear the difference, but I know that doesn't say anything about whether there is--either because of my expectations or because of artifacts in the process. Regarding AI to improve files, I doubt there is anything too magical, or I probably would have heard of it already.

 

[BDWoody]

Post in thread 'The comedy of some Hi res recordings' https://audiosciencereview.com/foru...dy-of-some-hi-res-recordings.6888/post-555591

Edit:
Archimago also did a broad survey which isn't like a peer reviewed study, but is an interesting read.

MQA Core vs. Hi-Res Blind Test Part II: Core Results

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

 

[N Fowleri]

Post in thread 'The comedy of some Hi res recordings' https://audiosciencereview.com/foru...dy-of-some-hi-res-recordings.6888/post-555591

Will do! Thanks!

 

[BDWoody]

Will do! Thanks!

That link is to a post that summarizes the meta study with the issues he identified. Basically...still nothing very convincing.

 

[DVDdoug]

For most people "CD quality" is better than their hearing. And most people (including audio professionals and audiophiles) who think they can hear a difference have not done the proper level-matched, scientific, blind, ABX tests. (The guys at HydrogenAudio "specialize" in ABX testing.)

Sometimes a high-resolution release is mastered differently than the CD. You have to do the downsampling yourself if you want to prove the difference is due only to digital resolution.

Sometimes a high-quality MP3 can sound identical to the uncompressed original or you may have to listen very carefully to hear the difference!

Pro studios record a 24/96 and a theoretical argument CAN be made for RECORDING at higher bit depth because when you when you record you have to leave headroom to prevent clipping. When you lower the level digitally you aren't using all of the bits and you're losing resolution. But I don't know that this has been shown to be audible, and most modern recordings are multi-tracked. When tracks are mixed they are summed* which increases resolution.


* Analog mixers are built-around summing amplifiers. In reality, there are level controls for each track, plus a master level control so it's a weighted average rather than a simple sum, but overall the resolution does increase.

 

[JeffS7444]

Remastering can make a big sonic difference. But even lossy audio can be difficult to distinguish from higher-res if you remove non-audio cues, such as differences in loading times, or flashing data activity lights. Try it for yourself by saving lower-res copies of your tunes, then comparing via free software ABX.

Amir's video provides details of how a trained listener can detect very minute differences under specific conditions:

My own experience with 24/96 downloads is that a few appeared to have been resampled from 44.1 kHz material, and none showed a lot of high frequency content at 20 kHz (-70 dB seemed typical). One or two showed a slight peak in the ultrasonic region which didn't appear to be musical content.

 

[DonR]

I wish @amirm would do some more testing of hi-res content. It is extremely revealing to me how shoddy some of these recordings are.

 

[Blumlein 88]

Here is the thread on the meta-study here on ASR.

AES meta analysis on audibility of hi-res

Paper freely available: http://www.aes.org/tmpFiles/elib/20160629/18296.pdf

www.audiosciencereview.com

My personal opinion is if hi-res made a substantial difference the answer to whether hi-res is good would have been determined a long time ago. The fact nearly 40 years after the introduction to CD we cannot definitively show an audible improvement means the difference if one exists is very small. Or may not exist.

So being generous, maybe some individuals with some recordings and excellent equipment can hear a slight difference. Even for them switching to CD instead would be no night and day difference or indicate hi-res is some kind of revelation in advancing the quality of recordings. When hi-res recordings are sometimes really better it is nearly always due to different and superior mastering. You can hear superior mastering with CD if it is available.

Another way to look think about it. There is one well done listening test which I won't name because it becomes controversial. This test was for lo-res vs hi-res. It also included one other difference. Using some of the finest equipment available in a special exceptionally quiet listening environment. With listeners trained to hear the specific difference beyond just the high rez difference they scored barely significant results just at the 95% level on two of three tests and just barely below that level on the third. The end result is that group could correctly identify the hi-res vs other recording 57 times out of 100 instead of a random 50 times out of 100. Do you think those extra 7 correct choices out of 100 means this is a highly significant difference that will greatly effect someone's enjoyment of music? I think it means you'll never be confident of what you heard without someone keeping score for you. And these are rather artificial conditions vs just enjoying music.

So the difference in audibility of hi-res vs regular res is somewhere between non-existent and so trivial as to be way, way, way down the list of things that matter.

 

[N Fowleri]

Thanks to everybody for your help on this topic! I now feel very comfortable that, not counting extraneous factors such as different mastering, that hi-res doesn't meaningfully change the listening experience. This is NOT to say that nobody can ever tell a difference at a rate better than chance, just that such a difference would be lost in the real world setting for me and nearly everybody else. Given the amount of time and resources devoted to the issue, and given the existence of many businesses that continue to profit from selling hi-res audio, if there is a difference, the burden is on these folks to prove it.

That said, I don't see the harm in spending a little extra to be "on the safe side." By a little extra, I mean a $2-$4 per album, but that is entirely based on my financial circumstances and superstitions. Plus, I do own a dog, and occasionally have mice, so maybe they enjoy hi-res audio more?

 

[fpitas]

Thanks to everybody for your help on this topic! I now feel very comfortable that, not counting extraneous factors such as different mastering, that hi-res doesn't meaningfully change the listening experience. This is NOT to say that nobody can ever tell a difference at a rate better than chance, just that such a difference would be lost in the real world setting for me and nearly everybody else. Given the amount of time and resources devoted to the issue, and given the existence of many businesses that continue to profit from selling hi-res audio, if there is a difference, the burden is on these folks to prove it.

That said, I don't see the harm in spending a little extra to be "on the safe side." By a little extra, I mean a $2-$4 per album, but that is entirely based on my financial circumstances and superstitions. Plus, I do own a dog, and occasionally have mice, so maybe they enjoy hi-res audio more?

the burden is on these folks to prove it

You are a funny man!

 

[REK2575]

So the difference in audibility of hi-res vs regular res is somewhere between non-existent and so trivial as to be way, way, way down the list of things that matter.

The strictly rational part of my brain knows that you are correct, @Blumlein 88 . The semi-rational part still selects the "High Res" recordings over CD-quality recordings on Qobuz.

It becomes a kind of audiophile FOMO. "Yeah, I doubt I can hear any difference at all between CD-quality and High Res, or even between OGG/Vorbis and High Res, but what if there is a difference? I don't want to miss out!"

I've subscribed to both Qobuz and Spotify for years now. I use Qobuz as my default for exactly the "reasons" mentioned above. But whenever I listen to Spotify on my good hi-fi set up, I always think, "Wow, that sounds really good."

 

[Joe Smith]

Yeah, what he said. I'm still on Spotify mostly because my wife is all comfy with their interface, and while I did trials with Qobuz and Amazon, could not hear a difference on material that I know well. So, fine for now. I continue to hope that Spotify will be forced to add a higher resolution option so at least it's available, but I'm not holding my breath.

 

[N Fowleri]

the burden is on these folks to prove it

You are a funny man!

You don't think my personal opinion will sway them?

 

[fpitas]

You don't think my personal opinion will sway them?

Give it a try. After all, the audio industry is noted for its straightforward adherence to integrity and truth.

 

[N Fowleri]

Give it a try. After all, the audio industry is noted for its straightforward adherence to integrity and truth.

A member of congress owns some audio companies. I'm surprised there isn't more cross-pollination.

 

[fpitas]

A member of congress owns some audio companies. I'm surprised there isn't more cross-pollination.

"There is no distinctly native American criminal class except Congress" Mark Twain

This has been going on a long time, obviously.