If you thought ‘hi-fi’ was marketing mumbo jumbo…
According to a study carried out by Queen Mary University in London, people are actually able to distinguish the difference between standard audio and hi-res audio quality. The study compared data from more than 12,000 different trials across 18 studies.
In the study, participants were asked to determine the difference between samples of Jazz and Classical music in different formats.
Dr. Joshua Reiss from QMUL’s Centre for Digital Music in the School of Electronic Engineering and Computer Science explains why the study was conducted. Reiss says that the motivation cames from constant discussions in the music industry regarding whether hi-res audio is necessary, or likely to make a any difference among consumers.
Reiss says…
“Our study is the first attempt to have a thorough and impartial look at whether high res audio can be heard. We gathered 80 publications, and analyzed all available data, even asking authors of earlier studies for their original reports from old filing cabinets. We subjected the data to many forms of analysis. The effect was clear, and there were some indicators as to what conditions demonstrate it most effectively. Hopefully, we can now move forward towards identifying how and why we perceive these differences.”
The findings of the study are somewhat interesting, as it validates the entrance of Tidal— Jay Z’s high-fidelity streaming platform — into the music streaming market. After launch, the service experienced widespread skepticism as many said that there is no need for Hi-Res audio, because it doesn’t sound any different.
Those people opted for rival services Spotify or Apple Music, which have a much lower sound quality and cost half the price. It is believed that Tidal hasn’t experienced as much success as its competitors in the music streaming market because people didn’t see the value in paying for a platform that had high-fidelity audio quality and lossless sound. However, it seems as though people can in fact distinguish the difference between sound quality.
This means that there is hope for Tidal, with price playing a more determining factor.
Reiss says…
“Audio purists and industry should welcome these findings – our study finds high-resolution audio has a small but important advantage in its quality of reproduction over standard audio content.”
((Image by Tess Watson, Creative Commons, Attribution 2.0 Generic (CC BY 2.0))
where da double blind data is at?
+1
You need to define hi-res.
People can certainly distinguish between good and bad mp3s, and, in some cases, between good mp3s and 16/44.
So, everything should be 16/44.1 today.
However, nobody can tell the difference between 16/44.1 and “better” in blind tests (“better” in quotes because 24/96 doesn’t make sense to consumers, while it is necessary during recording and mixing).
It seems like streaming uses way too much bandwidth. If we’re all being pushed to streaming, I can see why the quality is low – you have to have that data sent to you every single time you listen to a song?
Better to just download once, at the highest possible bitrate – how about 24/192 (although 32/96 or something like that would be technically smaller. My understanding is that going from 96 to 192 really just adds data way up there way past the ability of the human ear to hear, even subconsciously – something like 40K to 80K. Going from 48 to 96 gets you extra data from 20K to 40K, and it’s said that that range is perceived subconsciously. Going from 16 to 24 and from 24 to 32 is more samples, more bits, more pieces of information in a given time frame.
Storage costs – SD cards – are very very cheap now. Consumer recording devices should all be recording at 24/96. That’s smart phones, that’s point and shoot video cameras, etc. Over the last 20 years, from the introduction of DV to now, audio quality hasn’t gotten much better, even though it should.
“Consumer recording devices should all be recording at 24/96.”
Would be nice, but that won’t happen for a long time. Two channels of uncompressed CD-quality are still relatively new in high-end cameras.
“Over the last 20 years, from the introduction of DV to now, audio quality hasn’t gotten much better”
You could easily argue it’s gotten worse — much worse.
What I don’t get is that people are obsessed about resolution when they use sh!tty speakers and even worse converters.
If people wish to improve their sound, they need to:
1) Buy better speakers or headphones
2) Buy better converters
3) Buy their music instead of streaming.
In that order!
#2 and #3 are super essential: Computerspeakers, white earbuds and cheap soundcards are absolutely terrible, and there’s no way consumers who use any of those can tell the difference between bad old mp3s and 24/96.
…yeah well, that would #1 and #2…
Here’s a different comparison test you can try, between low and high quality MP3 and CD-quality WAV:
http://www.npr.org/sections/therecord/2015/06/02/411473508/how-well-can-you-hear-audio-quality
Of course people can hear the difference, but the question is: do they care? And the answer is: NO. I am a musician for 25 years, I play two instruments, I have a degree in classical music and jazz, I arrange and compose for big bands, symphony orchestras, pop and latin bands etc. I can transcribe ANYTHING you every heard in ANY kind of genre… YET I only listen to music via MP3 and Youtube. I tried Spotify premium and yes, I could hear the better sound quality, but the truth is: I DON’T FUCKING CARE! This sound quality argument is so fuckin meaningless! Because NOBODY CARES if the sound is better than an MP3!!! Even a professional, qualified musican like myself!
Well, the majority don’t care but there are, thank Christ, some that do care & so should you as you are banging on about how qualified you are & yet declaring your disdain about music reproduction. Let’s just go back to mono, scrap FM because who needs it, AM is good enough. While we’re at it, let’s also get rid of pesky sound engineers because we don’t need their expertise or them to bust a gut trying to get the best possible sound in a recording because it’s just not worth it. Oh yeah & fuck it – I’m going to put a clothes pin in my record player cartridge instead of a stylus cos who gives a shit how it sounds.
Misinformation alert.
Did anyone actually read the AES paper in question? The conclusions it draws are statistically dubious, and don’t really vindicate ultra-high bitrates, rather than seize upon a set of statistically insignificant and practically unlikely listening scenarios.
As always, the scientific reality of the ‘benefits’ of high bitrates are overstated; it is infinitely better to invest in good speakers, good amplification, good signal-path and acoustic treatment. 24-bit files are simply unnecessary, with correct dithering. Samplerates above 48k is simply unnecessary, with correct anti-aliasing.
“it is infinitely better to invest in good speakers, good amplification, good signal-path and acoustic treatment. 24-bit files are simply unnecessary, with correct dithering. Samplerates above 48k is simply unnecessary, with correct anti-aliasing.”
Truth. And you only want 48 for video.
” it is infinitely better to invest in good speakers, good amplification, good signal-path and acoustic treatment.”
And once you have the above, you will understand why it IS necessary to have 24 bit
Not really a “Study”, but a meta-analysis. This is where you run statistics on the results of a large number of previously published studies and try to find trends and correlations based on the aggregate. They performed no new testing here. Yawn.
Actually, I don’t give a shit about you idiots who want to listen to shitty quality music. Knock yourselves out, dumb it down to whatever you want. Use two tin cans & string for speakers for all I care, but keep the hell out of my choice to listen to music the way it was intended to be heard by the artist & recording engineer.
if anyone want to have blind tests you can try my app at
https://play.google.com/store/apps/details?id=com.bjsoundtrst&hl=en