Yesterday, the UK newspaper The Guardian published an article about high-resolution music. The article – entitled How much difference is there between MP3, CD and 24-bit audio? – starts with a leader:
“Debates rage over whether hi-res music is a gimmick. Three Guardian writers put four music formats — and their ears — to the test”
This is another example of The Guardian’s shoddy tech reporting. They take a hot-button item that can get them a lot of hits, and approach it in the worst possible manner.[1]
If The Guardian were to examine, say, the quality of meat sold in supermarkets, would anyone consider it valid if they had Tesco (the leading UK supermarket chain) provide the samples for their test? Probably not. But when they examined high-resolution music files[2], they went to Linn Records, a purveyor of such products. (I have nothing agains Linn Records; I have a number of their classical releases, many of which are wonderful.)
And, rather than perform a blind test – a bit of research on the internet would have shown the journalists that this is the only valid way to compare different music formats – they had someone from Linn Records play them in order from the most compressed file (128 kbps MP3) to what was expected to be the best (the high-resolution file). And they listen to them on a “high-quality system and speakers.”[3]
While their conclusions were mixed – they made it clear that they couldn’t always distinguish between the different formats – one comment stood out: “there’s a distinct quality difference between the kind of compressed, middling MP3 commonly downloaded from the major platforms.” This is disingenuous; if the person here is talking about 128 kbps MP3, not one “major platform” uses that bit rate any more, or has in many years.
I’ve seen many studies that involve blind testing of music formats, such as this study that I recently reported on, and they all conclude that, even among experienced listeners, it’s nearly impossible to distinguish between CD-quality music and high-resolution files. The study I link to above even includes a number of people who are musicians, or people who work with audio; in a survey after the study, those people were no better than the others, even when they felt relatively sure of their choices.
There’s a big difference between the high-profile investigative journalism The Guardian publishes – such as the Snowden leaks – and the tripe they pass off in the rest of their newspaper (and website). Much of The Guardian today looks like a blog, with a lot of stories about dating, weight loss and cooking. Their tech reporting often looks like that as well. Interestingly, they have an excellent classical music critic, Tom Service, who could have been a part of this test; I wonder why he wasn’t. Did he already have an opinion on the subject that went against what The Guardian wanted to publish?
This Guardian article reads like an advertorial. Lazy journalists didn’t want to take the time to examine the questions around high-resolution music files objectively, so they got a company who sells the product they’re reporting on, and pimped that company’s products, saying:
“…listening a very high-quality set of Linn speakers…”
“…on Linn’s superb hi-fi equipment…”
It’s clear what the agenda was here; this was not an article about high-resolution, but about Linn Records. The Guardian took the wrong approach to an issue that concerns (an admittedly small number of) consumers, and, rather than try and be objective, just wrote what would please a specific company. Take it for what it’s worth.
Read why I think that high-resolution music is a marketing scam.
Update: As a reader pointed out in the comments, the Guardian podcast allows Gilad Tiefenbrun of Linn Music make several misstatements. He suggests that MP3 or AAC are proprietary formats that are not “open.” Discussing one service that, on shutdown, led to the loss of music by purchasers, he equates the non-openness of specific formats as being an issue.
Unfortunately, he’s quite wrong. The Virgin Media service he discusses didn’t cause problems for people’s music libraries because of a format, but because of DRM. Any MP3 or AAC files you have now – at least since Apple dropped DRM – will be playable for the foreseeable future. He talks about being “tied in with either certain equipment or certain service providers.” Nope, unless there’s DRM, that doesn’t happen; and DRM is almost entirely gone for music.
But, it’s fair to say that high-resolution files do tie you in with certain equipment; if you don’t have something that can play the files, then you can’t listen to them. But this person conflates a number of issues in a few sentences: open source, DRM and a variety of formats.
Let us not forget that Linn Records is trying to sell a product; and a higher-priced product at that. So it’s clear that any person from a company like this is giving a marketing speech, nothing more. Technically, there’s no real reason for these files to cost more. If there is a premium for bandwidth, let them charge a pound or a buck more. Not £8 more per album, as here, compared to downloads in CD format.
And, while I’m at it, why does Linn sell a 24/96 disc for £18, and Hyperion can sell files of their records, at the same resolution, for only £10.50, the same price as a CD? (Some of Hyperion’s high-resolution releases are the same price as CDs, some are cheaper, and some are more expensive.)
I think Linn is gouging the market, because the only people buying these files are those who have already invested a lot of money in their audio systems.
- A nit-pick, but an example of the lack of seriousness of the article. They didn’t look at “four music formats,” but only three; they listened to MP3 files at two different bit rates. ↩
- The Guardian article states the following: “There is even debate what actually constitutes hi-res. As Linn’s managing director Gilad Tiefenbrun explains, “there’s confusion over what is and isn’t hi-res music. Is CD hi-res? Perhaps a high-quality MP3? Or does it have to be 24-bit music? For us, hi-res music is the 24-bit studio master – the original recording the artist made, from which all other files and formats are made.”” There’s no confusion at all. High-resolution music files are files at a resolution higher than that of CD; technically, the Red Book CD format, which is 16-bit, 44,100 kHz. They could be 16-bit, 48,000 kHz, which could be studio masters, in the case where a recording was not made in 24-bit. And studio masters can be higher than 24-bit; they can be DSD – Direct Stream Digital – which is used for SACD. ↩
- Any such statement should tell us: a) how much that system cost, so we can compare it to our own audio systems; and b) whether it used oxygen-free, unidirectional speaker cables. ↩
I would classify myself as an experienced listener. Perhaps even an Audiophile. I use 2 Arcam monoblock ams connected to a preamp and arcams “high end” d33 dac. Speakers in use are psb synchrony ones.
I was unable to discern a difference between 24bit and/or 96/196khz audio and the redbook format. It’s all about the mastering in my opinion. CD redbook format is a great format for distributing digital audio. And very Rarely can I hear a difference between a 256 AAC file encoded from the same redbook mastering.
Thanks for that information. I agree that mastering is one of the most important steps in the process, one that is often ignored.
I would classify myself as an experienced listener. Perhaps even an Audiophile. I use 2 Arcam monoblock ams connected to a preamp and arcams “high end” d33 dac. Speakers in use are psb synchrony ones.
I was unable to discern a difference between 24bit and/or 96/196khz audio and the redbook format. It’s all about the mastering in my opinion. CD redbook format is a great format for distributing digital audio. And very Rarely can I hear a difference between a 256 AAC file encoded from the same redbook mastering.
Thanks for that information. I agree that mastering is one of the most important steps in the process, one that is often ignored.
A nut pick perhaps, but the best tests need to be double blind, where neither the test subject nor the tester know which is being tested.
A nut pick perhaps, but the best tests need to be double blind, where neither the test subject nor the tester know which is being tested.
I think it’s a natural assumption that higher quality files must result in a better listening experience. An analog comparison might be DVD on 4:3 CRT and blu-Ray on a 1080p flat screen. It might also require “better equipment” to take advantage of the higher bandwidth file. But tests can challenge that assumption and it may be that our ears are not as critical as our eyes. You also point out that the article was not what was promised. Good post!
It’s a natural assumption, but one needs to find the point at which that assumption no longer holds true. The SD vs. HD example is pertinent; we can see a difference, but only under certain conditions. Apparently, at more than 1.6 x the diagonal of your TV screen, you can’t make out the difference any more.
This question will arise with 4K TVs as well. I see no need for a 4K TV at the size of my current TV (I think it’s 47″). But if, one day, I get a 60″ TV, which is likely in the future, then the difference between 1080 and 4K will be visible.
I was thinking about something this morning, after I wrote the article, while reading Swafford’s new biography of Beethoven. The audiophile vs. music-lover thing is just like the Enlightenment vs. Romanticism. The audiophiles are looking at the numbers, and convincing themselves that higher numbers are better. The music lovers who don’t care about audiophile stuff are more interested in the feeling they get from the music. Sound vs. music; science vs empfindsamkeit.
I think I’ll try and formulate some of that in a future article.
One observation that supports the point of your last paragraph:
Most musicians, of any caliber, are _not_ into high end audiophile systems at all.
They are quite happy with more or less regular hifi systems.
Maybe the reason is, again, that they “feel” the music. Maybe they only need the loudspeakers to trigger the music in their head, so to speak. For example, they enjoy technically rather bad historical recordings much much more than any audiophile is able to do.
I think it’s a natural assumption that higher quality files must result in a better listening experience. An analog comparison might be DVD on 4:3 CRT and blu-Ray on a 1080p flat screen. It might also require “better equipment” to take advantage of the higher bandwidth file. But tests can challenge that assumption and it may be that our ears are not as critical as our eyes. You also point out that the article was not what was promised. Good post!
It’s a natural assumption, but one needs to find the point at which that assumption no longer holds true. The SD vs. HD example is pertinent; we can see a difference, but only under certain conditions. Apparently, at more than 1.6 x the diagonal of your TV screen, you can’t make out the difference any more.
This question will arise with 4K TVs as well. I see no need for a 4K TV at the size of my current TV (I think it’s 47″). But if, one day, I get a 60″ TV, which is likely in the future, then the difference between 1080 and 4K will be visible.
I was thinking about something this morning, after I wrote the article, while reading Swafford’s new biography of Beethoven. The audiophile vs. music-lover thing is just like the Enlightenment vs. Romanticism. The audiophiles are looking at the numbers, and convincing themselves that higher numbers are better. The music lovers who don’t care about audiophile stuff are more interested in the feeling they get from the music. Sound vs. music; science vs empfindsamkeit.
I think I’ll try and formulate some of that in a future article.
One observation that supports the point of your last paragraph:
Most musicians, of any caliber, are _not_ into high end audiophile systems at all.
They are quite happy with more or less regular hifi systems.
Maybe the reason is, again, that they “feel” the music. Maybe they only need the loudspeakers to trigger the music in their head, so to speak. For example, they enjoy technically rather bad historical recordings much much more than any audiophile is able to do.
A worse blunder is on the associated tech podcast. Here the Guardian journalist claims (to paraphrase) there is “some sort of licensing arrangement involved with both the MP3 and AAC codecs”.
In the case of AAC this is absolutely untrue. It is open source.
Apart from his ignorant adoption of the popular misconception that AAC somehow “belongs” to Apple he is trying to use this calumny as part of his argument in favour of open source because such “proprietary” codecs may one day be discontinued by their owners. This is a perfectly good point to make, but you need to have the facts right and there are good examples where something similar has already happened.
If you didn’t hear the podcast, it’s at http://www.theguardian.com/technology/audio/2014/aug/20/mp3-about-to-die-tech-weekly-podcast. (To speed up this link, you can go straight to the forty-first minute at which the relevant part of the interview starts.)
That’s not true:
License fees are due on the sale of encoders and/or decoders only. There are no patent license fees due on the distribution of bit-stream encoded in AAC, whether such bit-streams are broadcast, streamed over a network, or provided on physical media.
http://www.vialicensing.com/licensing/aac-faq.aspx
MP3 licensing is murkier though.
Apple Lossless is open source. But, yes, AAC does not and has never belonged to Apple. It was developed by a consortium, and is part of ISO standards.
Thanks for the heads-up about the podcast; I’ll give it a listen.
A worse blunder is on the associated tech podcast. Here the Guardian journalist claims (to paraphrase) there is “some sort of licensing arrangement involved with both the MP3 and AAC codecs”.
In the case of AAC this is absolutely untrue. It is open source.
Apart from his ignorant adoption of the popular misconception that AAC somehow “belongs” to Apple he is trying to use this calumny as part of his argument in favour of open source because such “proprietary” codecs may one day be discontinued by their owners. This is a perfectly good point to make, but you need to have the facts right and there are good examples where something similar has already happened.
If you didn’t hear the podcast, it’s at http://www.theguardian.com/technology/audio/2014/aug/20/mp3-about-to-die-tech-weekly-podcast. (To speed up this link, you can go straight to the forty-first minute at which the relevant part of the interview starts.)
That’s not true:
License fees are due on the sale of encoders and/or decoders only. There are no patent license fees due on the distribution of bit-stream encoded in AAC, whether such bit-streams are broadcast, streamed over a network, or provided on physical media.
http://www.vialicensing.com/licensing/aac-faq.aspx
MP3 licensing is murkier though.
Apple Lossless is open source. But, yes, AAC does not and has never belonged to Apple. It was developed by a consortium, and is part of ISO standards.
Thanks for the heads-up about the podcast; I’ll give it a listen.
Hey Kirk – I’ve watched this debate play out on your site many times. I’ve even contributed a bit to it in the past. While I agree with some of the technical merits of your argument, I feel the cynical slant you take doesn’t do your observations justice. I think we have to accept the fact that not all people have equal abilities when it comes to hearing. Just because one person cannot hear what another person can hear, doesn’t mean a difference doesn’t exist between formats. I do testing for various friends who create handmade, out of my budget, headphones, cables and amps. (We’re talking $5000 headphone amps.) I can clearly hear the difference between especially 256 or 320 kbps tracks and CD quality tracks. Most often, I can clearly hear the difference between CD quality and 24 bit tracks, especially if they are 96/196 kHz or above. And it doesn’t take a $5000 amp to hear this. My own equipment is far less expensive …
I can describe these differences in detail, right down to the increased depth of sound stage, the spacial relationship between instruments, and the fully rounded tonal quality of the individual instrument or vocalist. Roll your eyes you may, but some of us (not likely many) can hear these differences. They are there, and they add great pleasure to our musical enjoyment. To consistently dismiss this ability – worse to discredit the people who possess it – is beneath the otherwise high standard you seem to have set for your blog.
Thanks for continuing to rollout for us your insights about the world around you …
I don’t consistently dismiss it. In the article I link to at the end of the above post, I say this:
“While there may be a small number of people who have the necessary audio equipment and good enough ears to hear this difference, those people are few and far between.”
I don’t deny the possibility, but I think it takes, as you you say, $5000 headphone amps or equivalent other hardware.
Just curious – can you hear these differences in a true blind test?
Hey Kirk – I’ve watched this debate play out on your site many times. I’ve even contributed a bit to it in the past. While I agree with some of the technical merits of your argument, I feel the cynical slant you take doesn’t do your observations justice. I think we have to accept the fact that not all people have equal abilities when it comes to hearing. Just because one person cannot hear what another person can hear, doesn’t mean a difference doesn’t exist between formats. I do testing for various friends who create handmade, out of my budget, headphones, cables and amps. (We’re talking $5000 headphone amps.) I can clearly hear the difference between especially 256 or 320 kbps tracks and CD quality tracks. Most often, I can clearly hear the difference between CD quality and 24 bit tracks, especially if they are 96/196 kHz or above. And it doesn’t take a $5000 amp to hear this. My own equipment is far less expensive …
I can describe these differences in detail, right down to the increased depth of sound stage, the spacial relationship between instruments, and the fully rounded tonal quality of the individual instrument or vocalist. Roll your eyes you may, but some of us (not likely many) can hear these differences. They are there, and they add great pleasure to our musical enjoyment. To consistently dismiss this ability – worse to discredit the people who possess it – is beneath the otherwise high standard you seem to have set for your blog.
Thanks for continuing to rollout for us your insights about the world around you …
I don’t consistently dismiss it. In the article I link to at the end of the above post, I say this:
“While there may be a small number of people who have the necessary audio equipment and good enough ears to hear this difference, those people are few and far between.”
I don’t deny the possibility, but I think it takes, as you you say, $5000 headphone amps or equivalent other hardware.
Just curious – can you hear these differences in a true blind test?
Just to be clear, I can easily hear the difference on far less expensive equipment. I have participated in multiple true blind tests, and more often than not have still heard the difference. I think it’s less about the equipment and more about a person’s hearing, or in some cases, their willingness to learn to listen differently. I’ve taken friends and had them re-listen to tracks they know well, but have never heard in higher resolutions than 256 k, and watched their eyes brighten, or jaw drop when musical nuances suddenly surface. I think it’s sad the most of the world has settled for a fraction of the quality of music that was originally produced. Worse yet most of them think 256 k is all there is … All that said, however, I have to admit – it’s first about the music, and next about its audio quality.
Fair enough. I don’t deny that some people have the hearing to be able to tell these differences. But the blind tests that have been done suggest that you’re part of a tiny minority.
But the main point is, as you say, it’s about the music first.
Just to be clear, I can easily hear the difference on far less expensive equipment. I have participated in multiple true blind tests, and more often than not have still heard the difference. I think it’s less about the equipment and more about a person’s hearing, or in some cases, their willingness to learn to listen differently. I’ve taken friends and had them re-listen to tracks they know well, but have never heard in higher resolutions than 256 k, and watched their eyes brighten, or jaw drop when musical nuances suddenly surface. I think it’s sad the most of the world has settled for a fraction of the quality of music that was originally produced. Worse yet most of them think 256 k is all there is … All that said, however, I have to admit – it’s first about the music, and next about its audio quality.
Fair enough. I don’t deny that some people have the hearing to be able to tell these differences. But the blind tests that have been done suggest that you’re part of a tiny minority.
But the main point is, as you say, it’s about the music first.
I like how your footnotes are reserved for snark : “oxygen-free, unidirectional speaker cables”.
At least someone reads the footnotes. 🙂
I don’t want to overencumber the text with snark; that’s not my goal. But I can’t help adding a bit in the notes.
I like how your footnotes are reserved for snark : “oxygen-free, unidirectional speaker cables”.
At least someone reads the footnotes. 🙂
I don’t want to overencumber the text with snark; that’s not my goal. But I can’t help adding a bit in the notes.
High Res Music? I love it and now I know why the typical “music box” has such an amazing effect on me: http://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0095464
A lot of people cite that study. It seems, however, to be the only one of its kind, and hasn’t been replicated.
High Res Music? I love it and now I know why the typical “music box” has such an amazing effect on me: http://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0095464
A lot of people cite that study. It seems, however, to be the only one of its kind, and hasn’t been replicated.
Popular music sound quality has not changed much from the early 1970’s. Listen to anything today and it sounds very similar to 40+ years ago. But listen to something 40 years ago from 1970!
I think there will be something that comes out in the future that enhances sounds, just like some sunglasses today enhance color. It will very noticeable and more pleasurable to the ear. At least for the last 40 years sound quality has been incredible. Listen to some early Elton John.
For the best sound today there are some inexpensive headphones (under 40 euros) that do make the difference between night and day.
Popular music sound quality has not changed much from the early 1970’s. Listen to anything today and it sounds very similar to 40+ years ago. But listen to something 40 years ago from 1970!
I think there will be something that comes out in the future that enhances sounds, just like some sunglasses today enhance color. It will very noticeable and more pleasurable to the ear. At least for the last 40 years sound quality has been incredible. Listen to some early Elton John.
For the best sound today there are some inexpensive headphones (under 40 euros) that do make the difference between night and day.
Really, high resolution and high end audio may deliver a better experience but just a bit more. And just to some people and just while playing some specific kind of music. And under some specific conditions.
I admit the best I have listened are under US$3000.0 audio systems and yes they sound great but my guess is that they sound great because their speakers are very carefully placed in a specially designed room.
No man with a $3k audio system put it randomly in his living room full of furniture and stuff all around with speakers obstructed, wrongly placed, etc, etc Not at all. People with that kind of money dedicate an entire room to music listening. They even build one and they pay architects for that.
I have listened in audio rooms from retailers and yes they sound great but hey nothing in that room is left to chance.
We live in a very advanced time of audio reproduction.
Going from bad to good you can have audio from your laptop or cellphone/tablet speakers (poor audio), then bluetooth speakers (a bit better) then boomboxes (at least they are stereo), then mini systems (very good quality) then integrated amps, receivers, then separates, etc, etc.
When you have a mini system (yes, the all in one cheap plastic crap you buy in bestbuy or walmart) you are getting already very good audio quality and you hardly need more.
You get a receiver? good but don’t expect great experience if you don’t place everything carefully.
You get separates? preamp? amp? dac? super expensive turntable? A pair of US$10000 speakers? good! but now you need a dedicated room specially built for music listening, otherwise you are wasting money and the end result while excellent, then it is difficult to say you enjoy 10000 times more than music coming from a cheap plastic mini system.
There are so many variables that you never end. For the most part it might be a waste.
The whole high end audio business is niche precisely because joy does not relate to how much money you invest on it.
Really, high resolution and high end audio may deliver a better experience but just a bit more. And just to some people and just while playing some specific kind of music. And under some specific conditions.
I admit the best I have listened are under US$3000.0 audio systems and yes they sound great but my guess is that they sound great because their speakers are very carefully placed in a specially designed room.
No man with a $3k audio system put it randomly in his living room full of furniture and stuff all around with speakers obstructed, wrongly placed, etc, etc Not at all. People with that kind of money dedicate an entire room to music listening. They even build one and they pay architects for that.
I have listened in audio rooms from retailers and yes they sound great but hey nothing in that room is left to chance.
We live in a very advanced time of audio reproduction.
Going from bad to good you can have audio from your laptop or cellphone/tablet speakers (poor audio), then bluetooth speakers (a bit better) then boomboxes (at least they are stereo), then mini systems (very good quality) then integrated amps, receivers, then separates, etc, etc.
When you have a mini system (yes, the all in one cheap plastic crap you buy in bestbuy or walmart) you are getting already very good audio quality and you hardly need more.
You get a receiver? good but don’t expect great experience if you don’t place everything carefully.
You get separates? preamp? amp? dac? super expensive turntable? A pair of US$10000 speakers? good! but now you need a dedicated room specially built for music listening, otherwise you are wasting money and the end result while excellent, then it is difficult to say you enjoy 10000 times more than music coming from a cheap plastic mini system.
There are so many variables that you never end. For the most part it might be a waste.
The whole high end audio business is niche precisely because joy does not relate to how much money you invest on it.
My friend is an Audiophile he spent £2000 on a Gucci shirt because of the higher thread count
Personally I have had best results when I spend the bulk of my budget on good speakers and headphones (The most important link in the chain)
An average amplifier is good enough for me
technically an average amplifier can produce sound with very low levels of distortion I can’t see the difference on my oscilloscope or hear a difference with my ears
What percentage of audiophiles own oscilloscopes I wondered
A 320kbs mp3 or Flac sounds the same to me
A £100 dac sounds the Same as a £500
Loudspeakers require better amplifiers because distortion is greater at higher sound levels but the tiny signals needed to drive headphones will suffer less from distortion this is why expensive headphone amplifiers are a complete con job
So I have learned to spend all my spare money where it matters most
Recently I was very pleased with the sound quality from Sony 1000xm3 wireless headphones With only an iPhone is used as the source AAC
My friend is an Audiophile he spent £2000 on a Gucci shirt because of the higher thread count
Personally I have had best results when I spend the bulk of my budget on good speakers and headphones (The most important link in the chain)
An average amplifier is good enough for me
technically an average amplifier can produce sound with very low levels of distortion I can’t see the difference on my oscilloscope or hear a difference with my ears
What percentage of audiophiles own oscilloscopes I wondered
A 320kbs mp3 or Flac sounds the same to me
A £100 dac sounds the Same as a £500
Loudspeakers require better amplifiers because distortion is greater at higher sound levels but the tiny signals needed to drive headphones will suffer less from distortion this is why expensive headphone amplifiers are a complete con job
So I have learned to spend all my spare money where it matters most
Recently I was very pleased with the sound quality from Sony 1000xm3 wireless headphones With only an iPhone is used as the source AAC