(Last updated: December 24, 2021)
Music forums can get quite heated about which is “better”, vinyl or CD? Seems everyone has a dog in this fight. You might as well ask which is “better”, chocolate or vanilla? “Better” is the wrong word, often shorthand for preferable, for things which are a matter of personal preference. I prefer vanilla. Better is properly reserved for how well something performs a task, like reproducing music faithfully. If one or the other does that better, you may well prefer it.
Improving your listening experience can be a lifetime mission. Other priorities may need to be put aside. If you opt for vinyl, you need to be prepared to spend more on vinyl and equipment, forsake the convenience and easy availability of the CD, and infinite choice through streaming. Basically, the audiophile listener has to decide which is better – for them.
Beware nostalgia, and tactile preference.
There are many factors that can lead people to prefer vinyl which have nothing to do with sound quality. They may like the cover artwork, vinyl may remind them of their youth, it may be a financial investment, all perfectly good reasons, but nothing to do with sound quality.
Beware equipment quality bias
Not all equipment is equally good at extracting and amplifying electrical signals. A very high-end CD transport mechanism, cost perhaps £100,000 , can bring the quality of CD well above that of of even a well-specified £20,000 turntable, and vice versa. At the rarified upper limits of affordability, the quality of “musical information retrieval equipment” may affect sound quality, arguably, even more than the choice of format. The question, “which performs better, CD or vinyl?” often depends on the level of financial investment. The best music retrieval systems I have heard are those where significantly costly attention has been paid to the stability and purity of power supply.
Beware the temperature analogy
It is commonly claimed in popular journalism that CD sounds “clinical” and vinyl sounds “warm“. If they do have sound colouration, the listener is describing an artifact of equipment, a symptom of a music system unable to render bass and treble in correct proportion. It should sound like the instruments playing during the recording, or at least what the engineer recorded. It has nothing to do with temperature. The difference is emotional. CD sounds pleasant, I can take it or leave it; vinyl sounds thrilling, it compels me to listen.
Beware the upper frequency 20kHz frequency cut off
There is a chain of dependencies: the source instruments and recording environment, the microphones, the tape machine, the mastering and lacquer-cutting process and vinyl pressing. the playback system, and lastly, your hearing.
It is sound engineering orthodoxy that human hearing cannot detect frequencies over 20kHz “The average hearing range of the human ear from 50Hz to 16Khz” – Wiki reference – mine currently fades at around 8kHz. This has been the justification of the sampling rate of the compact disc, of 44,100 times per second, because it enables a recording storage frequency range between 20Hz and 20kHz , supposedly “the range of human hearing”. The compact-disc standard assumes that there is no useful information beyond 20kHz and therefore includes a brick-wall filter just above 20kHz.
The evidence for this 20kHz hearing upper limit is based on sine-wave audiometry tests, indisputable. However it is not the whole story. There is also a substantial body of scientific research that frequencies higher than 20kHz are perceived when mixed with a wider range of frequencies, because those frequencies generate harmonics elsewhere in the frequency range, change what we hear. Additionally, sound is carried by bodily conduits other than just by the human ear. Frequencies higher than 20Khz are both perceived and preferred when present. That is also the science. There is no scientific evidence that the quality of sound is perceivably improved by cutting out higher frequencies. The best they can argue is that it is unnoticeable. Audio-scientists say different:
The human auditory system can analyze hundreds of nearly simultaneous sound components, identifying the source location, frequency, time, intensity, and transient events in each of these many sounds simultaneously and develop a detailed spatial map of all these sounds with awareness of each sound source, its position, character, timbre, loudness, and all other identification labels which we can attach to sonic sources and events. I believe that this sound quality information includes waveform, embedded transient identification, and high frequency component identification to at least 40kHz (even if you can’t ‘hear’ these frequencies in isolated form)
Musical instruments generate higher frequencies and they can be preserved
Many musical instruments generate a large frequency range, in excess of 20kHz threshold, some over 100kHz (crash symbol and claves), rimshots 90kHz, the trumpet (straight mute) generates frequencies up to 85kHz, and even human sibilance to 40kHz. (all measurements documented lab-based by CalTech, James Boyk 1997)
Of the trumpet frequency range and volume, nearly a quarter of the sound output of the trumpet is beyond the theoretical 20kHz frequency upper limit. It is relatively low in volume, but it is still there, twice the level of background noise.
Professional magnetic recording tape as a storage medium will record the full frequency output of all the instruments, in so far as the microphones in use were able to capture them. Microphone sensitivity upper limit is an area of uncertainty, some reckon up towards 30kHz, certainly a lot higher than 20kHz. So the information is there on the original master tapes.
Can a vinyl LP storage medium hold these uppermost frequencies? Mobile Fidelity were able to cut a vinyl LP with frequency range as high as 122kHz.
My Linn 242 speaker tweeters output up to 38kHz frequency. Super-tweeters can be added to lower range speakers with beneficial effect.
Not all vinyl is equally good. Mastering and lacquer cutting engineering decisions affect the final result. This frequency analysis compares a Japanese pressing with a US pressing, of nominally the same recording – frequency analysis by the excellent Youtube ANA(DIA)LOGUE:
I’ve superimposed above a screengrab of the two histograms, which is more revealing than side-by-side. The two pressing histograms are very similar until they reach the upper frequencies, where they depart. The Japanese pressing begins reducing the volume of higher frequencies from 10kHz and upwards, with little beyond 20kHz, and nothing beyond about 26kHz.
I suspect the frequency analysis reveals the characteristics of a digital/ intended for CD source and not Japanese mastering/pressing vinyl preferences, as the upper tonal range looks typical of digital/CD output – rolled off top-end highest frequencies, “because no-one can hear them”. Analogue productions will likely have remastered from original tapes, which preserve the higher frequencies.
Summary: The output of musical instruments extends well beyond the 20kHz limit, up to 100kHz. The full frequency range is capable of being recorded on magnetic tape, subject to the limitations of the microphone used, and capable of being cut into vinyl. Turntable tonearm/cartridge will retrieve these upper frequencies from the vinyl, and pass them to the amplifier and speakers, which will give the listener the full frequency range experience.
Which is why vinyl and live performance are a closer experience than a CD cut off at 20kHz. Listening almost only to vinyl, I can tell immediately when the higher frequencies are missing, not because vinyl can’t reproduce them, but because some engineer in the process has cut them off. It also explains why you should not buy vinyl records sourced from CD, or other digital sources.
Sound Quality – Inputs and Outputs
There are two important variables in reproduction of music: the INPUT – the quality of the original recording and mastering, “the source”, and the OUTPUT – the ability of the Hi Fi to faithfully retrieve and replay what was captured in the source. Good hi-fi should reproduce simply what is there, without adding anything. Admittedly, all the hi-fi systems I have listened to tend to impart a unique character of their own. Some are dry and analytical, others are punchy and bass-heavy, each more suited to particular types of music, so this may be difficult to achieve in practice, but it should still be the goal. The best hi-fi returns what was recorded, nothing else.
With a vinyl record, there are variations in mastering and pressing of different editions. You can influence potential quality for better or worse by your choice of vinyl edition. In many cases (though not always) the best sounding edition will be the first edition, closest to the original tape master. With CD/digital source, you have got what you got: the file is the same from one copy to the next. Some argue the merits of high resolution files, 24bit/192 kHz or higher, re-mastered from original tapes (if they still exist in usable condition) In my limited experience, that may close the gap, but does not change the final winner.
The output side – audio reproduction – is something you have much more control over, a wide choice of main component equipment and expense. Also power management (dedicated mains supply, balanced mains transformer, purifiers) and interconnects ( high-end plugs and cables) will also make a huge difference to sound quality, allowing you to get the most out of your main components. Don’t take my word for it – or anyone else’s – try it for yourself.
In my experience, what sounds best is “analogue end to end”: from acoustic musical instrument recorded with valve-based microphone, to magnetic tape, to cutting head to vinyl master (no digital preview), to stylus to preamp, and amp to speakers. Every step is continuous physical signal. However on a cautionary note – good hi fi wont make a bad recording sound better. Counter-intuitively, improved equipment can enable you to hear more clearly how bad a recording is. On the other hand it will often breathe new life into a good music collection, which is a big return on investment.
To some people, music is just music, an audio quality dimension does not effectively exist. I guess that applies to many people who have never heard a high quality vinyl pressing played on a highly tuned vinyl system. They are unaware of what they are missing, so can’t see what all the fuss is about.
There is a loss of musical information in the transfer from the original musical performance onto various digital storage media – CD or download-quality MP3. Simple illustration, as they say, not to scale, not everyone accepts that the difference is audible.
Vinyl stores continuous information. Digitally recorded music is a store of sampled information. CD sampling rate of 44,100 times a second not because it provides the best resolution of the sound image, but to fit the typical length of music around an hour onto the CD medium.
Irrespective of the issue of sampling, CD playback has physical problems in information recovery due to jitter and other unwanted interference. Streaming is a better solution for digital files – no moving mechanical parts – however it too merely “closes the gap” with vinyl, and mostly remains dependent on solid state circuitry.
To audiophiles, sound quality contributes significantly to the enjoyment of music, particularly when the instruments are acoustic, as in most jazz. Acoustic instruments move air, and capturing moving air and reproducing it faithfully as moving air requires a good hi-fi system. Information that can be recovered from vinyl is the very smallest detail – the “brushstrokes” – the difference between the experience of a real painting and a printed reproduction. It also includes frequencies which are theoretically outside the range of human hearing, which some studies find evoke a more favourable response when included.
Listening trials have been done to test whether people can correctly identify different grades of audio equipment – usually, a very expensive component and a cheap one. Good journalism – subtext : you are being conned – but not always very scientific – we did this test with “some people” on “some equipment” and “they” couldn’t reliably tell which was the more expensive. You must draw your own conclusions. Comparing sounds has a host of issues – noticing new information due to repeat listening, listening for the wrong things (more bass vs less bass, deeper bass vs tuneful bass, louder vs less volume) never mind expectation bias. Very often things sound “different”, but which is the better “different”?
I listen on high quality equipment to both vinyl and streamed cd and I can tell the difference between digital and analogue sources in the opening seconds. I also compare vintage pressings with modern pressings of the same recording, different generations of vintage, and the difference can be significant.
Opinion at LJC is based on physical listening tests – A:B:A:B:A:B. I do my own “science”, based on observation and experiment. More than a few times, my expectations are confounded, demonstrating the primacy of listening over thinking. Thinking often merely seeks to protect previous opinions, to prevent that house of cards that is the sum your opinions from collapse.
Often you hear the disclaimer “but I don’t have golden ears”. No-one has, or more correctly, everyone has., everyone hears the same thing. However not everyone has built up a palette of experiences, a thousand points of comparison, or a vocabulary of comparison. Think of your first glass of wine. Now think of a thousand glasses later, when you can understand, and describe what you are tasting when you taste it, because you have developed a vocabulary to articulate the difference. Audiophiles might talk about “timing”, and “musicality”.
Vinyl vs CD : LJC Verdict
In listening pleasure, vintage vinyl beats CD around 90 – 95% of the time. Every now and then, something surprises me. Much modern vinyl (pressed in the last two decades) is disappointing and offers little advantage over CD, sometimes little more than a CD transferred onto 180gm vinyl.. Bear in mind, not all modern vinyl. In just the last couple of years (2019-21), some audiophile vinyl manufacturers have made incredible improvements through all analogue processing mastering from original vintage tapes with quality pressing (Blue Note Tone Poet, Music Matters Jazz 33, Vinyl Classics Series, some Pure Pleasure).
The vinyl resurgence has brought out “lost tapes” issued on vinyl. Lost because they were not artistically up to the mark, or were radio recordings, or bootlegs, which should have stayed lost. Personally I don’t care if vinyl is “trendy” or not. There is too much vinyl that is not manufactured properly, and it really doesn’t matter if it is 180gm vinyl.
Declaration of interest: my verdict is based on having owned both a high end CD/streaming system (Linn Akurate) AND a high end vinyl system (100% tube-based). You must have experience of both in competition, otherwise you are just putting money on your own horse, marking your own homework, not a long way from “liking” your own posts. You have to listen to both candidates in sound quality competition.
The effectiveness of Hi Fi with different types of music
I have listened to some modern sound sample based electronic music on LPs – laptop composers – and if this is your thing, personally I would just get a Technics rig with the maximum bass and volume amplification. You can not recover what is not there in the first place and I wouldn’t bother with hi-fi sophistication unless you enjoy equipment and mod-ing in its own right – though that is a legitimate interest.
Life at the upper level of hifi
Once you get to the point of good source faithfully reproduced, you can move up to the next level, the music experience: artist coherence, rhythm, timing, emotion, easier to follow, freshness. It’s a good experience.
However you will have to struggle with a number of issues with vinyl, such as storage volume, cleaning, and sky-rocketing auction prices. Research has never been done on this, but I would bet money it would prove that marital difficulties ensue when a record collection exceeds 3,000. At only 2,400, my own is often the subject of robust discussion.
Since my beloved is neither a hi-fi gal or a jazz-lover, the topic returns frequently: It’s too loud, could you turn it down? You have too many records, the house is being taken over by them! Those speakers are too big, you never asked me before you bought them. All those hideous wires! Add your own.
Then there are clicks pops and scratches, because original owners of your records in the Fifties will have played them on a vinyl-destroying machine, known as a portable record player: Eight to twenty grams tracking weight arm, the autochanger, and a chipped sapphire stylus that was never changed. Not every vinyl seller is entirely truthful.
Realistically, you need to equip yourself for the best of both worlds…
Further audio research references