CD vs. 24-bit streaming - Sound of the past vs. sound of the future (Turntable tips)
HTML-код
- Опубликовано: 14 фев 2023
- Would you rather have 16-bit CD audio, or 24-bit streaming which is 256x better? Will you hear any difference? Check out the Audio Masterclass Music Production and Sound Engineering Course at bit.ly/3W3tpKo
DAVID MELLOR'S MUSIC
David Mellor's music on Bandcamp - davidmellor.bandcamp.com/
David Mellor's music on Spotify - open.spotify.com/artist/6OkaD...
Available on all good streaming services
The CD revival - So wrong on so many levels - • The CD Revival - So wr...
CREDITS
Studer A820 - JacoTen CC BY-SA 3.0
Dolby 361 - PJ CC BY-SA 3.0
Sony 1610 - Akakage1962 CC BY-SA 3.0
Sony DAT - DRs Kulturarvsprojekt CC BY-SA 2.0
Compact disc - Kulshrax CC BY-SA
Spotify graphic - Spotify
EQUIPMENT USED TO MAKE AUDIO MASTERCLASS VIDEOS
CAMERA - Sony Alpha A6600 amzn.to/3uj7Dtq
COMPUTER - Apple Mac Mini M2 Pro amzn.to/47tW6Wg
MICROPHONE - Sennheiser KH 416 amzn.to/3QWWvuO
ALTERNATIVE MICROPHONE - AKG C414 amzn.to/3SLqCad
BUDGET MICROPHONE CHOICE - Rode shotgun amzn.to/3G0ni36
MICROPHONE PREAMPLIFIER - Comica LinkFlex AD2 amzn.to/40SMHFT
TELEPROMPTER - Padcaster Parrot alternatives amzn.to/40A55TJ
TELEPROMPTER SOFTWARE - PromptSmart Pro promptsmart.com/
AUDIO INTERFACE - Behringer UMC202HD amzn.to/3ugj6db
HEADPHONES - Sony wired amzn.to/3sEjlht
KEYLIGHT- Amaran 100d amzn.to/40A5Gor
FILL LIGHTS - Switti RGB amzn.to/3ufhOPu
SOFTBOX Phot-R alternatives amzn.to/3QGoTjJ
LIGHTING STANDS Phot-R alternatives amzn.to/3QWXOKe
MONITOR - Samsung UJ59 amzn.to/3MEoBsk
MONITOR ARM - Invision alternatives amzn.to/40SLqi5
MIC ARM - Neewer amzn.to/49Ob0sD
TRIPOD - Velbon alternatives amzn.to/46gu2oa
GREEN SCREEN - Elgato alternatives amzn.to/3SG2F3L
DESK - FlexiSpot E7 Pro amzn.to/3QVIWvS
VIDEO EDITING - Apple Final Cut Pro www.apple.com/final-cut-pro/
IMAGE EDITING - Adobe Photoshop amzn.to/3FZ7Ww0
AUDIO EDITING - Avid Pro Tools amzn.to/3QA5Wz7
NOTEBOOK - Leuchturm 1917 hardback amzn.to/47zonKX
PEN - Sakura Pigma archival pen Sakura Pigma archival pen amzn.to/47zz3cP
OSCILLOSCOPE - Hameg HM 203 alternatives amzn.to/47tWN1O
BEAUTY - Philips 3000 nose hair trimmer amzn.to/40BdLJo
Disclaimer: Videos on this channel may include affiliate links and may feature software that has been received free of charge with an NFR (not for resale) licence, equipment that has been loaned or donated by a manufacturer or distributer, or equipment that has been bought for less than full retail price.
In my experience, how a CD is mixed/produced has a much bigger impact on how it sounds than whether it is 16 or 24 bit. But maybe that is not the point of this video.
I agree 100% on that one, I much prefer the audio on my old vinyl, even though it's long been converted to 16 bit digital and de-clicked, I never listen to the actual vinyl, as in stick it on a turntable, as every time I do so it will be worse than the last time - that's called mechanical wear and tear.
Most modern audio seems to be mixed for a mobile phone speaker, or a crappy bluetooth speaker. HiFi is not in the shops anymore, so why mix for something not generally available.
The (audio) world is going to pot.
@@DW_drums Yes, most dacs don't even bother to read more than 4 to 6 bits. They skim through the rest and fudge it. So cd's have so much potential that most people are not even using now and then they want us to go on to another format? How many bits are most dacs going to really read of a 24 bit? Give me a well recorded 16 bit cd on a good dac on a good amp/speakers system.
@@rft2001 No. You would hear dramatic quantization noise. Even 8 bit quantization is clearly audible, even by a deaf person. So no, this cannot be true. A totally different story is from 16 bits to 24 bits. 16 bits already give you 96 dB dynamic range, and this means that in a normal home listening environment, where "natural" background noise (the fridge, the traffic, the buzzing of some switching power supply adaptors...) is rarely less than 20 dB, you would need to listen at 116 dB in order to hear any faint quantization noise. Let me say that 116 dB is UNBEARABLY loud.... So, 16 bits are already more than enough to grant ZERO quantization noise in normal listening environments. 24 bits is overkill. And indeed, even the best DACs cannot truly resolve 24 bits, because even the best designed ouput stage cannot reach high enough SNR. The best ones can reach about 130 dB SNR (again, hugely overkill in any case), equivalent to about 21-22 bits.
@@paulstubbs7678 I still play all mine. As I get older my hearing gets worse and our days left on the planet get less so I say enjoy your LPs mate. There’s nothing to beat the experience of putting on a record, cleaning it and looking through the sleeve notes whilst it plays.
I usually find that I will play several records once I’ve got into the groove (pardon the pun). I think it’s because once you’ve gone to all the effort, you’re more inclined to listen to more. Digital music is easier but I don’t end up listening to it as much when at home.
Sorry, I went off at a tangent. Enjoy your vinyl mate… that’s what it was made for!
People are constantly making this argument but coming from someone who’s been into HiFi for over 30 years I can tell you I listen just as attentively and get the same enjoyment from the act of putting on a CD and in the meantime it actually sounds good. Believe me I’ve done the vinyl thing and I know what it takes to actually obtain good sound from vinyl. Fact is even if you have a great table, great cartridge, great tone arm, great phono stage, great cleaning system and a zero stat that’s as close to perfect as you’ll ever get from vinyl and it’s still limited in resolution and dynamic range compared to the CD. Not to mention vinyl holds a static charge no matter what you do. It’s just the nature of the material. Static not only causes cartridge interference but it also attracts micro particles which creates further noise. You can limit this with an anti static treatment like a zero stat but that only helps the problem. It doesn’t completely negate it. I love vinyl for nostalgic reasons but there’s a thousand reasons why vinyl realistically sucks. I can still put on a CD and enjoy the liner notes without all the negative aspects you get from vinyl which is a lot. I’m a firm believer in hard copy media because as a music enthusiast I absolutely enjoy every aspect of an album right down to the cover art. You just simply give up to much with streaming but from a sonic standpoint even streaming is better than vinyl. It’s certainly less engaging though. I’ll give you that.
I used to be an audio designer. I designed some of the best commercial digital to analogue converters, amplifiers and speakers I or my customers had ever encountered. I hold 3 international patents on different aspects of audio reproduction technology.
As you rightly say, CD is good enough. Not only because 16 bits is sufficient to store enough levels for even 20 year old ears, let alone 60 year old ears, but also because the magic of D/A upsampling actually gives you more virtual headroom than the original signal contains, due to the fact that musical instruments only produce smooth waveforms when viewed above about 6 KHz or so, most of them much lower than that. So 20 bit D/A is actually quite standard from a 16-bit signal.
So, you are absolutely right. CD is not only good enough, if you (as a human) are listening to the very best audio equipment (everything through to the speakers, very important) you will not be able to hear the difference between a CD reproduction and 24-bit streaming reproduction of the same original audio master.
As a comment below from "tweaker man" says, the mastering (and recording) is much more important than the medium.
I totally agree with this. When the engineers designed CD, they decided that doubling the absolute maximum frequency any human could hear would be enough. They actually went beyond that with 44 kHz sampling , so as stated it's not only good enough. It's better because it; Uses less data, therefore smaller files on your computer or phone. The other thing worth mentioning is that many people listen to music on inadequate speakers, or headphones. As stated above, this is very important to the quality of the sound. I'm my opinion, this is the reason records are making a comeback because people are actually listening to them on a good stereo instead of on earbuds from a phone.
@@garycard1826 44.1Khz for audio, 48Khz for video, insert technical reasons here.
I do always record anything in 24bit, 48Khz. Size is not the big issue.
Yes, a fair bit of modern mastering/editing sucks, that's why I like the sound off my old records.
I'm not sure what they are editing for - iPhone speakers?
Shannon and Nyquist spent enough time verifying this specification.
Upsampling doesn't increase the amount of information stored on the CD, as your comments appears to imply.
It's all about the mastering of the music, I have Lots of redbook CD's, that are mastered extremely well & they sound outstanding , I have lots of CDs that sound poor, same goes for sacd and Vinyl, so for me it's all about how good the recording & mastering is, no matter what format its on, great video👍
See my post above. I agree. I feel bad for some of my unnecessary bashing in the past. So much to type, but much of the recordings aren't "bad". Some composition choices, performance quality have been more of a problem. I heard it all nice and clear, with all the emotion and or lack thereof. Get the noise and data right and it all opens up. Pretty entertaining stuff when the overall sound is not confusing. Dancing Queen is my test track (can induce wincing, ticks, mute), it's as enjoyable as I think it should be. For better or for worse, I haven't heard anyone sing like that since. 😆
Clean dynamics cost money, folks can't afford it, so I get that pop does get the full dynamic range treatment. Those songs with aggressive compression sound interesting, and I wonder if that was the effect they were after (otherwise sounding bland on lesser gear).
Unfortunately it’s very easy to find a dynamic range compression on CD than on a vinyl. Only the very first CDs has no it, and most of SACDs. Streaming is the worst: the latest re-issues made by people who learned few plugins for a sound editing software, and still the compression, cause despite of radio death, now music is mastered for airpods and sport activities.
There used to be a website that listed a title and the different variations of each catalog number for more common artists. It was helpful in tracking down specific cds based on how they were mastered. 50s and 60s music is hit or miss for one hit wonders and also for mono versus stereo versions in digital formats. Though I do miss vinyl and it's large cover art / nostalgia, but it's crazy money to minimize the damage caused by simply playing it and I'm not paying more money than a basic car for a laser turntable.
@@ericcrippen8634 I remember that. Roon has a dynamics figure which is kind of helpful, but I wish they had more information in the vein of that website we're thinking of.
On what forum I can find what release of individual album is a best version?
I have some 16 bit CD's that blow away 24 bit. It all depends on mixing and mastering. I'm not ever going to give up my CD collection for a streaming app.
so true.
For me, this 24 bit is just snake oil. I bet 99% of people cannot tell the difference between 16 and 24 bit mastering. Oh yes, everything goes through mastering expertise (Emily Lazar is a well known individual who is able to destroy, ruin and cheapen even the most beautiful recordings).
you do have SACD (dvd) and HFPA (bluray)
HFPA is encoded as 24-bit/96 kHz or 24-bit/192 kHz linear PCM ("high-resolution audio"), optionally losslessly compressed with Dolby TrueHD or DTS-HD Master Audio.
if you prefer physical.
so right about the mix/ master. Because I have listened to a Cd , excellent, I have listened to the same album in 24 / high res audio and it was awful.
I never understood streaming. A 512 sd card costs 40 dollars and can hold most people’s music collection. Ready to play anytime anywhere without a need for internet
Excellent summary of why 24 bit audio is so important in the studio but not so important for the consumer experiencing the final mastered mix. Well stated.
Have to agree with you. I have found that the mastering of many “hi-res” recordings to be much better so they should better. I have also heard some amazing 16bit 44khz recordings. Wish they would focus more on better mastering and less on bits and sample rate.
Absolutely
most people will never have the ability to even hear the "hiss" in the silent parts of cd recordings. a cd can make 20kHz, if you can hear 20kHz, wow, that sounds like "AIR".. but the hiss noise due to dynamic range will be more frequencies lower than 20kHz and so what we are really talking about is how much "hiss" you hear when you run a CD player through a SUPER high quality mixer (like a 100,000 dollars) through a pair of speaker point blank ( ???? maybe 10,000 watts... certainly at least 500 watts... which may not be enough to hear the hiss on a really well done cd )... because that hiss is what we are talking about eliminating and I'm telling you : most people don't eliminate all the noise.. not even big studios publishing hits even always do that. we're talking about the general public FAR exceeding the studio with some of this gear ( in some cases ).... for older music, forget it.. its kinda like watching the original wizard of oz on a 4K television.. sure, there are 4K cameras and even 8K and higher BUT ! ! ! ! with that sensor, how much of the video is in focus to that degree and also "stabilized" ??? many movie producers film in 4 K with intent to ultimately publish HD.. or they will film in 8K, then optimize down to 4K using software stabilization.. but now you have issues where if someone maybe has hairy legs, its gonna become part of the movie now.. but these older movies : I mean the cool things is when they scan film, they can pretty much scan a 4K scan but honestly THIS is the real issue : the LIMITATIONS TO MAKE GLASS LENSES !!!! there's only so much detail these lenses can focus. so you may be getting 8K but is it really better??? well yes with some lenses, yes it is..but with some lenses, its really not "better".. I don't know.. I'm torn on the advent of new tech. I would not have been torn if all we had was a telegraph and we were about to come out with a telephone, in that case I would be all about it.. but now, we have reached a point where I am in no rush what so ever to advance the technology.
FLAC files sounded way better than any cd ive ehard though
@@jhoughjr1 Really comes down to the mastering. I can't hear the difference between a cd quality FLAC file and a 24bit 192khz flac file when the cd quality file is created from the 24bit 192khz file properly. Have you done a blind test yourself? I think you would be surprised with the results.
All this bit bs 24 bit stuff is to take sheeple to the cleaners. Hell, make28 bit to sell to suckers, then sell 36 bit to suckers. It is all bs.
Thank you for creating these videos. You've done a lot in answering many questions I've tried to figure out and the internet is full of incorrect information that does not make any sense when compared to my personal experience with hifi. From your videos it has all started to make more and more sense and I can understand things I hear better and can pass the knowledge on. Please keep making more!
As an ex recording studio owner 16bit is even more than enough. When we record a single track we record below the 0 db threshold because when mixing multiple tracks total loudness is multiplied. What I'm saying is if even 16bit recording is done with low noise equipment then mixed properly there is no way the end user will be bugged with noise. If your hifi is not up to scratch 24bit signal higher dynamic range will even be more taxing. So guys and gals get your hifi equipment de-noised and don't fall for marketing greed. CHOOSE YOUR HI-FI CAREFULLY OTHER WISE EVEN 50 BIT WON'T CUT IT. IF THE RECORDING EQUIPMENT IS LOW NOISE EVEN 16 BIT SOUNDS WONDERFUL..
No, that's not quite right about the difficulty with 24 bit.
The only issue in hifi equipment that is "not up to scratch 24 bit signal" is poor design of the 24 bit DAC converters and associated circuitry.
24 bit DACs give a higher signal to noise ratio (than 16 bit). To get the maximum benefit from using 24 bit DACs the analogue design elements need to be of lower noise design, otherwise the effective signal to noise ratio ends up not being the 144 dB it should be, and you have lost the benefit of using a 24 bit DAC.
It's easier to design the electronics for a 16 bit DAC which has a signal to noise ratio of 98 dB.
The idea that a hifi needs to be good enough to support 24 bit dynamic range is a fallacy.
The signal going through the HiFi, through the amp is analogue, it's a voltage changing with time, it makes no difference to the amp whether that signal started out from an analogue source or if it originated as a 16 or 24 bit data stream.
The signal going through that amp is just audio signal plus noise. The difference between the 16 and 24 bit DACs is simply the amount of quantisation noise that added on to that audio signal.
24 bits is used in the mixing console to reduce the amount of distortion that occurs when manipulating the 16 bit data streams, to prevent numerical errors creeping into the audio data stream (see my other post on this).
The HiFi amp just sees an electrical signal which it then amplifies, as linearly as possible across the full voltage range output from the DAC. The difference is that to get the best from it, the 24 bit DAC should be designed with good low noise circuitry. Downstream from the DAC, nothing cares.
Don't get mixed up between dynamic range and distortion.
Dynamic range does not determine and nor is it a measure of how good it sounds.
THD is the measure of how accurate the output signal is relative to the original sound signal. That metric looks at the change in amplitude of the harmonics that make up the waveform as the signal passes through the system.
Your last sentence is a little concerning referencing noise along with "Get your hifi equipment de-noised".
But noise doesn't determine sound quality does it? And you can't denoise a system unless you replace components of the system. The noise is fundamentally a function of the design of the components and operating temperature of the semiconductor devices in the components.
These are not aspects the user controls, except by replacing the equipment.
There are several key parameters: THD, THD+N, SNR, dynamic range.
You need to be clear in your mind which affect the sound quality.
As far as I am aware, and I hasten to add I have been out of the industry for many years, there is no published recommendation that says in order to handle 24 bit digital audio recordings, that a certain level of THD, SNR, dynamic range are required for the HiFi playback equipment.
@@deang5622 Fantastic explanation. Finally someone who has a thorough understanding of analog, digital and the engineering considerations that must be considered on the *entire* signal path end to end.
You are of course 100% correct. Unfortunately, many people just don't understand the technology or its limitations at all, so, sadly, they will always latch on to some marketing hype that appeals to them and believe that. Vinyl is a case in point.
@@deang5622 Well explained! There is so much marketing fallacies and average Joe unfortunately doesn't have the knowledge to reject all false stuff. Every sound equipment has its own transfer function, the sum of all transfer functions accumulate along the signal path up to the listening room which has its own and finally to your ears! People should invest into their listening room first, that's the most non-linear place on the signal path!
@@guyboisvert66 Yes, agreed.
While working for Dolby Labs, we had an awesome listening rooom with some very expensive Meridian kit. We often did blind listening tests and we could not tell the difference between CDs and higher bitrate. A lot of this was done when designing the original Dolby Digital codecs (6 channels stored as digital info stored between the sprocket holes). It had to be highly compressed so lots of listening tests were conducted. But, it had to be damn good.
Yes but you're talking about bitrate here, not specifically encoding resolution.
But did you listen long term? Instant switchover tests whether blind or not assume we have perfect discernment at an instant. We don't... years of experience tells me that listening over extended time will better demonstrate differences than do instant A-B comparisons.
@@socksumi Yes, it was done over a long period. Care was also taken to sit in a dark room with eyes closed with a period of silence before each test. Amazing how much more you can hear when your eyes are closed for a while.
@@KangoV What I meant by long term is when say comparing components... listen to one component over several days in a completely familiar listening room and playback system such as at home. Then listen to the other component for a while.Then conclude if there are audible differences.
I remember listening to audio gear that sounded wonderful in the store... then when I get it home it might sound good at first... but after extended listening sessions, subtle but annoying colorations began to manifest that I didn't notice on first listen. As my brain latched on to these colorations and distortions, they became easier to detect over the long term.
@@socksumi Listening tests were all conducted in a fully sound isolated room. The insulation was very thick with those foam "spikes". All the rooms (listening rooms(2), projection room and 54 seat cinema) floated on huge springs to isolate them. It was pretty hard core. They knew what they were doing when they built it.
Far more important than number of bits is the quality of the DAC and it’s associated analog output stage. Most CD players and computers have low quality DAC sections. The DAC is where the digital bits are turned back into an analog output signal, which is what the sound being recorded usually started off as, prior to being converted to a digital signal for mixing and mastering. The magic always happens in the DAC if it is a good one…. Same for streaming.
That is completely true. the DAC makes is fair share as well in quality... i have an old Sony CD deck from 1994 that sounds amazing in both optical and rca with headphones
Yes absolutely, 100%
Actually you'd be surprised how good even cheap DAC's sound compared to high end hardware. Here is a $100 Sony Stereo receiver compared to a NAD C388. The NAD costs $1499. ruclips.net/video/zbqFOGAId20/видео.htmlsi=IqlrRsAExSrDsYAT
i see it the same way. i have an old rotel 16bit dac from the 90s and many cheap 24 computer dacs from 10 years ago can't compete because their power supply is shit. you need a power suplly for the digital and analog section that is separated. also for the analog section you need a very stable supply to have enough channel separation. better to have separated left and right power supply stages to have good channel separation.
a good dac chip does not cost much, but the power supply add much more to the overall cost. some dac chips can be used in mono mode and using two of them with two separated power supply stages makes the difference. all that adds cost but it is necessary to really get the most out of 16 bit. this must be on point before talking about 24 bit.
whether it adds to the listening experience to have a few bit more is difficult to say if you compare an extremely good 16bit dac with an average 24bit dac.
24 bit means a dynamic range of 144dB. but you need the power supply stage as well to serve that 144dB. including the channel separation. and what does the output buffer stage to the power rail that is connected to the analog power supply of the dac?
what is a listening test say if the 24bit dac is not done well.
i would rather have 96kHz to reduce modulation effects in the 20kHz region and increase the time resolution between left and right channel. double the frequency automatically gives you 1 bit more dynamic resolution as well.
In a professional recording studio you have multiple channels of music that are summed together when they are mixed. They need the 24-32 bit precision so that when each channel is mixed the noise floor remains low. Otherwise, the noise will accumulate in the final mix. Consumers are just playing the music so they don't need, and likely can't hear, the difference. Chances are the ambient noise in a listening room is much higher than any noise introduced by a 16bit playout format.
16 bit is enough for every detail to have it's own place, basically Mixing and Mastering are easy straight forward processes that many overcomplicated because they couldn't accept that it's that easy task for Mix and Mastering
i can not see a direct dependency between noice and resolution
I dont need no expert to tell me facts, my golden ears can hear better than bats. Even though im old.
@@michi9816 the bit depth has direct correlation to the noise floor. It allows for 98dB signal to noise level and is related to quantization (rounding errors between the analog input voltage and the digital conversion). Quantization errors sounds like tape hiss. Since most tracks have a dynamic range of under 20dB this is of no concern at all. Nobody in consumer audio needs more than 16bit! Also, the beloved vinyl record has a much higher noise floor and that’s apparently fine? 😅
It is not to do with keeping the noise floor low, it is to prevent distortion.
A 16 bit value combined with another 16 bit number (as happens when mixing multiple channels, or processing the 16 bit stream) will often result in a number requiring greater than 16 bits to represent it.
If the number is too big to hold in 16 bits then you have to do something to it to make it fit in 16 bits, such as truncation, and results in the final analogue waveform being massively distorted to the original that was recorded.
So it's not about noise floor, it's about preventing distortion that occurs because you have an insufficient number of bits in the mixing stage to hold the digital values which have been produced as a result of processing on 16 bit data. (numeric overflow type situation).
I completely agree that 16bit for the final product is perfect ... but having dabbled in recording since the analogue days I recently had a "life changing" experience of the advantage of a 24bit recording. It was simply a "master" where a whole section of a live performance was significantly quieter than the rest of the excellent recording. To match the levels in analogue days would inevitably have resulted in a huge increase in background noise but doing the same with the 24bit recording was like magic ... simply matched the level and still no audible increase in the noise floor! Brilliant! The advantage of bit rates above 16, as explained in the video, is the scope it gives for manipulating the source rather than being necessary for the final mixed version.
24 bits is usable for recording. But 24 bit over 16 bits will not give you anything extra in the final mix that you will listen to.
24 bits is usable for recording because you might want to make some element of the recording more prominent in the mix later. That often means compressing that element so that it will sound more up front in the mix. Doing that compression means that you'll also be boosting noise, and often a lot of it. So in that context, having 24 bit accuracy can be a better starting point before going into the mixing stage.
During mixing, you'll typically have almost unlimited headroom in a DAW because they typically work with 64 bit floating point accuracy. That's ridiculously high accuracy that will never become an issue in that context. It's just not an issue there.
For the final master, 16 bits will be more than fine.
Your favorite audiophile Pink Floyd album isn't going to sound any better in 24 bits because it wasn't recorded with that kind of accuracy to begin with. It was just noisy tape while recording on top of another noisy tape while mastering.
There are better reasons why such albums still sound so great, but the number of bits has nothing to do with it.
Exactly. In electrical engineering, we learn all the digital theory, we have the advanced / applied math to calculate, we have the DSP courses, etc. And don't be fooled by the infamous "staircase representation", there's no such thing: It's just a bad representation of the actual data which is "data points", not staircase. The number of final bits refers to dynamic and noise floor: With 16 Bits, we get 96 dB, more than enough. The 16 vs 24 doesn't bring you "precision" or anything like that! To understand how it works, Monty Montgomery's great video: ruclips.net/video/cIQ9IXSUzuM/видео.html
correct
I wonder if that's why home recording interfaces will do 48 or 96khz; you have more headroom for not getting your initial input range properly set and end up with more room for correction.
all digitized analog recordings are 24bit
so why degrade it by releasing 16bit files from the original 24bit one?
@@roverwaters3875 I think that's part of the points made in this video - 16 is 'good enough' once mastered correctly based on the range you can hear from floor to ceiling. It'll be better for recording as you have more room for adjustment but once it's mixed/mastered the differences are potentially negated once mixed down.
A few years ago I set up a test in a good listening room, where I gradually reduced the bit depth on a recording, and asked people to say when they heard a difference. Most people couldn’t hear a difference between 12 and 16 bits, and those that could new exactly what they were listening for (noise).
it's true. there's not that many bits in a lot of music. Of course a good dac is still a good dac. Just because you don't need a ferrari because the speed limit exists doesn't mean one should buy a shitty driving car
Well, it also could be that they didn’t really KNOW what to listen for, and if they did, that their ears were not trained to detect it.
How many people are still satisfied listening to their 16 bit, 64kbps MP3 rips from ten years ago? Sadly, too many.
This is a complex topic, to be sure. But I have simplistic thoughts about it. My feeling is the more bits the better because then it will be more like analog. And the more bits there are, the more it is likely to sound more analog. After all analog is continuous, digital is bit space bit space bit space. I like to think music is filling up those spaces with more and more bits.
I use bit-crushers a lot when making music, under a certain point the sound is totally artificial, and even lower it gets destroyed! But yeah, over 12 bits only a trained ear can tell the difference! (noise+high frequency artifacts, such as phasing).
@@sylviarienzo6955 No, the human ear/brain combination do not provide continuous sound: that is why we can only hear within a certain frequency range. The ear/brain combination only give the impression that there are no "gaps" in its "sampling" of changes in external air pressure. Don't confuse sine-waves as displayed on an oscilloscope and analogue audio.
Unless you have a VERY good system, and VERY good ears - and - you listen to music where it can possibly matter (classical), then 24 bit matters not vs 16bit.
You mean my 24bit AC/DC album will not sound any better - no way 😀😀😀😀😀
Actually if you drop AC/DC to 4 bits an audiophile would probably not pick it.
Heck AC/DC sounds better run into an amps 'brick wall', get the last 5 Watts even if its at 40% THD.
*And* are under 21...
There's no music that requires a dynamic range from a whisper to a jet engine. And if there were, no amplifier with the tolerances required for faithful reproduction and no chance of finding speakers with that capability either.
@Alexander Ratisbona Stockhausen's Helicopter Quartet maybe.
Good analysis. In summary, 24 bits allows for comfort in recording, without having to bother about recording level, and 16 bits are enough to reproduce the actual dynamic range once the recording has been processed and fine tuned for optimal digital storage taking full advantage of a 16 bit range.
Some graphic illustration would have been useful to ease the non engineer viewer’s understanding, though.
First time I've listened to your channel and I agree with everything you've said. And to that point, I really appreciate being able to buy so many cheap used CDs at the used book store. 16 bits really is enough and if a CD sounds bad it's because it was poorly produced. Adding bits won't fix that.
Agree 100% that's an issue I have been trying to drum into Hi Res fanatics who swear that it is audibly superior to CD even listening on equipment which isn't Hi Res compatible when I have listened more than closely and fail to hear any difference at all. CD is a wonderful medium when it's done properly but try and tell that to vinyl junkies who are listening to a dynamic range of 70db at the very best, which includes inevitable tracking distortion and noise, and you'll be told in no uncertain terms that you need a hearing test. I still buy CDs, I still get much musical enjoyment out of CDs and I would never go back to the horrors of setting up a turntable ever again and straining to hear the system rather than the music OR wasting my money on the con that is streaming.
👍Yes! My kind of "audiophile"
There is one very good reason why some vinyl DOES sound better than it's equivalent CD. You cannot take a digital master that has been recorded balls to the wall HOT, and even think about making that work in a vinyl cutting process. This will only create records that totally skip all over the place. Hence, they have to be MASTERED for the medium, which means in many cases, the vinyl will actually have some dynamic range, whereas the CD does not. IE, The Arctic Monkeys first release from 2006 is totally compressed and full volume on the CD, but the vinyl version is very warm and so dynamic. It was like listening to an entirely different recording, which in a sense, it is.
The people who did the tests clearly didn’t have high quality cables. They probably used simple twisted copper wires. Yes that is sarcasm. Some people have extraordinary hearing, in their own minds.
Totally agree, I just buy cd's now for digital playback that is.
Man. I know that’s right!
Man I really appreciate your vids…, context is decisive 🙏🙏
24 bit is useful, if not compulsory, at the recording, mixing and mastering stages. Because mixing and mastering engineers will process the signal. Same process with photography : you may recover a full viable picture from a hugely underexposed photography captured with a 14bit full frame sensor whereas you will never manage with an 8bit Jpeg.
Thank you for the photography analogy, I was thinking the same thing. However, in most situations Jpeg is just fine.
@@maze400 fully agree. The analogy goes even to your point : 8-bit jpeg can only be a final master. In photoshop / lightroom you will put the 14-bit RAW file.
@@maze400
Yes, a decent quality JPEG is analogous to a 16/44.1 file used for final PLAYBACK.
Shoot & Edit in RAW/Record & Mix in 24 bit, then output in JPEG/16-44.1 for consumption.
Love the photography analogy! Raw files to jpegs!
Jpegs are essentially the same as mp3s. Lossy compression. Both pretty much suck. If you must compress a photo, imo PNG is a better format, but still lossy.
I totally agree with you - 16 bits - the best 16 bits out of the 24 or 32 floating bits we use in studios - is good enough for the listerners market :-)
It can be said objectively that CD encoding is plenty of dynamic range. After all, the world’s greatest music exists, and has been fallen completely into love with, with CD playback. My producing career spans the years 1975-today so I’ve personally used all the formats. The fundamental reason we moved on to 24 bits was to better accommodate the common studio practice of copying, re-bussing, and other techniques that could easily reveal artefacts in 16 bits. In that, it has worked splendidly. Best to think of it in photographic terms. In 24 bit, we can zoom in losslessly quite a bit.
It's not unlike shooting 8k video for 4k delivery.
Here is my take on this… but, please note that I am NOT n audio engineer. Were I a recording professional, I would prefer to mix down from a 24 bit original at 96 MHz, even though my final product is 16 bit 44.1 MHz. And I’m willing to bet that the final product would sound better than if it were to be sourced from a 16 bit 44.1 MHz original. Any takers?
As a graphic designer I've always made a comparison to resolution. An effective 1200 ppi image doesn't look 4 times better than an effective 300 ppi at the same physical size. But if you want to start doing effects or repurpose for a larger physical output size, those extra pixels are great to have in your pocket.
@@TucsonBillD You have it wrong. It has to do with accumulating errors (noise) while mixing. Mixing in digital is calculating numbers with a software on a computer (a digital console is a computer!).
Entertaining, pragmatic and accurate content. Good job.
A voice of reason is always good to hear. Thanks for the video!
I’d say you did a pretty good job as an introduction to the topic. As someone who dabbled in mastering for CD delivery back in the day, I would recommend Bob Katz “Mastering Audio, the Art and the Science”. There is no shortcut to understanding dynamics, and that’s the easiest to understand, yet comprehensive reference on the topic.
You probably listened to Ana Caram's recordings done by Bob for Chesky. Anyway, I called my friend over to listen which crossover version he likes better for particular 2way monitor. He listens and he says there's some noise, some interference coming and going away. I looked over the cables, connections, etc. Turns out the noise is in the recording, I guess something sneaked into recording chain, but knowing how hard is to record live in the middle of a big city they didn't repeat the session. Pretty decent set of loudspeakers for catching things you don't want have in your recording, just too analytical for my taste ;)
If anyone thinks they can hear a difference between 16bit and higher bitdepth, there is actually a way to listen to the isolated difference by itself. Load the say 24bit 44.1kHz track into audacity and export it as dithered 16bit same sample rate lossless (wav, flac, whatever else true lossless). Load the 16bit track parallel to the original into audacity and invert it. Now everything that is identical in both the original and converted track is cancelling out. Exporting the audio of both tracks combined into a new 24bit track leaves you exactly the difference of these tracks, simple substraction so to say.
Play this as loud as you want. In the very unlikely case that your audio chain offers enough gain to make this track audible, you will agree that the amount of volume would immediately alert everyone in a one mile radius and blow your speakers in seconds if there was either the original or the converted track playing.
You can analyze the spectrum of the residual track and find that the noise actually sits as low as -130dB across most of the frequency spectrum. Total signal to noise ratio confuses people a lot.
Sony's SpectraLayers pretty much confirms everything you've said. I've a fair amount of techno music studio hardware that operates at 24bit, so don't quite need to get wondering about how a track on Tidal is a 'bit' messed up from the master.
Very good video .well thought out and executed. I love it when you have comments stating sound & streaming. But as you already said there are a limited amount of steamer companies available. Back to 16 & 24 bit I am quite happy using a good quality cd player and getting the best from that .I must say again a Very good video Gerald 😊
I loved your explanation and I will check out more of your videos. I just got the 32 bit looper by boss and hope to dither down to 16 bit for performance and recording. I have much to learn and I'm grateful
Thank you for this video.
I've been mastering my own music at 16-bit for many years and I don't see a problem with it. I wasn't aware of the volume increase capability of 24-bit but honestly, anyone who listens to music at a ridiculously loud volume is going to destroy their high-frequency hearing, lol. At the end of the day it's all about enjoying the music anyway :)
It’s almost like the CD had some deliberate research behind it. Shocking they didn’t just wing it! 😂
As a pupil of the (then) professor that before worked as one of the four designers of the Red Book specs I can assure you that Philips went through extensive theoretical and practical excercises before even moving on to listening tests (which I reproduced for the curriculum he set up). Double blind listening tests confirm what you're saying :-). A lot of thought had gone into the sampling rate since it determined how steep the cut-off filter had to be and the well known (phase) issues that go along with it, versus the costs involve to improve it rendering the product less suitable for mass market. And although I agree with Rupert Neve that having a frequency range available far beyond the audible (other discussion), the 16bit/44.1kHz 20kHz audio range spec has not been detected as less than any other higher speced system by more than a few in the world
@@ErikDobbelsteijn thanks for the info! 😃
THANK YOU! I can't believe after all these years this is still a subject of debate. I have a Deezer subscription which provides 16 bit, 1411 kbps FLAC streaming, and I couldn't be happier.
absolutely great!!! > you get to the point..........
warm greetings from germany and from a german musician
What matters the most is the quality of the recording and instrument placement/separation in said recording. Yes you can tell higher bit recordings if they are done right, they sound more involving with a wider soundstage. Then again, the gear you are using also helps to an extend.
Excellent video. Thank you
To be honest, I started writing a comment half way through the video as I was getting annoyed (the video was still playing in the background).
But then.. I backspaced and started all over again, as you mentioned the most important part when it comes to 24 bit: it is very useful in mixing.
So - I do agree 100% with your explanation. Nice work!
Long live CDs.
Well the will definitely live longer that your average mobile phone full of mp3'd 24 bit flac audio.
@@paulstubbs7678 I own CD’s that I bought almost 40 years ago, and only one no longer plays (and it’s actually only 23 years old). For me it’s Flac (i ripped the all to a server) and MP3 in the car. I’d o Flac in the car if my system would play it, but I doubt I could tell the difference there. Also use MP3 when exercising, but if I’m paying attention in a quiet setting, I can hear the diff in the cymbals.
I RIP CD's
24 bit sounds better however.
Depending on the recording.
Thanks Qobuz!
Their trying to bring them back. They developed these new Hyper CD-ROMs that are 1 petabyte per disc. The equivalent of 250 million songs
I used to have time to sit down and just listen to music, up to the time my first child was born, then life interfered. I now have the best stereo equipment that I've ever owned, but hardly ever just sit and listen to music. I imagine it's the same for the vast majority of people. So I seriously doubt 16bit vs 24bit matters.
And by the time you get back to that great gear, your hearing has likely degraded.
@@cbug4196 Sadly that's me😢
I like the sound of 16 bits 'done right' - much better than mp3! I use Tidal and it sounds pretty good - I still like the sound of a nice CD though. When I'm mastering I use 32 bit floating point with no dither, for my final mixdown, and 24 bit and 16 bit with dither for the masters I'll deliver - headroom is comforting! Great video again
Very much appreciated information. Most of it fly's over my head but I understand enough to as i sad appreciated the information so thank you for this. (First time watching btw).
I do hear noise floor in many of my CDs...... of 70's artists. You can hear the noise floor of the studio master tape via the quality of CD. Remember those labels that said "Because of its high resolution, the Compact Disc can reveal limitations of the source tape"?
I remember those labels like it were yesterday- trying to explain to the average listener that an analog master converted to cd would expose limitations of analog recordings. Cheers!
Exactly! The big master tape recorder used offered an equivalent of 11 bits noise floor. Not only that, mixing add noise all the way to final product. The analog cult guys on the internet doesn't have a clue and when they're talking about "digital signal staircase", the show their complete misunderstanding of the digital theory let alone the engineering behind it. ruclips.net/video/cIQ9IXSUzuM/видео.html
Or your bad equipment! That’s your speakers, amplifier and wires / interconnections all makes hissing sound.
@@Scott-px6mi That's a fair point, although I've listened to other artists well into the digital era at a similar volume, without hearing noise. But I wouldn't be surprised if my (relatively nice) headphone setup hit its noise floor before the CD format does. I certainly wouldn't want to try it while music was playing lest I go deaf from the volume!
@@Scott-px6mi Unhook your speakers from your amp, put an ear up close to a speaker and hear how much hiss the speaker cable generates.
I remember reading about a study which sought to answer whether a difference could be perceived by humans, between 16 bit CD's and SACD's.
From memory they had around 500 audio engineers listen to some A/B tests and note which they believed to be which. The result? None of the participants could do better than a 50% success rate.
In other words, their perception was as good as a coin toss. Or in other other words, no perceived difference at all. Meaning that 16 bit audio seems to adequately meet our needs.
I feel you did and excellent job explaining the practical application of the different bit rates on this video.👍
You’re good mate. Keep the vids coming 👍😎.
Remember the JVC XRCD's? They proved it's all about the way they were recorded. Their CD's smoked a lot of SACD's of the same album in sound quality. Sad when they gave it up because most people didn't care.
@Alexander Ratisbona I was at the Las Vegas CES show and a representative from JVC was explaining the whole process and showed comparisons with other CD companies. Using the same master tape showing how much better their recordings were. Quite stunning how much detail was left out by the other companies. Elusive disc sells them.
i searched now about a format released in 1995. i didnt know about
I've only mixed one album in my life, so maybe there's something I'm missing. But I used 96/24. And when I downmixed to 44/16 I couldn't hear any difference whatsoever with my professional grade equipment.
well then you just havent spent enough on equipment then /s
@@herbyverstink christ how much would I have to spend
@@dannyb3663 ALL OF IT?! 😂
The difference is much less than doing eq 1dB of eq, if done with decent resampler and dither.
To be honest the MP3 format will be more dynamic than any other format out there if it is to be streamed. But if you are doing 96k at 24-bit recording or editing you should be able to notice the difference in particularly if using a loss-less format vs. Lossy. Not too much difference in perceptive sound change will be noted if the streaming level is 192 kbs at 24-bit or better.
It is one thing to understand complex things. It is quite another to be good at explaining them. Your explanation was, to my mind, perfect. I learned quite a lot. Thank you.
Just enough is not enough! Loved it.
Cd collecting will remain relevant particularly for collectors.
There is a huge amount of music that never gets enough attention nor streamed, and is collected by record and/or cds.
Also, the noticeable difference among most people of 16 bit to 24 bit is not generally differentiated to the listener. Just look at the record vs cd debate! People enjoy their medium and truth be told audiophiles with their home gear will not pick up on the differences very much either unless their life long journey system can pick it up. And then again the subjectivity of ones own ears and personal taste remain a bigger factor.
I like vinyl and the fact that it still hasn't died says something. I fully accept that it is not as accurate as digital, it colours the sound, but in a pleasing way making music enjoyable. But isn't that the point?
@@millzee60 absolutely. Pleasing to the ear is more than A okay by me any day of the week.
CD's will remain relevant for people who want to actually own their music collection and not license it. What Apple giveth, it can taketh away.
@@jamescarter8311 Absolutely. At least with cds you can give it away to a friend or family member with similar taste and interests.
@@jamescarter8311 I can't remember what it was called but the BBC had an online digital store which went under. But before it did, the BBC allowed users to download their content. I imagine that any online store would have to do the same were it to be closed down. Apple users have bought their content, not rented it.
Well done, I agree with virtually all of this. What I have said for years is that correctly dithered 16 bit is more than enough AS A DELIVERY FORMAT. Of course we track at 24 bit and we mix in 32/64 bit (whether we choose to or not - but many don't people seem to get that part). But as a delivery medium that will no longer be changed, I am happy to have 96dB of dynamic range be the headroom you mention for my rock/pop/R&B songs that will barely have 20 dB or dynamic range anyway. I see you already have a video tackling vinyl which I am yet to watch. Hopefully it supports the premise that it's a heavily flawed format that has some lovely euphonic tones, but is objectively technically horrible when compared to CD.
Thank you for your precise explanation, that's reveal a lot of thoughts about what the people like me can listen on a CD or a Vinyl or a Reel to Reel and whatever. Indeed in these times of surrounding mass missinformation a lot of people talk about numbers and is suggested to listen better when a system can deliver more "numbers" than another, it's a world of confusion and noise for nothing. My experience is that listening some very good vinyl i have the same pleasure of listening the same good CD, with some little difference due to the type of reading system but always pleasure and satisfaction. Thanks two times.
I record in the field, using 24 bits lets me keep level low enough for the unexpected peaks but still have decent resolution and noise when the final mix is put together.
I'm skeptical, always have been, always will be! To me there will always be more of a difference between two different masters of something, than the difference between the same master stored on 16/44 or 24/96.
The only difference I see, is an audio engineer tasked with producing 24bit audio will try way harder than an engineer tasked with making an 'everyday' 16 bit CD, a 'loudness wars' release will (hopefully) be the last thing on the minds of a 24 bit release engineer.
@@paulstubbs7678 Unfortunately many 24bit releases of pop/rock music are just as smashed towards Full Scale as the 16bit versions! It's madness.
100% agree! Many times the difference you hear with streaming is that you are hearing a different mastering job. You're not comparing two streams of the same source.
@@paulstubbs7678 Do you think that the format of CD itself (16/44) placed constraints on what could be put on it, in a dynamic range context or bandwidth context? That the format itself might have partially contributed to the modern loudness race?
@Zickcermacity
No. If the mixing and mastering engineers were doing their job and optimizing the gain structure throughout the production process, there is plenty of headroom for the full-scale reproduction of ANY type of music or real world sound in the 16/44.1 format. That's why it was so carefully chosen when the digital audio format was first developed for widespread commercial use.
The 16/44.1 format itself did not contribute to "brick walled" music releases on CD or otherwise.
The path towards "The Loudness Wars" was either due to lazy/unprofessional engineering (not maximizing gain structure & Dynamic Range during production), but WAY more likely, it was 99% due to record label execs demanding that the engineers that they hired maximize the OVERALL PERCEIVED LOUDNESS in order to make their artist's track STAND OUT and GET NOTICED compared to the Previous Song or CD that was played on the radio or a CD on your HiFi system.
See how you noticed my ALL CAPS "YELLING" text more than the previous or following text that I typed in this comment??? That is the equivalent in written communication to the purpose of a Brick-Walled audio recording. IT'S TO GET YOUR ATTENTION and Sell More CDs and Music! ;-)
I have a few cd's that have been mastered in 20bit but reduced to 16 bit to work on the cd format...these sound truly remarkable.Its all in the mastering and not the bit rate.
It's always great hear about sound from the goat himself. Thank you Mr. McCartney!
It's enough for me too 🤗 great explanation.
Truth is, TOO much is never enough when you have audiophiles with more money to throw around than common sense. They hear these minute differences in recordings because they spent so much money, they'd damn well BETTER hear the difference. The placebo effect alone can convince such people that indeed, the $8700 they just spent on a component in their equipment chain really DID bring back amazing improvement in........well, it just did. Just don't force them into a blind listening test.
Yup. By the time they've done all their little tweeks it'll sound 200% better than the original master. The 350 dollar audiophile wall outlet is what really makes your system shine. Then listen to mind numbing audiophile music to really hear how good your system sounds. It's gotten stupid, if it looks better it must sound better. And all the phony techno babble to sell placebo. I can't even get anyone to give me a techno nonsense explination of how a wall plug improves sound.
@@bradt.3555 Oh, and let's not forget about the $10k speakers cables too. Those things REALLY do make a difference haha. NOT! At that price, they need to be made of gold strands, not copper. So many of these dealers are laughing all the way to the bank.
@@jimhines5145 So true. What I don't get is the denial of placebo, which is provable science. I'm all for getting the best possible sound but there's a point where there's no more info to be got from a particular recording. With today's technology you don't have to spend hundreds of thousands. Seem's much of the audiophile world is eye candy.
@@bradt.3555 Thanks for the reply. Rather than adding an "audiophile" power outlet, a decent UPS will get the job done, probably better and much cheaper too. While I do consider myself an audiophile, I still have lots of common sense. As a broadcast television engineer for 37 years, I have tons of knowledge and have had my hands on and inside of some super expensive and super sweet a/v equipment over the years. Many times the money one pays for a piece of equipment is just a pure waste of money. Even in the broadcasting world. When Blackmagic hit the scene with great products that were 1/10th the cost of the big boys prices, things changed. It is also the same in the consumer world, but "pure audiophiles" still think more money equals greater sound. Not at all true any more. Just know what chipsets the equipment you are buying has in it and you too can be a cheap audio audiophile lol.
@@jimhines5145 I'm a retired electrician. Done power for many large, small commercial, residential, industrial etc., etc., projects and been an audio hobbiest since H.S. Took electronics in H.S. so I too look at this stuff with some background. That's why the audiophile receptacle just left me speechless. Unless it's defective, it just cannot change the sound..........at all. Oh, I have a little 15 dollar Radio Shack DAC I got years ago just cause a tv I got only had optical out, no analog RCA's. Found out the other day it actually uses very good chips, so whata ya no. From utube to roku to TV to R.S. DAC to system still sounds pretty good. Thanks for the chat.
I have a pretty good system and find the quality of the recording has more to do with how good it sounds than if it's 24 bits vs 16 bits (CD vs DVD), my PS Audio Transport can play either in it's native format. Also a disk sounds a little better than the same file downloaded from a sampling service, not sure why but it does. The red book recording uses a brick wall filter that kills anything over 22.05khz but filters being what they are tend to get a bit wonky as you get close to the cutoff so 22.05 is cutting things a bit fine. Better to use a 48 or 96khz scheme (gets you to 24 or 48khz filter) to get that filter well out of the way of top audio frequencies. 196 should be better but I can't tell the difference - maybe it's my 75 year old ears?
A top notch 24 bit copy of a track really does sound better than a 16 bit version of that same track 9as long as you up the sampling rates) but on many systems you won't know it because the system just isn't capable of resolving the difference between the two.
Fascinating.
Great information as always and real world understanding, no nonsense here. Good.
My biggest issue with streaming is...
When I have grown up listening to a song unfortunately the members of the band that recorded the song (who are now older and wiser)have decided in their mature wisdom to remaster said song. The streaming services only stream the remastered version which to me sounds worse. So...
Find the cd if you want to know what it's supposed to sound like.
It is unfortunate that remastering often makes things worse. Perhaps at some point there will be a trend for de-mastering. A lot of music lovers would like that.
Usually it is done because they don't own the rights for the original recordings.
20 bit is enough for audio because analog amps can not control the voltage smaller then that. But since computers work in multiple times 8 bit we ended up with 24 bit. What’s also interesting is the 32 floating bit converters. This one’s can’t clip (they say) which makes them perfectly for field recording.
32bit sound is for first time or new project begging but as far as mix and Master goes 16 bit is enough but let's say you can hear and recognize 14 and higher Khz then think, do you want to listen and be able to hear more details then avarage listeners or you want to deliver sound that they like .
Really impressed with your explanation, A great channel, hi from Aus.
Absolutely brilliant! Thank you 🎉
Everyone knows that real audiophiles don't listen to music. They look at frequency response curves, modulation transfer functions, and S/N ratios.
16bit is perfectly good and I totally agree with you that professional usage needs some more headroom. The same goes with sampling frequency. The dynamic range of the CD format is even more than enough. The first time I heard CD back in 90's I was impressed by - the silence! Enormous amount of the silence and "air" in recording which made me fully enjoy the music and sound. AAD and ADD were and still are perfect for listening. As soon as DDD came out all this has crashed because the profit ruled of music production and started "loudness wars". The profit today is trying to sell more and more bits and more and more sampling frequencies although it doesn't make sense. The saturation of all this things evolved "vinyl revivals" and "cassette revivals". And what they have? Dynamic range much lower than digital technology, but even more, this type of media require much more care in mixing and production. That's why mine "best way" is to have CD format, but audio production more or less like you'd make for vinyl or tape. In other words, bring back AAD and ADD where you can even use plugins as A because nowadays they are incredible and at the end limit conditions like you're recording on vinyl...
You explained the most problematic problem of a large part of the music industry these days... They're ruining music and fortunately, there are still (but too few) good companies with competent sound engineer that produce pristine stuff. With a good DAC and a good soundsystem (i have an Ayre QB-9 hooked to Rega Elex-R and Ryan R-610), the sound i get leaves the vinyl in the dust! Sold my Linn-Sondek LP12 (which is an excellent turntable) and won't ever go back to vinyl and all its limitation / noise / crackel, lack of dynamic, etc!
Nice to hear a man with technical knowledge and intelligence!
The fact that Red Book CD format is still the Gold Standard for end-user audio in 2023 is a testament to those Sony and Philips engineers of the late 70s / early 80s who settled on the format at the time.
Does a listener need 80dB of dynamic range? look at the dynamic range of most recorded music, and situate that in the context of the ambient background noise level where it's being listened to, and how loud it's socially acceptable to have the peaks so that that quietest parts are still above the ambient noise level.
Put another way... CD is a building with 96 floors, but probably only the first 20 or so are in use.
I think you mean the top 20 floors.
The reverb/delay effects in the background can be as low as -60dB and they should be audible on a good audio system easily + decay of the guitar strings and drums. The DR measurement takes into account only average level of summed elements vs peak level of summed elements which is totally inaccurate for representing how much dynamic range is being used in the reality
I'm having a lovely time right now listening to Screamadelica on CD and I am blissed out!!! I may also be blissed out if I listened to it on Tidal but right now I'm not. I'm in heaven!!!
Any format like tape/vinyl, cd 16/20/24 bit or 24 bit steaming will do fine if you have great recording and mastering, excellent audio equipment, and amazing hearing.
As the hapless commentor who pointed out the 256x increase in resolution in 24bit, I totally agree that subjectively this is likely completely undetectable by any actual human ears. I have thousands of cds and am delighted with the format. Most pop cds don't even use most of the dynamic range available with 16 bit effectively because of brickwall mastering techniques.
The best channel on RUclips. Takes the cake.
I love your videos, they are so informative. Can you say something about the CD loudness wars that ruined CD’s?
I just want high quality recordings and to enjoy the music. I’ve heard some 24-bit Hi-Rez that sounded horrible and have many CDs that sound excellent so I agree it’s all about the recording process and the sound engineer
Please note you do not get completely unchanged or uncompressed 24 Bit through Streaming. But you can get that if you instead download the music from some websites that offer WAV or AIFF. But to make it extra complicated I can also mention DSD which is not PCM at all, instead it is a one Bit with an extremely high sampling frequency. Then there is DXD up to 32 bit and 352.8KHz
On a tour of the studio of the BBC natural history unit in Bristol. We were shown the multi channel audio sound system. On the mixing desk were two very cheap speakers. I guess this is a good example of head room. Hearing it as it could be as opposed to how most would hear the audio through a domestic TV. This was early 90s.
Thank you. A really great highlight reel of a few nuanced topics in play: tradeoffs and headroom in engineering, variability of playback/listening/recording environment & equipment, logarithms, sensitivity, semantics, the power of marketing and carefully selected statistics that encourage keeping up with the Joneses, and biases encouraged by past investment, for just a few.
Engineers build bridges and airplanes with redundancy and slack to keep people safe and allow flexibility in repairs.
Blowing a take due to constraints in gain staging might not be an existential catastrophe, but there might only be one chance to capture an inspired performance true to the original, with enough padding that it's unlikely to have overs, high noise floor, or get jacked up in post.
I can’t tell the difference between 16 bit and 24 bit recordings. My main medium for listening is CDs. I also have Apple Music however I prefer CDs because my playback system is better with CDs than it is with Apple Music.
There are lots of additional advantages with CDs not least that with the CD what you have won’t change, it isn’t going to be replaced with a remastered version or something like that.
It won't change unless you scratch it lol! The first thing I do when I buy CDs is rip them to FLAC files!
Just imagine the people bragging about their lossless streaming service and output the music to Bluetooth.
so just dont scrach them. i have many cds from the 90s that are almost like new because i had very carefull when removing and putting on the external case. i used many times and they play fine allways. in the 90s i had less than 10 years of age so if i can avoid scratch a CD with 7 years of age, you can too ahah
It’s sample rate that makes the difference in sound quality. When I used to work in studios we do blind tests when everyone was worn out from take after take. I wanted to add some fun in so we did this pretty often. Pretty much everyone could hear the difference between 44.1k and 96k. No one could hear a difference in 16-24 bit.
Sample rate only makes a difference when you're young enough to hear the tiny drop in the level of high frequencies towards the top of the range where the anti-aliasing filter is applied on a 44.1k master. Aside from that tiny roll-off shaving a couple of dBs, there is no difference between the data capture of 0-22050Hz in the 44.1k file or the 96k file. A 96k file does not pack more slices into the range of human hearing; it only extends slices beyond the range of human hearing into which inaudible frequencies can be stored.
What can happen is that some DACs are mysteriously better at playing back one sample rate compared to another, but when that's the case that difference is not going to persist outside of your studio.
@@joechapman8208 agreed on most of that. We were all early 20’s. :)
The biggest differences to me were how cymbals sustained. The initial hit sounds the same, but how it sounds afterwards is the key. 44.1 you start to hear it diffuse a bit earlier, where 96k it stays super clean.
I do think that 99% of the general public wouldn’t hear the difference. You have to know what to listen for.
@@apinkfloydsound Ok, but 96k is unnecessary, then. 48k places the filter decently far above human hearing. It would be nice if we had a standard a bit beyond 48k for total peace of mind, but 88.2k+ is only of use to sound designers who want to see what happens if they record ants walking on tin foil and then bring it down in pitch until it's audible. For anyone else it's just a waste of storage and performance, and a risk for intermodulation distortion.
@MF Nickster I think you messaged the wrong person, but to play their position for them, defenders of high sample rates will claim those frequencies are felt rather than heard. I know, I know.
@@joechapman8208 Oh definitely, if played loud enough the dog will certainly bite you to stop the torment!😅
There are some marginal differences pending on how it is applied. I would think of it like a TIF photo verses a JPEG photo. They both display the same image, but in context of editing and removing noise and detail the TIF is superior. It is the same with the depth of processing in a 24-BIT encoded audio over mere 16-bit.
I love these videos. They explain and debunk. I have CDs that sound absolutely marvelous, without any unwanted noise and also SACDs/DVD-Audio that sound horrible, so it is not a matter of bits...if we have enough. Someone said that "perfect is the enemy of good"
I pretty much agree with everything you said here. In my very humble opinion the difference between 44.1/48khz sampling rate and 88.2/96khz is much more significant (and audible) than the difference between 16 and 24 bit. 24 bit is nice to have but I suspect that very few people can reliably detect the difference from 16 bit.
Agreed. Why is the author and everyone apart from you not talking about sampling rate?!!
@@mfnickster9754 44.1 can only cover 22.05 if the sample is perfectly in phase with the signal. If it is 90° out of phase it won't record it at all. This, admittedly, is a worst case scenario but the problem extends well down into the range of human hearing. And we haven't even started to talk about filters!
@MF Nickster Digital sampling frequency is the same as a analog sound frequency range 🤦♂️. Ffs, is it asking too much for people to have a modicum of knowledge on topic before they post an ignorant comment? 🤣🤣
@@barrymiller3385 This clown thinks digital sampling rate is same as analog frequency range🤦♂️. Just ignore the ignorance for your own sanity 🤣🤣
20 kHz is usually for quoted for Hi-Fi but the vast majority of people cannot hear it. 15 kHz is perfectly adequate for music reproduction, this is for instance the bandwidth of FM radio. In this context 44.1 kHz is sufficient to capture all of the useful signal.
Grabbing popcorn for the comment section
Great Explanation Sir
Absolutely agree with you mate
Overall, I think you make a good case for the topic. If I disagree with any of it, it won't be by much. However, I don't think you are taking all of the factors that effect sound quality into consideration. And just to be clear, I believe you're being 100% truthful. There's just more to the story. I'll just list a few examples that you may find useful'
The first thing that needs to be considered is measurements and specs. They can only take you so far. For me personally, when I listen to a stereo system, the most important thing I listen for is timbre. For example, does a cymbal sound real or does it sound like a piece of metal dropped on the floor? Specs can't tell you that. I have done some research into the matter, and I found a few institutions that were trying to come with a way to measure timbre. They're not even close. The reasons cited were timber is a very complicated quality, and that makes it very difficult to measure. Its not something simple like frequency response.
This next point is really a continuation of the measurements problem I mentioned above. This has to do with the quality of your CD player. The difference in sound quality between different CD players can be huge. When SACD's first came out, I went out and bought a decent low to midrange player to try the new format. (For anyone who cares, I believe it was a Sony 9000ES, but it was a long time ago, so I don't want to say I'm 100% sure.). When I compared the same recordings on CD and SACD, the SACD sounded better. At this point someone always asks if I was comparing a remaster to an original. No. It was the same recording. However, things became much more interesting when I took a CD and played it on my expensive, high end CD player, and compared that to my new SACD player. No comparison. The higher specs on the new format could not overcome the lesser format played back on superior gear. I believe this is why there's always a heated debate on the topic. Not everyone is listening to the same system with the same music.
One last thing I wanted to address concerning CD players and DAC's, is how most people evaluate them. Most judge these products based mainly on the digital porting of the player, such as the DAC chip. The chip plays an important role, buy by no means tells the complete story. How the chip is implemented is far more important. Most of us usually ignore the analog stage of a CD player or dac. The analog section can easily have a much bigger impact on sound quality than the dac chip. If you gave 5 designers all the same transports and DAC chips and asked them to build a CD player, all 5 would sound different. How much will vary, but there will be differences.
I do have quite a bit of experience with these types of products, and sometimes I'm asked to give advice what to buy. My answer is don't listen to anyone. Use the advice you get as a guide, but you make the decision. If you can't try a product and evaluate it for yourself, don't buy it. If you don't hear a difference worth paying for, spending money on it is foolish. Just because I'm getting good results does not mean you will. Try before you buy. One thing I see happen all the time is when someone buys a product and they don't get good results. It gets labeled snake oil, and a negative opinion is formed. Down the road, you have different equipment and more listening experience, and the product you labeled as snake oil may enhance your system. That's why its important to make the best choice you possibly can.
I believe an important aspect affecting the 'feel' of the final sound is timing. Too much variation in the exact timing of dac chip conversions to create the analogue signal (jitter is the term) will affect rhythmic perception and make a song a less pleasant listen. A good dac unit apparently re-clocks the digital data to present it in uniform time to the conversion chip(s). In short, expect to spend as much money for a decent streaming dac as you might for a best-buy CD player.
@@philproffitt8363 Overall, I think you have it right.
Historically, jitter has been the biggest problem facing digital playback's sound quality. Before they figured out how to control it, your best bet was to get a single box CD player, not separates.
Since then, the industry has made some huge improvements. The designers don't always get the credit they deserve, but if you compare old DAC's to new ones, you'll hear major improvements. Anyway, to deal with timing errors like jitter, the master clock is the major component. You need a DAC or re-clocker that is able to take over and function as the master clock. Aside from that, the transport makes a big difference, as well.
I used to save my audio in 16 bits , 48kHz, then I switched to 24 bits 96kHz and I must say, the sound is much richer. Of course it wont matter much, if you listen on your phone speakers, but in my home studio I can clearly hear the difference which is probably not 256 times higher, but at least twice as high, judging with my ears alone. Also nice video, I like the fact, that you don't cut every 2 seconds like modern RUclipsrs do.
The 96kHz will be giving you a smoother, non digital sound (check out 192kHz and the "digital" sound doesn't exist)... If you are listening to pop and louder music, the 24 bits won't make much difference...
Prove it to yourself with a blind test, any other test is subject to placebo and invalid.
@@SethTaylor1 I can definitely hear the difference between 44k, 48k and 96k as far as playback engine... There is no difference between 16bit and 24bit for "loud" music...
what difference do you notice? i notice the difference with eyes closed and with instruments. tge voice i dont notice difference from 24bits to 16 bits
@@davidsucesso2419 I notice a more clear sound, when I listen in my studio, the difference is not noticable on regular speakers though
Great video.
It all comes to samplingfreq, and the DAC at the end. Just to keep the samplingnoise down, and liniarty in the DAC.
Lot of the older CD playes suffers with quantisasion noise, but when the 1bit/bitstream was introduced this improved a lot.
But still, a good mastered CD in 16 bit will stil be quit enough for most persons.
Interesting video and I agree with it. Sounds better in theory than in reality :)
But on the other hand, listening to a CD, it turns out that the CD player has 24, 32 and even 2x 32 bit converters to improve the sound. I noticed that it is better visible (audible) on lossy files such as mp3, where such a player can fill in the gaps. So the better the audio equipment, the smaller the differences in listening.
I have heard so many albums in different formats and heard flac's with around 800kbps bitrate sound better than 24bit-192khz files. Mix, capturing, so many factors that contribute to it sounding good or not
Adding to the comments about mastering, the streaming services have their proprietary loudness and dynamic range standards, in order to offer a consistent level. They modify the level on top of the mastering. On top of that the modern records and also the modern remasters offered on the streaming services are affected by the so called “loudness war” of music production, that reduces the dynamic range to absurd levels in sake of absurd loudness. That remastering is done using tools such as compression , limiting , saturation, aural exciters etc, mostly on digital domain. While that is supposedly done to a industry standard now, when I compare some of my old CDs to the remastered “master” version available on Tidal, the difference is very obvious for me, the streamed remaster is louder than my old CD but that reduces the quality . I consider those 24 bits are wasted in many cases. But I can only speak about pop music.
Wait what? Tidal modifies the level of the songs we stream?
In Spotify, there's an option you can turn off called "Normalize volume". Is there something similar in Tidal?
Thanks :)
@@1337sim1 Tidal has that option as well...I turn normalizing off.
Loudness war isn't going on anymore at least not as bad as it was. Most of the time if you have a good DAC you can overcome that.
@@LetsRideIllinois 😂
@@rabarebra laugh all you want. That just proves that you'll never be able to explain to me why albums that I've had from when I was back in high school sound more expansive and dynamic today than they did back then
Having been a huge supporter of physical media my whole life, I think I'm finally going to start looking at digital options for the home. I use Spotify and Tidal on the go, but I think I really need to start doing more research on a good setup for hi res digital audio at home.
100% agree with the thinking. Another topic is sampling rate but that's a discussion for another video. 🙂
More than 16-bit is really only useful in multi-track recording, as the noise level of many tracks together could (at least in theory) make a meaningful difference between 16 and 24 bit. But as stated in the video, when we're talking about a mixed track (even if it's surround sound), 16-bit offers more than enough dynamic range for even the most critical recordings. As such, buying 'High-res music' is a bit of a scam. People that claim to hear the difference with 16-bit are most likely hearing the difference between two masters of the same material. In other words, if the 24 bit version was directly converted to 16-bit, nobody would be able to hear the difference in a blind test.
Streaming = you don't own it.
excellent piece.
finally a wise point of view about this topic
For sound quality? 16bit and 24bit will sound identical, 24bit has a higher dynamic range but its utterly pointless for any music listener, im not sure why people keep dragging up this old argument, it was proven in 1982 that a redbook CD will be all the sound you will ever need. All this high res nonsense is there to make money nothing else, in fact it can sound worse due to harmonics and crosstalk in the high frequencies.
What i find laughable the most is when people spend thousands on turntables and amps to try and make a vinyl record sound bit-perfect like a CD, vinyl was never capable of reproducing perfect sound, and it never will be no matter how much you spend, a cheap £20 cd player from tesco will sound better than any record deck.
And BTW only classical music can saturate a cds dynamic range, rock/pop etc are absolutely horrible, most songs today have 12db Dynamic range due to the over use of sound crippling compressors
A rolex is 256 times better than a casio, but they both still tell the correct time.
It's all about making money, are you really gonna hear the difference with the human ear? it's the actual source quality used to put on CD's, and Vinyl is what makes the most audible difference... and of course the source used to upload music to all these streaming services...
Yup @@miked91101
We already should have 32bit/384Khz audio CDs
Very interesting. Thank you!
Great video, from the guy that sells SunLife Insurance. Hope June is well. Envious of your cheese and tomato sandwich.