008 MQA part 1; Why 24 bit 192 kHz audio?

[@TheHansBeekhuyzenChannel]

To the viewer that wrote: "This German knows its stuf": there is nothing wrong with being German, but i'm Dutch:-)

[@TheHansBeekhuyzenChannel]

:-))

[@rrrandommman]

+The Hans Beekhuyzen Channel Thank you very much for this video! I love your relaxed and detailed explanation, it makes it very easy to follow.

[@TheHansBeekhuyzenChannel]

+rrrandommman Thank you, you're most kind.

[@TheHansBeekhuyzenChannel]

I don't think this remark will make you popular in The Netherlands.😗

[@TheHansBeekhuyzenChannel]

It has been some years since I was involved in recording. I am not the right person to answer this question since my knowledge of contemporary equipment is not up to date. Sorry.

[@jaysheth2090]

World needs more teachers like you.

[@TheHansBeekhuyzenChannel]

+Jay Sheth Tnx, you're most kind.

[@eduardocampos3643]

@@TheHansBeekhuyzenChannel I know nothing about digital sound, but I am a mechanical engineer. I understood the reason behind most of the things you explained, but I want more. In about 20 minutes I learned so much. Will follow you and keep learning. Thanks!

[@adrianallen5347]

Perhaps the best explanation of sample rates and depth I have heard so far. Thank you Hans.

[@petrofski88]

Agreed!

[@davelogeman]

This has got to be the best explanation for hi-res audio that I have come across. Well done.

[@merrillfalk]

This is all Physics. I taught Physics for 32 years, and I am VERY impressed with your work. Brilliant. Thank you for the video.

[@owlmuso]

Thanks Hans, this is the best explanation of hi res audio I have seen. Most commentators only focus on frequency and not time resolution. So your focus on the effect of time resolution on how we experience sound is really illuminating. Thanks once again

[@wwoolworth]

I wish KZbin had more people explaining subjects as thoroughly as you do. I can not imagine the time you must put in to do just one video.

[@TheHansBeekhuyzenChannel]

Well, it's about 20 to 30 hours per video. But I think it pays of.

[@bjornahh87]

this IS the best explanation I`ve come across in a long time, I allways knew that 44.1khz/16bit wav sound file is missing everything that you as a listener need in a "soundpicure" for me I'ts always been like looking at a black and white picure when you could look at the same picture live in colors with the live inviroment that comes with it. I allways use the higest available bitrate and codec beacause i want the most clear, crisp and sounds in the backround, eccoes, endings of the instuments, reverb, and i can go on and on ...I will show this to anyone i meet that tells me that 44.1khz/16bit is all we need..

[@stevenjackson8226]

Hans, you are the man :) Your typical clarity, directness, and useful organization and development of the material. And, of course, a bit of your dry humor. Nice, and thanks.

[@brianabdb]

Hans Beekhuizen, you answered EXACTLY my question I searched on the internet for. Thank you for sharing and for the way that you share this information! Keep spreading the gospel ;)

[@ProjectOverseer]

Wow Hans. This has to be the very best description of human hearing possibilities and why 192kHz sampling does have benefits - brilliant 👍

[@adleneboulebtateche156]

Your pedagogical method is unrivalled and your english is perfect. At first, I thought you were British.

[@TheHansBeekhuyzenChannel]

🙏🏽🙏🏽

[@EnriqSandoval]

One of my favorite of your videos yet. Thank you so much. I tune in almost daily to see your video. I re-watch many of them because there seems so much information in them. Again thank you.

[@anmolagrawal5358]

Excellent explanation. Fabulous! I searched about HiFi audio on Wiki, numerous forums and threads and read many articles but your video was very explanatory and cohesive in its delivery. Cleared my doubt about this topic for eternity.

[@walkinthrutheparkbymr.melo3905]

Thank you from those of us who have been advocating 24/192 Audio!

[@TheHansBeekhuyzenChannel]

🙏🏻

[@gullydeluxe]

Great explanation!! Simple, informative & professional!!!

[@burakcelik3752]

Simply an excellent explanation of core concepts about digital audio (and even some good information about analog sound and biological hearing process!) in just 17 minutes that usually take someone around a year to have a good grasp through getting them from different sources. Extremely refined and easy to understand !

[@TheHansBeekhuyzenChannel]

Glad you enjoyed it!

[@johnaweiss]

4:45 Musical octaves may help people understand logarithmic frequency. At the bottom of the piano keyboard, octaves are about 30 Hz apart, but at the top of the keyboard octaves are about 2,000 Hz apart. But to the ear, an octave sounds like an octave anywhere on the keyboard. The lowest C on the piano ("C1") is about 33 Hz. The note one octave higher (C2) is double that: about 66 Hz. The second-highest C (C7) is about 2,100 Hz. The note one octave higher (C8) is double that: about 4,200 Hz. But to the ear, the distance sounds the same: an octave.

[@sdnalednas]

I wish I were able to give this video more than just the one 'thumbs up'. With the release of the Pono and other various high definition audio options out there, the banality of linear thinking criticism has been very frustrating. Your explanation of the density of resolution in high def audio is what all naysayers need to here/see. As you outlined, our listening environment will provide functional limits to the transduction of a source file. With care given to our signal chain and environment the rewards are well worth it. Thanks for putting this out there, and I'll do my best to spread the word.

[@TheHansBeekhuyzenChannel]

Tom Sandelands Tnx Tom. But I wonder whether the 'linear thinking brigade' will be convinced by it. They have their measurement equipment to judge, we use our auditory system. Don't forget how the church 'proved' that our planet was planar: those ships that didn't return went over the edge and those that did return clearly came from the same direction they departed in. So the world couldn't be globe shaped.😐

[@davidswain1718]

Superbly explained, I applaud you. This should be compulsory viewing for anyone interested in digital audio HiFi. Congratulations on your excellent presentation.

[@TheHansBeekhuyzenChannel]

@ Mary White: nowadays I would record in 24 bit 176,4 kHz and convert - when needed (for for instance cd production) - down to 16 bit 44,1 using a quality downconverter like the Weiss Saracon.

[@TheHansBeekhuyzenChannel]

+The Hans Beekhuyzen Channel Makes completely sense.

[@belcantobrasil]

Very good video!!! The only one that explains bits and samples not only applied to the effects on the non hearing part of the music, but also on the hearing frequencies!!!!!

[@SimonCash]

That is the the most straightforward and easy to comprehend analysis of this subject I've seen so far. It was concise and informative and put the whole Hi-Res audio subject into perspective for me. Excellent video. Well done Hans, I'm off to watch Part 2 now :)

[@TheHansBeekhuyzenChannel]

Wow, thanks!

[@johnmarchington3146]

Thank you, Hans, for that extremely lucid explanation of sampling and bit depth

[@TheHansBeekhuyzenChannel]

Glad it was helpful!

[@johnmarchington3146]

@@TheHansBeekhuyzenChannel Time-smearing is still a bit of a mystery for me

[@MoonwalkDancer]

Thank you for all this information. I use 24bit - 192Khz when I record from my vinyl player to flack files. The only sad part is that KZbin converts the audio inn to aac so I can't really give people the real Sound Quality experience. I love your work and your passion for Music. Merry Christmas & happy new year 2018.

[@seraphthecreator]

The sampling rate is ok but 24 bits is a waste considering you'll be lucky to get more than a dynamic range of 60db on vinyl

[@krane15]

Correction, youtube compresses the file

[@EssenceofPureFlavor]

This is really well done. I read about mqa recently, and didn't really understand the point. This helps a lot.

[@TheHansBeekhuyzenChannel]

Thank you

[@luisgambao1255]

Another great lesson, Mr. Hans. Your explanation for what you say between 6:09 and 6:18 could help people who don't agree with high resolution audio to understand things better. I thank you.

[@TheHansBeekhuyzenChannel]

Thanks for your kind words. Spread the word😃

[@alexrichardson6461]

Like so any others here, this is the best description of digital audio I have ever heard - or seen. Thank you so much

[@TheHansBeekhuyzenChannel]

You're very welcome!

[@MagnaHifi]

Very well explained and well-researched story, congratulations Hans!! Not to open a discussion here but I truly believe every word you say but we have to be realistic and in the end it comes down to the record quality and whether your home system is good enough to allow high-resolution touch. With a steep learning curve, CD quality (digital audio) is improved a lot over the last 20 years while mainstream consumer was steered towards MP3. Looking forward to MQA and hopefully this is going to be a widely accepted technique and we finally set and or 192kHz/24 bit as "the new default" standard!? Thank you Hans, keep up the good work! - Jos - Magna Hifi

[@BVcello]

Bijzonder interessante uiteenzettingen. Hartelijk dank om dit zo gedetailleerd met iedereen te delen. Mvg

[@danicooke346]

Amazing clearly explained and occasionally with some dry humor. Love this.

[@TheHansBeekhuyzenChannel]

Glad you enjoyed it!

[@andrew8992]

Astounding video. There is so much knowledge packed in here i had to watch this several times just to take it all in. Thanks Hans.

[@TheHansBeekhuyzenChannel]

Glad you enjoyed it

[@thomasgrichen7206]

Great video. Small correction, when talking about the choice for 24 bits over the desired 20 bits. You state that computers always need to work with powers of 2. First 24 is also not a power of 2. But the reason is that data is still mostly stored in bytes, sets of 8 bits. So a sample is stored in a multiple of 8. Still 20 is not a multiple of 8, but 24 is. You probably knew this, but for the people who might got confused.

[@hansbeekhuyzen7770]

I have corrected that earlier but thanks nevertheless.

[@martinsmithtimes637]

I just add it have nothing to do with computers (in a sense of general purpose cpu like intel motorola or risc) becouse for such machines there is no diference (pass 16 bits) to work with 20 or 24 bits or 17, 25 or 32. instructions for arithmetic logic and data access is either 8, 16 or 32 bit. So why 24 bit ? because this logic did not come from computer but DAC and dedicated microcontroller dsp circuit. here you need to distinguish between bit resolution for storage file format and conversion DA/AD in DAC first quality cd palyers used phillips TDA'40 chip whixch was 14 bit and to many it sounded better then contemporary 196 16or 24 DAC ! (but for different reasons), the 16 bit become standard but in TDA'83/85 chip only major 13-15 bits worked in practice hi end dacs used 18 then 20 bit ( HDCD usues 20 but stores it in 16 bit file) so it was hen dac evolution reached 20+ bits it goes for for third byte of file data ( so far 2 bytes=16 bits) third mean 3x8=24 for justified usage of 24 bit resource 24 bit dac to Dir ctly interpret the file format started to.make more sense. ultimately increased compute r power and RAM allowed for mastering on desktops in 24 bit. where it makes utmost sense since you store 16bit sample plus 8 bit volume ! that allows for people.(like me ) who write dsp FIR IIR FFT MEM or mixing algorithms in CPU code for reqltime antialiased performance to freely use 32 integer ( back in 1999-2004 FPU was slow) routines that performed on general.purpose computers with highest quality. The experience garhered during this R&D I will be happy to share in some other post if anyone is interested to make clear why 96khz or higher frequency and bitdepth is crucial to quality and how come digital domain signal procesing is so difficult for stereo positioning soundstage or so..

[@MichaelOZimmermannJCDECS]

Extremely informative and very well explained. I have not yet heard anything like it! Thank you Hans!

[@jeffjardine8411]

Simply an excellent video that makes something complicated understandable for anyone who is willing to take the time to pay attention. Doing so shows a certain mastery of the subject. Thanks for sharing your knowledge!

[@ArjanAdriaanse]

At 14:40 you imply that 24 is a power of 2, which it is not. A more plausible explanation for the use of 24 bits instead of 20 is that 24 is a multiple of 8 and therefore one sample can be contained in exactly 3 bytes, which are 8 bits each.

[@TheHansBeekhuyzenChannel]

If you scroll down you'll see that this was already corrected by me. You are right it is about 3 bytes.

[@hellsacolyte]

Fantastic video Mr. Beekhuyzen. I learned quite a few new things with it!

[@VitalityFactors]

Thank you so much for the great explanations. It perfectly answered the questions which I had in mind on Hi-Res audio.

[@zorankalina4399]

Ah...so nice to finde such a brilliant reports From a better times of all of us 🙂

[@TheHansBeekhuyzenChannel]

👍🏼

[@wojciechczupta]

Great job Hans

[@homosepian1234]

dear Hans! your videos are superb and pleasent to ther ear and eye, i just cant stop watching them 2-3 times each. thank you so much for the lovely lectures and examples! keep on lecturing us :)

[@TheHansBeekhuyzenChannel]

You're most kind. Tell the world and help me reach a larger market so I can continue doing this. I love it too:-)

[@stephencosta6814]

I just watch this video again and once again I am floored by you beautifully explained you're the best

[@ezrazski]

Perfect - no nonsense, all fact, all in context. Too bad this type of information is so hard to find online. I'm going to share this link with the types who continually come to wrong conclusions about digital audio.

[@hansbeekhuyzen7770]

ezrazski I rather like the absence of this kind of info on the web, it strengthens my 'position'😀 Thanks for the compliment and thanks for sharing.

[@magoostus]

xiph.org

[@TheHansBeekhuyzenChannel]

???

[@TheHansBeekhuyzenChannel]

Yeah, that is knowledge of 10 years ago. And predominantly based on assumptions. Really....

[@MarkvanderLoo76]

Weer een interessante video hans, ik wacht alweer op de volgende!

[@TheHansBeekhuyzenChannel]

Tnx, spread the word

[@agosto.mp3]

Awesome video! So what are examples of good filters to use that generate less amount of artifacts?

[@TheHansBeekhuyzenChannel]

I don't understand your question, I'm afraid.

[@slam854]

Great refresher course. I got a scare today when I learned CD transports are ending production. Time to step up to the real world plate if I wish to continue the audio passion. Thanks Hans.

[@Kangnam81]

I am fully impressed after viewing your video. Really professional presentation, perfectly ordered knowledge, modest but reach form. I remember that TV was like this in 80s I our communist Poland. There should be a separate free of charge TV change in the whole EU with people like you.

[@Antony_blr]

Hello there Mr Hans, this is one of the coolest videos on the topic and it is spot on!!! Keep it coming.

[@TheHansBeekhuyzenChannel]

I will,, thank you.

[@capwkidd]

Excellent video! I knew some of this info, and learned something. You packaged it all into a nice clear and understandable video, bravo!

[@danielmartin7505]

I always feel a bit privileged to listen to his explanations

[@carlosfantube]

Thank you Hans, amazing video.

[@ferakles]

At 11:30 you say that auditory time resolution is somewhere between 5 μs and 10 μs. Can you pls cite a source of such an information and how it was estimated? Thanks

[@TheHansBeekhuyzenChannel]

There are plenty mentioned in the reference list of A Hierarchical Approach to Archiving and Distribution by J. Robert Stuart, Peter G. Craven, Audio Engineering Society Convention Paper 9178

[@tigertiger1699]

Your explanation even had the wife interested..🙏😂… truely high praise for your style Hans😂🙏🙏🙏

[@TheHansBeekhuyzenChannel]

😁

[@cameronproaudio]

About 15 years ago I attended a weekend seminar for STE (society of television engineers) and SMPTE members in Palms Springs, California. One of the guest speakers was the head of engineering for THX. THX had conducted an experiment over the course of a weekend at Fox Studios in Century City of all the various flavor of audio encode/decode for various record and playback systems used in record industry and motion picture production. They invited "golden ear" types from all over the world to attend. Record producers, engineers, re-recording mixers, audiophile types, etc. Over the course of the weekend, THX conducted many double blind ABX testing sessions of these systems with sampling & bit depth resolutions from 44.1k/16 up to 192k/24 and everything in between. These test were conducted in what is considered to be one of the best acoustic environments through one of the best monitoring systems in the world. At the end of the testing, Much to THX's surprise, the tally from the double blind testing indicated two main things: nobody could tell the difference between any of the systems or various sample rates/bit depths; in some tests, 44.1k/16 scored slightly higher than average. The conclusion was so startling to the THX folks that they planned on redoing the testing to make sure they didn't mess something up. I don't know if they ever got around to it. But in my opinion, it sound spot on based on my personal experience. 44.k/16 performance far exceeds frequency response and noise performance of any analog audio recording method for real world use. And IMHO, higher sample rates and bit depth for end user delivery is a waste of disk space and processing power. Until you can prove your case with true double blind ABX testing, I'm a skeptic that there is any tangible improvement with your methodology. 44.1k/16 with a lossless codec is more than good enough.

[@TheHansBeekhuyzenChannel]

+cameronproaudio Let's not go into the ABX test discussion here. It has been discussed many times on the web and elsewhere without any chance on consensus. I wonder if you have watched the entire video since I do agree that 44.1 kHz exceeds the bandwidth needs for our auditory system. That's not the reason I am pro 192 kHz. Whether 16 or 24 bit is needed, depends on the equipment used and whether the listener is experienced enough to identify the 16 bit artifacts. BTW, 24 bit usually is pointless since analogue audio has a maximum dynamic range equal to 20 to 21 bit. It's just that digital equipment wants to work with bytes (group of 8 bits) and three bytes sum up to 24 bits.

[@petekay67]

Amazingly concise and educational. This is the real stuff. Thanks very much.

[@TheHansBeekhuyzenChannel]

Glad it was helpful!

[@HarshadContractor]

Sir, Such clarity in your description in both parts of your video. I'm 70 years old and use FLAC to convert my CD's to FLAC, I am able to hear the quality difference even when I re-record to a reel to reel tape recorder. Am I wrong somewhere? Please advise.

[@TheHansBeekhuyzenChannel]

I am not sure I understand the question.

[@jvandekant]

Great explanation! One thing that keeps bugging me: “computers need powers of 2 and 20bits isn’t”. But 24 isn’t either, so I don’t understand that part.

[@TheHansBeekhuyzenChannel]

24 bit is 3 bytes, 3 x 8 bits

[@jvandekant]

@@TheHansBeekhuyzenChannel Thanks Hans, that makes sense. I should have thought about that. Theoretically 20 as well as 24 is the sum of powers of 2, but arranging into 8 bit blocks is obvious🙄. I started going through other topics on your channel and I really like them!

[@slimyelow]

also new to me was how expenses increase in low pass filtering technology and how it can muddie the mid range, especially when sampling below 96khz.

[@georgeparaskevas8091]

My name is George Paraskevas and I thank you for making this video.

[@BFArch0n]

In a world of 4k video streaming....we can handle 192k/24bit music. Especially when we have the option to download it to reduce streaming needs.

[@MrLawrence0071]

A truly fantastic video! Thanks for that. Dank U wel! (Belg) ;-)

[@TheHansBeekhuyzenChannel]

+Lawrence O My pleasure/graag gedaan. Spread the word.

[@roycspary8923]

correct and very good, few in the audio world seem to appreciate the evolutionary consequences of our hearing, thefolds in the outer ears cause a variable time delay of high frequencies, which is how we perceive the height of a sound, so imagine a primitive hominid, pre Australopithecus, up in a tree at night, that twig snap could be above, below or level, now if that is a leopard, which we know hunted early hominids, well that ancestor became cat food, the ones that got it correct passed on their genes for exceptional hearing in the time domain, as an audiophile in my 20's spoiled by a dream system, I came to correlate jitter and filter types with sound quality, thank you for not only confirming this, but explaining the how and why, that earns you another subscriber

[@TheHansBeekhuyzenChannel]

My pleasure

[@nyrbsamoht]

thankyou so much. you just put together so much different stuff I have been reading and alot of it it suddenly makes sense.

[@al166012]

This video is good even for those who dont know so much in this topic. So simple explanation, thanks Hans

[@TheHansBeekhuyzenChannel]

My pleasure.

[@jonaswox]

And btw the referenced "philosophy" is not some abstract weird concept. Its relatively clear that since no person is in the room, nobody is around to percieve the sound, hence "there is no sound". From a philosophical perspective this is a very straightforward and reality bounded claim.

[@TheHansBeekhuyzenChannel]

Let's not go there, there have been discussions going on for 50 years on this topic......

[@MSJourney8967]

you are talking all about high-res of music, how about voice? can you tell about why we need a super wideband or full band voice

[@TheHansBeekhuyzenChannel]

The same applies to voice. Watch kzbin.info/www/bejne/nZbEoHiqYp6JY9k

[@tothehilt]

Excellent video with very high quality considering the compression time.

[@TheHansBeekhuyzenChannel]

Thank you!

[@justinhu672]

WOW, everything is clear and authentic as always. Though we can hear only up to 20kHz, but some believe the resonance and the pounding effect of the frequencies that we couldn't hear might have some effects on the "color" of the music as a whole, really curious about that, would you share your view on this. 192kHz 24 bit is reasonable now, but I mean should we sample as much as we can in the future since storage and computing power as well as mechanics have advanced so much, not just to increase the resolution but to preserve the unknown effects.

[@TheHansBeekhuyzenChannel]

There are those that suspect interference between signals above 20 kHz might be of influence. This is highly unlikable. Lets say two frequencies, 25 kHz and 26 kHz would produce a 1 kHz tone (the difference signal), this signal would be lower in amplitude than the amplitude of both HF signals since it is the difference signal. In audio the power spectral density (energy or power per Hz) is inversely proportional to the frequency of the signal, meaning that there will be very little energy at 25 and 26 kHz to begin with. Our ears use a kind of automatic level control that adapts the sensitivity to the sound, like the eye varies the iris to the light conditions. And like the eye is incapable to see into dark spots when a bright light is next to it, the ear is incapable to hear a low level signal when a loud signal is next to it. This is called masking. given the 1/f rule, there already is a 28 dB difference in loudness between the 1 and 25 kHz signals. So it is highly unlikely the difference signal will be noted to our ears.

[@maartenvankranenburg7780]

very well explained and yes video makes it much easier to digest especially on larger articles. Jos says at the end it comes down on record and playback quality. With today's converter and headphone quality i am not so worried about the playback side. Since majority often rules it would be interesting to know what real significance it makes to the general public if presented with the two options. Thanks, maarten

[@TheHansBeekhuyzenChannel]

Wait till the second video comes out. As it seems now MQA will also be able to improve 44.1/16. Still working on the last details.

[@Hexspa]

I just recorded a Rhodes part out of Kontakt in Ableton using a MOTU Ultralite MK3 @ 44.1, 48, 96, 176.4 and 192kHz. I listened just now many times through an Avantone MixCube in a quite silent, acoustically treated room. I would say the differences are clear. To me, 44.1 is unsuitable for recording anything to be released. 48kHz, which I have been recording at, seems most suitable for rehearsals. 96kHz could record a band or live performance if buffer was a factor. 176.4 seemed like a compromise and 192 seemed live, as though the recording wasn't even a recording. 44.1 seemed trapped in the speaker and clearly an "audio file" while 192 seemed to touch the fingertips of the performer. The reverb seemed to have more definition and the depth placement (how far the dry sound was relative to the reverb room) was far more clear as the resolutions went up. I would like to hear what others say about this "timing" factor; the widespread argument up to now is all about frequency-response. Even regarding that, the transients seem sharper with higher resolutions at 96kHz and above. Yes, it was a MIDI recording of lower-resolution samples but I could still hear a definite difference and wouldn't be surprised if I could pick among the three in double blind tests. Thanks again for this video. Now I just need a way better computer.

[@grizcuz]

+Hexspa I think, unfortunately in many ways, it doesn't matter whether you can hear the differences, you obviously know what to listen for and how to describe the differences. I'd be interested if you think an 'average' person who enjoys listening to music but isn't an audiophile could also hear the differences you describe unprompted? I've heard some younger people actually say they prefer the sound of MP3 files these days compared to uncompressed CD or WAV! It's going to be the normal, everyday consumers of music who will determine how music's encoded and the formats of future releases, rather than people like you and me who want the most faithful representation of the musical performance possible. Yes, we'll probably be able to find services that will offer some kind of lossless format or another for an increased price, but I'm afraid the majority of consumers will probably still be listening to their music collections in a way that's inferior to CD. It's very odd that in a post CD world the fidelity of music has decreased. That was probably understandable when digital storage was limited and expensive, but not now when there's ample inexpensive ways to store lots of large files, both locally and in the cloud. That's without even considering the fact that lots of people these days don't own any kind of 'HI-Fi' device or even store physical files outside of what they can stream through their phones. I shudder to think what the bit-rate of music streamed through something like Spotify is? And I think I've read that Soundcloud replays all music at 128kbps? I'm actually quite concerned in some ways about the future of hi-fidelity music. When you examine the budgets that artists/labels expend on recording/studios and just how much they've decreased in recent years. Then couple this with how the music is probably going to be mostly consumed [streamed through phones]. Then it's not much of leap to see a future where the highest quality possible recorded sound isn't valued in the same way it has been since sound recording was first developed. Although, I suppose the availability of far higher quality musical technology equipment that isn't nearly as expensive as it once was may go some way to counter this possible trajectory. When people can produce music in their own homes that was once only possible through attending massively expensive professional studios, then it's going to be the skills of the people involved, rather than the size of their own, or their record company's, wallet that counts. And the death of lots of recording studios around the world will attest to this. I hope I'm being unduly concerned and perhaps we've been going through a period where people have been force fed the fast food of MP3's. Once they savour the gourmet dish that is uncompressed, high bandwidth music they'll never want to go back. But to be honest, most of the infrastructure behind how many people consume their music is pointing towards the former rather than the latter at the moment. Apologies for turning this reply into a mini essay that drones on as well. And I haven't even touched upon how music is mastered these days and the whole dynamic range/loudness wars debate!

[@slimyelow]

The most amazing part for me was the explanation of human auditory time resolution (7 micro human seconds) I have engineered for 25 years and have never heard about this perception of sound.

[@hansbeekhuyzen7770]

+Slim Yellow That's understandable. It is only due to relatively new neuro-scientific research this became known.

[@slimyelow]

after my last post I started recording up tp 12 channels at 192. When I use UAD software and Unison Preamps for tracking guitar through vintage Marshall amps and effects at 192 I am just blown away, and will never go back.

[@roderik1990]

+Hans Beekhuyzen That sounds.... awfully pseudo-scientific. How in the fuck could you even ascribe a well-defined temporal resolution on the order of microseconds when neuron activation times are on the order of milliseconds? And the ear closer to a filterbank in working, than something that samples with a certain resolution?

[@slimyelow]

I believe the flaw really is in the design of them mother fucking hardware low pass filters, built into the converters. They really fuck shit up in the mid range. I assume that at 192 they are obsolete, and therefor good-mother-fucking riddance. -yes?

[@TheHansBeekhuyzenChannel]

+Slim Yelow Well, they mostly are. That's an important point of higher resolutions.

[@tomstickland]

The CD system designers were very astute. Their choice of sampling rate and resolution were well matched to the human ear. There's blind tests out there that have shown that people cannot determine any improvement beyond a certain resolution. If MQA is claiming higher bit rates and resolution but at an equivalent data rate to a CD then it has to be more lossy, which it is, or at least partially. I really can't see any need for high bit rate FLAC files, let alone MQA distorted versions.

[@tomstickland]

Regarding timing, you have to ask how a microphone is ever going to deliver an instantaneous step change which the higher sampling rate would detect to higher temporal resolution. Even an instantaneous rising edge being fed to a speaker and then passing through air to a microphone, the microphone would pick up a ramp and the digital file would have samples of the ramp which means that the temporal position would be better than the clock rate suggests.

[@TheHansBeekhuyzenChannel]

Microphones are a given fact. And when listened to them prior to recording, there is nothing wrong with them. It is after passing through the anti arising filter and the reconstruction filter, the timing of the signal has changed, harshness in the mid range appears and pace & rhythm are changed. That is what MQA is fighting.

[@mbsnetwork2650]

Hoi Hans! your insight and ability to comprehensively explain complex topics in an interesting and enjoyable way is masterful. I trained as an electrical engineer and it’s amazing how much more I can learn from watching your videos. Thank you so much!

[@2604victor]

Hans , you are great ! please , continue this way teaching us about the misterious digital world! Thank you!

[@TheHansBeekhuyzenChannel]

Well, If you like my work, support it using Patreon or Paypal: My Patreon page: www.patreon.com/theHBchannel Paypal: www.paypal.me/theHBchannel

[@pabloosvaldopenizzotto1098]

Hi Hans. Thank you for your detailed explanation. I have see many hi-res audio KZbin videos that sound amazing in my hi-fi system. Even sounds nicer than a CD recording. Nevertheless I have seen many comments saying that KZbin streaming is unable to reproduce hi-res audio files. So my question is: what is the actual streaming rate of such youTube videos ? The good sound quality we can get in such videos is more related to the way the sound track was mastered? I look forward to your comments. Thank you!!

[@yuvals6409]

Please see my post below. It is true that KZbin files are compressed, but using the set up that I describe below, I get a marvelous SQ. And yes, they sound nicer than a CD recording (and I have a Teac CD-P3450SE Player (tweaked by Ken Ishiwata). The question I can't answer is what is the actual streaming rate of such youTube videos. At any rate, I suggest you buy the set up I have described below - the total cost is $25. I am convinced you will like it.

[@fritzwalter8933]

Very interesting and straight forward article with good explanations about the issue wich is often not understood even by prof audio engineers (when it's about the digital domain). One thing I would like to get, is a source link for the 7µs auditory time resolution you mention in your video (at 11:40). I only found references being in the ms sec range, not µs wich would render that argument void.

[@TheHansBeekhuyzenChannel]

On page 6 of "A Hierarchical Approach to Archiving and Distribution" by J. Robert Stuart, Peter G. Craven those sources are mentioned. You can download the paper here: www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKEwjkk4mp5oLyAhUVO-wKHaf0C78QFjABegQIBhAD&url=https%3A%2F%2Fwww.aes.org%2Fe-lib%2Fbrowse.cfm%3Felib%3D17501&usg=AOvVaw15Of_jjRB09VOM4U81S2Hz

[@fritzwalter8933]

@@TheHansBeekhuyzenChannel Great, thank you! For those who are interested, here is the paper for the mentioned source: www.ejta.org/en/kunchur1 wich describes the scientific analysis for the 7µs auditory time resolution.

[@ScrewballMcAdams]

Very easy to understand, even for me as a non-audiophile.

[@jaysheth2090]

Hans, can you add in or do a quick one about the Frequency response curve and its relation to the audio quality.

[@TheHansBeekhuyzenChannel]

+Jay Sheth No;-) Just kiddin', It's a very simple question that can't be answered with a simple answer. Ideally you want a straight frequency response over the audible range (say 20 Hz to 20 kHz) but ideally you also want a straight phase response, no noise and so on. Depending on the device trade-offs have to be made. For reconstruction filters in d/a-converters for 44.1 kHz sampling many manufacturers choose a very slight roll off at 20 kHz (a few tens of a dB) to have a better phase response. If you look at the frequency response of a good speaker, you see deviations of dB's for there the trade-offs are different. We still understand far too little on how our brain processes what we hear. I measure all devices I review with one of the best measuring devices around but it will never tell me how good a device sounds. It can only tell if there are faults in a device of what compromises are chosen by the designer.

[@harishparvatham2906]

Hi you said 196khz is required for falling into the less than 10. But does does it come to my ears in that rate when my speakers don’t produce such high hertz. Or they don’t require to. And does high res speakers that go till 40khz have an impact wen compared to 20khz speakers in the context of source resolution. Thank you

[@TheHansBeekhuyzenChannel]

You find the answer in this video: kzbin.info/www/bejne/nZbEoHiqYp6JY9k

[@n.r.2258]

Does it makes any sense to resemble from 44.1 to 198 khz like you can do in Volumio ?

[@TheHansBeekhuyzenChannel]

The quality of the information in the file doesn't improve. But it makes it easier for many affordable DAC's to convert it to analog.

[@RobHTech]

Very good vid. Professional. --In the part explaining the ADC process, I am a bit curious. For some music that is created in the digital environment, such as a lot of Techno music that I listen to, is there any real analog-to-digital conversion? Is this the case, or am I misunderstanding? And, if so, then I would presume that the maker can synthesize in just about any sampling rate, etc. that one's computer is capable of. But, playing that music than becomes a challenge. ---And, Tidal, for example, offers hifi quality, over CD quality, streaming. But, my iPhone 8, which is my main source of music in my car, caps the output at a max of 48kHz. Otherwise, an external DAC that bypasses the internals is required, such as DragonFly Red. --Iphone 8 plus and car: I use bluetooth a lot, like many. Investigation reveals 8 plus uses BT version 5, supporting A2DP codecs: 1. AAC--16 bit/44.1kHz/264kbps--and SBC (default codec)--16bit/48kHz/320kps. So, it's either accept SBC or something else, like a wired connection, which brings up a host of other problems. --The lightening port goes up to about 480Mbps. --The apple lightening to 3.5mm jack converter actually has an internal chip set that includes, to name a few, a DAC and an amplifier.

[@TheHansBeekhuyzenChannel]

The software used to generate sound is usually not of such high sound quality that high sample rates do pay off. But theoretically you are right. If sound quality is your thing, you should not use a smartphone as source.

[@Kapitaen_Flauschbart]

Very nice work, thank you!

[@McCulsky]

Absolutely clear and great explanation. 2 thumbs up!

[@josephloesch8575]

Great teaching. And I have a question for you, Hans: If I take my 192Khz/24 bit ProTools recording of a singer and acoustic guitarist, and want to preserve as much nuance and fidelity as possible, what do you recommend as a mix-down or "bounce" approach? My skills are those of a singer and acoustic guitarist, rather than an engineer. And I make one recording after another at 192/24 that has shimmer and finesse, but apparently stumbles into degradation issues when I put it onto a CD for someone to hear. I found your site today because I went looking (once again) to see if I can discover any transfer/mixdown system, or any considerations (dithering?) that I am misapplying. Clearly there are some fine sounding CDs being produced--what is the bridge between my 192k/24, and that clarity? --Aside from issues of $10,000 microphones and so forth. Just, how to get what sounds so good to translate onto a CD. Any answers are welcome! And if you don't have time, thank you for the excellent teaching you have given. Joseph Loesch, Los Angeles, California

[@TheHansBeekhuyzenChannel]

It depends on what your means of distribution is. If it's cd, you better produce in 176.4 kHz sampling since that is a multiple of 44.1 kHz and you need to decimate from 24 bit to 16 which needs proper dithering. But if you only offer downloads or USB sticks, you can downsample to 48 kHz 24 bit. If you need to convert to MP3, try to find a plug-in that can convert directly from 192 kHz to MP3. I can't help you with the choice of plug-in's, I'm not following that market anymore and thus amp not up to date.

[@josephloesch8575]

Thank you for taking the time to answer my question, and for mentioning the different considerations. The ProTools 10 that I am using has a save-to-MP3 option that allows 320kbt/s (48000Hz); and maybe there is a plug-in that renders higher quality, which I will look for. ProTools includes "PowR" dithering; I'll see what I can learn about "better dithering," and so forth. --When I'm happy with the sound of what I'm working on now, I'll send something to you. Thank you again.

[@tocavlad]

your explanations are great....first time when i'm speechless

[@TheHansBeekhuyzenChannel]

You're most kind. Thank you very much.

[@heberthybr]

OK so... smartphones with 24bit/192khz phone jacks are able to reproduce high resolution files, right?

[@TheHansBeekhuyzenChannel]

in theory at least. But my guess is you'll be hard pressed to find one that really offers an analogue audio result equal to that.

[@heberthybr]

The Hans Beekhuyzen Channel You sir mean the right smartphone or the right track? For instance: I have a Sony Xperia Z5 Premium with 24bit/192khz audio jack and Im playing an album from HDtracks.com. This means that Im hearing a high res track with a high res player, right?

[@TheHansBeekhuyzenChannel]

I have not tested many smartphones, so I can't really comment on models and types. But the smartphones I know are limited in sound quality compared to home equipment. 24bit/192 kHz does not sound better if the hardware - the smartphone - does a poor job in rendering. But the Sony you mentioned might be better than I know. I really can't tell.

[@andrew8992]

Hans, can you explain (in simple terms !) why 1 bit is equivalent to 6db? The only explanations I can find are mathematically quite involved and beyond the processing power of my merge brain. Thanks in advance.

[@TheHansBeekhuyzenChannel]

Every bit doubles the amount of levels that can be coded. The doubling of a voltage is 6 dB.

[@MrApplewine]

How much better is 24 bit 384khz (DXD) over 24 bit 192khz PCM for added improvement to the timing (not audible pitch) for listening? My hi-fi integrated amp/dac can't decode DXD (384) just 192khz.

[@TheHansBeekhuyzenChannel]

I wouldn't loose sleep over it. There is hardly any music available either in 384.

[@gerhardgroen]

I have a theoretical question about sounds we hear, I am thinking about lately. We never hear one frequency at a time, as we know. Allthough that’s what we talk about mostly. We in fact allways hear interference patterns of many frequencies combined, harmonical or not. These patterns must be different (in my opinion) when a frequency of say 30.000 Hz is added to this interferencepattern. In my opinion it cannot be excluded that a human ear in fact does hear a difference there in lower frequencies. What is your opinion about that?

[@TheHansBeekhuyzenChannel]

That's an old theory and when you analyse what is happening, you will see that if these interference patterns already excist, they will be at extreme low level and thus be irrelevant cq inaudible.

[@MuenchenBob]

Very nice explanation! I really like your twig snap example.

[@TheHansBeekhuyzenChannel]

I’m afraid the twig example is not mine. I just borrowed it😌

[@DDastrup1234]

Thank you Hans, early in the video you taked about 256kb more than sufficient in your car. my question is what part is the limiter of sound fidelity coming through. Are the speakers the limiting factor or is it the noise while driving the car? thanks again this video has been very helpful.

[@TheHansBeekhuyzenChannel]

First, I was talking of 256 kbps AAC, not 256 kbps MP3. That makes a difference! The noise a mid class car makes while driving is somewhere around 75 dB SPL A or worse (at home it's tens of dB's lower). That will mask a considerable part of the detail in the music. Whether the speakers or the electronics (amp, player) will further reduce the quality depends on the equipment used. But if you could afford a current Rolls Royce, S Class Mercedes, 7 series BMW or equal (on gas, not diesel), the noise might be lower (that's what I read in the car magazines) and when your car is equipped with a Burmeister or other very high quality car stereo, full CD- quality (ie FLAC, AIF or ALAC) might be a better choice. But I'm no car journalist and have never driven such luxury cars and therefore can't comment on that.

[@DDastrup1234]

Thanks, another question I trust your answer on and I am working on putting together a blind test. There is so much on the web saying that flac and alac is worthless. Is flac and alac truly lossless audio or do you lose some fidelity with the conversion?

[@hansbeekhuyzen7770]

Derek Dastrup I have occasionally heard differences as well, but they were due to the hardware that was not able to decode properly. Furthermore one must ask himself that if he needs a blind test to tell the difference, that difference is significant enough and if it wouldn't be wiser just to enjoy the music. Doing a true double blind test takes at least three persons to execute the test and tens of test persons.

[@Catalonia]

What are your thoughts on recording? I.e. I can record as high as 192khz and 32bit float. Is it worth the extra tax on my system?

[@TheHansBeekhuyzenChannel]

+Evan Rowe That depends on the purpose of the recording, the equipment used, the budget available and so on. Hard to answer from here. I would record anything at the highest sampling rate and use 24 bit LPCM for that excludes no market. But that is six times more date that has to be stored......

[@Catalonia]

+The Hans Beekhuyzen Channel I would be recording layered music production, including mics, line in tracks, etc. I'm not sure what LPCM is as an option, but i'm using studio one. I can go as high as 192khz and 32 bit float (which, from what I understand is 24bits but with some extra room or something).

[@TheHansBeekhuyzenChannel]

LPCM stands for linear pulse code modulation which is the technical term used for digital recording as done for cd etc. If you want to record at such high sampling rates is something you should decide yourself based on the sound quality, demand from customers and so on. Apart from that, this subject is rather off topic here.. Sorry.

[@Catalonia]

Sorry, I guess what I was aiming at was sort of at the heart of the debate. On one side, there are those that say, recording above 24/44khz is unnecessary and does not provide added value. While others, presumably yourself, believe that there is perhaps value in it.

[@TheHansBeekhuyzenChannel]

There is no absolute truth and it is really off topic on this channel. You can see my point of view here: kzbin.info/www/bejne/qJDaqYV9nqeigMk

[@johnzimmermann6856]

Thank you for your excellent commentary.

[@TheHansBeekhuyzenChannel]

Glad you enjoyed it

[@MehranJ]

Very nice tutorial. Thank you. I recently bought a brand new Elac discovery (paid only $600 and with Roon essential installed I think it was a bargain). Up until now I 've been using my Bluesound Vault 2i as my ripper/music folder/streamer but would like to use it as a network drive for Elac so that I can use Roon's software. Is this possible and if so what's the best way to accomplish this over my internal network. Regards,

[@TheHansBeekhuyzenChannel]

In Roon goto settings -> Audio and enable the Vault2i. Now you can choose the Vault as endpoint. To have Roon 'see' the music on the Vault, goto settings -> storage and add the Vault by typing the network path. When entered, Roon will automatically start indexing the music and download the additional metadata.

[@buluthanna9745]

I was searching why we have 192 kHz even we can hear until 20kHz. And you explained it with timing. Thanks for it... Many things I couldn't understand but it's because of my English. I wanted to also ask: Did you put some maybe 20 kHz sounds behind this video? Because it sometimes beeps...

[@TheHansBeekhuyzenChannel]

to that I know of.

[@issafarhoud760]

Hi Hans - Excellent and informative broadcasts. My question is , I have many HI Res SACD, can I rip them into some kind of MQA to stream and archive in my house, thus reducing file size and bandwidth. I know I can archive as normal FLAc but interested in HI Res MQA type file

[@TheHansBeekhuyzenChannel]

There are two problems to what you want: first, ripping SACD's requires a special version of the Playstation 3 and archaic software. I can't help you there since I haven't investigated that (I found someone that ripped my SACD's). I have heard that ripping now is also possible using a Pioneer or Oppo player. If you google 'sacd ripping' you find more info. Encoding the result into MQA isn't possible either. First, SACD uses DSD for encoding. That should first be converted to PCM and if you would like to maintain the quality, you should use 24 bit 352.8 kHz, which will give huge files. But then you would need to get access to a facility that is licensed to encode MQA and they will not be willing to help you, not even if you would pay them. There is no consumer version of the MQA encoder.

[@marywhore1]

Hi, If you were to record a Musical Album, what would you record in, 16bit 24bit? In the end does it have to be converted to 16bit for audio cd? How do the pros do it? Thank you.

[@yepyep5247]

What is the maximum sample rate, bit depth, and bpc youtube allowes? Thanks in advance.

[@TheHansBeekhuyzenChannel]

I wouldn't know, but the audio on KZbin is always lossy compressed using, currently using MPEG4 audio as far as I know.

[@ludolupinski4724]

Explication super claire. Very clair explanation ! Many thanks

[@TheHansBeekhuyzenChannel]

Glad you liked it

[@maximilianogarzon1078]

Hello Mr Beekhuyzen !. I really enjoyed your videos :). I understand the superiority of the audio in 192Khz over 44.1 in terms of the steepness of the anti-aliassing filter, but regarding the resolution of time I don't think it's a valid argument to say that since the human resolution is 7 micro seconds, we need a sample equal to or less than that, I think it is not the case because no matter how much the human is capable of determining 7 micro seconds, when digital audio becomes sound, there is no longer an interval between sample and sample, has been reconstructed as a continuous stream, they are not separate frames when analogized. I have the feeling that there are scientific reasons to use higher sampling frequencies, but there are no subjective reasons to do it. so, what do you think? Greetings from Colombia !

[@TheHansBeekhuyzenChannel]

It's not about the time resolution as a result of the sampling frequency but about the time smearing caused by the reconstruction filter.