24 bits or 96 kHz? Which makes most difference?
Comments on this video
You can comment on this video at YouTube
You can comment on this video at YouTube
Thursday January 4, 2024
David Mellor
David Mellor is CEO and Course Director of Audio Masterclass. David has designed courses in audio education and training since 1986 and is the publisher and principal writer of Adventures In Audio.
@Manuel_Manzotti: If you have 2 identical audio projects, one in 48 kHz and one in 96 kHz, with a lot of plugins, effects, reverb, delay, distortion and so on, if you export a wav or mp3 file, whatever from both projects at the same sample rates and bits, the file exported from 96 kHz will sounds completely amazing than the other one from the project at 48 kHz. Just make the test and especially null test! It's night-day difference! The file exported from the project at 96 kHz has much more depth, clarity, better transient, space, anything just better. Of course your PC must handle a project in 96 kHz but that is other story.
Is like you shoot a video or photo in 4K then exported at 1080p and shot the same video or photo in 1080p and exported at 1080p. Of course the video or photo captured in 4K will be much better even you export at 1080p. The same is in audio.
@thepracticalaudiophile: For every person who convinces themselves they hear something they don't, there are people who convince themselves they don't hear things they actually do.
@edwarding4355: I wonder if my cat hears a bunch of distorted rubbish at the high end.
@edwarding4355 replies to @edwarding4355: We produce audio for human ears, but house cats are stuck with our audio.
@AudioMasterclass replies to @edwarding4355: Worry you it. Your human will food and pets.
@Captain_Mike: From what I understand there are some parts to consider in playback systems. If you have a DSP in the playback chain, it good to have a higher bit rate. For DAC chips e.g. ESS to perform better its good to have higher sampling rates.
@ZodiacVoodoo23: I went searching for this info because my soundcard runs at either 48k 24bit or 96k 16bit with the same latency and while the hive says 96k is better....I wonder...
@markc9871: People say the human ear can't hear above 20Khz anyway. That's PITCH not SAMPLING Frequency. So everything above 20khz is noise. That's the principle of the DAC. Push all the noise above the range of human hearing where it can't be heard, then filter it off without distortion. No ballocks.
@stephencharlesdawson: Just a reminder: some of the DACs most highly respected by some audiophiles operate as "NOS" DACs. That stands for non-oversampling. Ie: they use a precision resistor ladder the old-fashioned way. All that's well and good, but the ones I've tested also dispense with the "reconstruction filter", and tout that as a positive benefit. Look at their output on an oscilloscope and you see the stairsteps that the filter would remove. Look at their output on a wide-range spectrogram, and you see imaging way, way into the ultrasonic. Look at their frequency response, and you see the high frequency falling significantly from 10kHz and up. One example, the $10K+ Aqua La Scala MKII Optologic DAC.
@POEMS466: By Jove, why not record music that only dogs can hear? Because dogs are not paying customers. Humans are. And the CD was originally designed to cover the full range of the human ear.
@Welwynnick: You didn't answer the question! 🤨
@Raptorman0909: How about 24 bits AND 96 kHz...
@gibson2623: Hi, great Video as always. Here is my take on this: 96 kHz is hard on CPU and the files are the double size.....Thing is, in practice most plugins introduce a lot of aliasing at 48 kHz. It s easy to check with just a simple sine sweep, and there is the aliasing on the spectrogram even with the plugins at neutral state. Now, if I mix the tracks with the plugins at 96kHz, you know what s gonna happen to the folding back of the waves, they ll be further ahead in the frequency spectrum.....So I bounce the final mix back to 48 kHz for listening and finished project and I don t see a great deal of aliasing. I tested this with sine sweeps at the appropriate SR. I use R8brain in reaper for SRC, I could also use iZotope RX. I don t Dither anything also, since I keep everything at 24 bits. Would you care to criticise my setup? Is it overkill with no spetacular benefits? Thank you very much once again :)
@t.c.494: What kind of two bit analysis is this? I need a bit more.
@smanzoli: None of them, blind tests proved it.
@JLProB: is not at least twice the freq is twice , is a math problem not an inspirational issue
@Audiorevue: Well let me put it this way,
In high school I had two friends, one of them was named Willie bits and the other was named Richard Hertz, we called him dick.
As much as I love dick, he's not the man you want fighting beside you in a trench, you want Willie bits, he's the man to take you all the way.
@lawrencearmour5310: Not that I can hear it (either) but when the frequency range goes up to 20 kHz, you might ask yourself what you should be able to hear if you could. If the audio signal has anything more complicated than a sine wave at 20 kHz, then there will be harmonics and you won't be able to hear those anyway. At the top of the spectrum, you're mostly hearing the harmonics of the sounds whose root notes are much lower down.
@dizwell: Sample rate buys you higher frequencies that no-one can hear. Increased bit-depth buys you higher dynamic range, which might be discernible to some, though if the mastering has already participated in the loudness wars, that'll be irrelevant too. As far as discernible audio goes, Red Book CD is still just about as good as it ever needs to get. When I rip my SACDs, I convert to 24-bit, 44.1kHz. It's all I'll ever need.
@kenwilliams7597: Audiophiles can't hear the difference between 48 or 96 kHz. They are lying. 96khz is only best to use during production as EQ & effects processing is done at a higher resolution. This will eliminate possible artifacts rom processing. Once mixed, no more processing will be done. You will not hear the difference between 48 kHz or 96 kHz.
@ismiregalichkochdasjetztso3232: I'm using an LTC2208 ADC to sample signals at 16 bit and 130MHz sampling rate. Actually two of them, for coherent operation with two antennas.
It's in the receiver portion of a direct sampling software defined HAM radio. Nothing for Phil. 😂
@otto-r7h9l: But music, or sound in general, is not made by pure sinewaves. It is, actually, but it's the sum of "infinites" harmonics, each having different levels. With 48kHz, a sinewave of 20kHz have 3 sampling points, good enough for a pure sinewave, not good enough for real sounds. DAC will have to do some guess, and invent the sound. It does that quite good, but that's not the point.
I am not pretending to hear more than 18kHz, but I want the waveform to be sampled at higher frequency so to be able to reproduce the correct shape. In my opinion that's the reason some people likes analog audio more, just because it does not lose that details, even if it comes with higher noise.
Anyway, no one cares about CPU load, so having more sampling can only be better or, worst case, it will be same quality.
Just my 2 cents.
@createdflat5034: Ear syringing makes the most difference
@endrizo: digital filters in dacs cut off everything over 22 khz...so...digital audio is not hi res. ultrasonic harmonics are missing..no matter what bit depth or sample rate.
@analogue33: 24 bits makes such a difference with a 44.1k.😮
@AnimusInvidious: Sample Rate is certainly more important than Bit Rate for final delivery.
@BirdArvid: I am so very pleased that 16bit/44.1Khz is good enough for me. I watch my movies in my hometheater at 48Khz because that's what they're put on the disc at, but I don't believe that it makes much of a difference. I own two DVD Audio discs and I believe they both have stereo tracks which use 192Khz sampling rates, and they do sound nice, but not having anything to compare them to, I have no idea whether it's the sampling rate or the production itself which makes them sound great. Anything higher (and yes, I've tested!) is wasted on me. So my 2000 CD collection is staying put and being enjoyed! And I am absolutely certain that your listening room acoustics have a far greater influence on the sound you enjoy than the sampling rate/bit depth!
@morten1: I've used 24-bit and 44.1khz for years now. It's fine and I see no reason to switch to other/"better".
I think a lot of these technical hyper-detail perfection-hunting topics comes from boredom and a lack of focus on the MUSIC and being CREATIVE.
You know, the good ol' essentials. Music can also be "perfect" coming from an old kitchen radio
@killjoy4862: I prefer audio in 44.1/16. Depends also how is audio recorded.
I have a friend who have a band.
They record in their studio everything in 48kHz/24bit or 96kHz (I'm not sure tbh).
But I found one of their albums "pirated copy" on some website which says that they have the "master record" from studio in 192kHz/32bit.
As far as I know, audio format (flac/wave) can not be "upscaled" so to speak.
I'm I wrong?
Well, that website was an obvious LIE btw.
@synthetic_films: 48kHz came from the film/tv world to match 24fps since 24 is half of 48...
@synthetic_films replies to @synthetic_films: also, 88.2kHz had it's day...some of us have old recordings that are in this sample rate because we knew our delivery was CD and the conversion was easier and avoided interpolation. i have gone back and exported these masters at 88.2/24 to listen on my walkman and they do sound "better"
@synthetic_films replies to @synthetic_films: correct me if i'm wrong but the argument that higher sample rates rob your computer's ability to process audio is also sort of misleading. if you're set up for it computationally then you can do it and now-a-days, with any of the apple M1-M4 chips you can run multiple tracks in pro tools at high sample rates no issue. the reason to use high sample rates like 192kHz etc is for digital signal processing. your plug-ins can "stretch" and "mold" your sold files further without aliasing and without the sample falling apart. you can process more extremely if you need to and your sound will be cleaner overall. it give the plug-in more to work with.
@alanhosman8185: They are both indépendant parameters and of course both are effective individually
@gohan2091: If my MP3s / FLACS are 44.1Khz or 48Khz, and I set in Windows, 32bit / 96Khz, my DAC (Topping Centaurus) will be playing the audio at 96Khz sample rate. Is there any harm in this? ChatGPT says it's ok. LIike you said in the video about higher overhead and aliasing.
Why 88.2kHz or Higher?
1⃣ Reduces Aliasing in EQ Processing
Equalization (EQ) modifies frequency content, and if applied at lower sample rates (like 44.1kHz or 48kHz), it can cause aliasing—unwanted distortion that occurs when new frequencies are introduced but can't be properly represented due to sample rate limitations.
Using 88.2kHz or higher moves these unwanted artifacts further out of the audible range, making them less noticeable.
2⃣ Better Interpolation Accuracy for Parametric EQs
In PeaceGUI / Equalizer APO, parametric EQ applies curved frequency response adjustments that require accurate interpolation.
Higher sample rates (88.2kHz, 96kHz, etc.) allow more precise frequency adjustments because they have more data points per second to work with.
This results in smoother EQ curves and less phase distortion.
3⃣ Prevents High-Frequency Roll-Off from EQ Adjustments
Some EQ algorithms cause high-frequency loss when working at lower sample rates.
If the sample rate is too low, the EQ might struggle to accurately adjust high frequencies (above 15kHz), making the sound duller.
Setting the sample rate to 88.2kHz or 96kHz ensures the entire frequency spectrum is handled correctly.
4⃣ Avoids Harmonic Distortion When Boosting High Frequencies
When you boost frequencies near the Nyquist limit (half the sample rate), aliasing distortion can occur.
For example:
At 44.1kHz, the Nyquist limit is 22.05kHz.
At 88.2kHz, the Nyquist limit is 44.1kHz (giving much more headroom).
Boosting treble at 44.1kHz can introduce distortion, but at 88.2kHz or higher, there's more room for smooth EQ adjustments.
@genesmith4019: I got the highest resolution DAC and now I can hear farts from the control room. I’m looking into some “Golden Mean” cables too so I can smell them as well!
@sjors01: Considering the average age of audiophiles is around 50 and up. 20-20khz, hearing is deteriorating more and more. We may think that it is not so.. we older people can afford this expensive hobby because of our career and the time we have left on earth. Very nice hobby. if you look at the content of electronics, there is no more than € 100 in parts, why does it have to be so expensive in mass production is the question.. High-end nice same story, there is no more than € 100, - but whether you really hear it. of course there are differences in sound, because every factory has its own preference for sound. that is why I take older reviewers their well-intentioned version to heart, with the back of my mind that I also encounter that your hearing is deteriorating more and more. so Phil you are absolutely right high RES is cd quality for little to get both new and second hand for € 1, -. we do not need streaming at all, but flac while .wav gives the best result. only a pity that no metadata can be entered.
@audiononsense1611: I'll just say that done correctly PDM is superior as I've never experienced anyone who has heard it on my system or at the dealers I've worked with say otherwise. Most believe it's analog....
@SimonP2: No mention of harmonics?
@deetgeluid: Keep in mind that frequencies above 20k do produce audible harmonics below 20k.
@deetgeluid: I believe 44.1k was used to be able to integrate anti-aliasing filters. They need to be analogue so can’t be brickwall. So Philips used 44.1 to have the extended frequency range for the slope of the filter to reach the point where aliasing is not heard anymore below 20kHz. Okay, I need a flipover or whiteboard to explain. Sorry!❤😊 Also, look up Nyquist theorem on why we use 40k instead of 20k.
@AudioMasterclass replies to @deetgeluid: Oh shit. I've been working in audio for more than 40 years and somehow I forgot to learn Nyquist. As for why 44.1 was chosen, I think you need to look up that.
@deetgeluid replies to @deetgeluid: @ Sorry, looking uo Nyquist was not meant as a personal comment, but for people who might be interested and never heard of it. I’m Dutch, maybe something was lost in translation. You sound a bit angry, I’m sorry if I offended you in some way. 😔❤️
@petrofski88: I never understood much about "these things", but love to learn. My headphone amp shows freq. rates up to 11.2M but is always on 48.0K and I never knew what that was, now I get it - I think. What I wonder is... if 96.0K is already going beyond great, and to many/most sounding just the same, why push something like Quad-rate DSD? It seems that it just creates more information noise.
@a64738: CD`s sound horribly distorted over 15khz, it is just that most people is deaf over 15khz so they can not hear the horrible distorted sound CD have above 15khz (over 15khz it is less and les sound and more and more just white noise the closer you get to the 22khz max theoretical sound (not really sound but noise) it can make. The 22khz max sound freqency only apply to a pure sine curve, anything more complex need higher sampling frequency then 22khz to be reproduced as they contain sound components / information that need higher sampling to be reproduced because music is complex signals and not sine waves...)¨'¨. In short 22khz is to low sampling frequency to reproduse MUSIC because music contain information (timing) and frequency components that even if they are outside your hearing frequency it still makes huge impact of HOW the sound sounds like and introduce horrible distortion when they are filtered out because of to low sampling.
@a64738: Use a 24 bit upsampler, it makes everything sound identical to 24 bit (I can not hear difference from original 24 bit recording and a 16 bit recording run trhough the upsampler in blknd test, but I can hear if it is 24 or 16 bit immediately because the difference is clearl and present).
@alexschwarzmeier7061: You can't hear the difference between a CD spoiled by the loudness war and 24bit studio quality? Are you kidding me?
@djdrummernick: 384 32 bit 8000 sasmples amzing playback in asio on vdj with ifi black label to mixer with lexi mx300 and bbe
@Vanflowne: If you like the high frequency noise of all that Hi Res crap you like to buy 24 bits 192kHz. The excellence of audiophile stupidity is called INTERMODULATION DISTORTION
@driz77: What makes the most audible difference is the objective distortion at low-but-perceptual levels (down to, around, -60dBFS). This is why multi-path architecture excels over any legacy single-path design.
@molotulo8808: 3i just subscribed. You are a fine young man who seems wiser than their age.
@AudioMasterclass replies to @molotulo8808: Actually I'm hoping for my wisdom to catch up with my age. It's taking a long time.
@giventhyjimothy replies to @molotulo8808: @AudioMasterclass +2
@brucerosner3547: I can hear dynamic range; I can't hear high frequencies.
@KLiNoTweet: The moore the better, right 😊
@nine27: We should not free ourselves from CD 💿 being that so many independent artists are pressing up and selling them again.
@assaf3468: So what I need to use for music on windows? I have 660s2 and fiiok7
@kevinmills5293: I had an SACD player and the sound stage it could throw out was stunning. No CD I had could do the same. Many years ago when single bit DACs started to hit the market I was on the hunt for a new CD player and it was easy to identify the single bit machines by the scale of the presentation. Don’t really know why but the high bit rate machines just seemed better.
@SteveInDub: As a producer I work with a lot of legacy 44.1kHz 16 bit source material (CDs).
My DAW is set to work with a resolution of 24bits, but I tend to stick with a project sampling rate of 44.1 kHz, beholden to the idea that converting from 44.1 to 48 might create artifacts in the audio, and not changing the sampling rate is best practice.
I suspect I might be dwelling in the stone age here, but I'd like to try and catch up!
@brians7094: I have heard great quality recordings on CD that seem to have most all of the music. I don't consider 24/48 to be high resolution, but now I know why it gets that label. I find that 192 kHz is generally more than twice as good as 96, assuming a great recording, like Wes Montgomery's first album. And I don't buy this application of Nyquist theorem to audio. Look at a sine wave on a digital scope, and you can clearly see that it would take about 10 times the sampling rate to accurately reproduce it. Yeah, to me, sampling rate is much more important than bit-depth.
@brians7094 replies to @brians7094: @ViklasSvenske Thanks. I watched that. But I still hear a difference between 44 & 192 kHz sampling rates with a good NOS discrete resistor DAC. Of course, these differences cannot be heard on less resolving systems.
@ALed-j7u: I still haven't come across a 96k recording that is not 24 bits also.
@AudioMasterclass replies to @ALed-j7u: Yah. As a producer, 24 has practical benefits. 96 has costs.
@frankd.b.9233: CD 24 Bits 44100 Hz, DVD 24 BIits 48000 Hz all studiokwal. pc sound = + like 192000 Hz
@JAO53JAO: I have recorded with 24 bits at 96Khz for years, not because I can hear a difference in sound, but because I believe that I get greater headroom and much less aliasing with this choice. And as I typically work with about 100 tracks in a song, reducing the problems created by lack of headroom and distortion from aliasing seemed like a good idea, even if there seem small. It costs more to be able to do this, but I do it anyway. Maybe I am wasting money and effort ?? Thanks!
@JAO53JAO replies to @JAO53JAO: You are right on the money. This is not about hearing better sounding music - our ears and brain cannot perceive and process the music in a way that would find much (if any) difference. It is important to note that most commercial formats you will use, will not accept 96k file submissions and do not broadcast them, so you will usually not be able to play them in your car, on your phone, etc. However, working in the industry, you may find it useful when mixing and mastering to have the higher headroom files and aliasing is a real thing and it can be audible. Sometimes quality is not about what you hear, but about what you do not hear.
@joesalyers: High sampling rates for playback are insane 48K is the accepted norm for all platforms and is a happy medium but even 44.1 is fine for playback in the final product. But for music production using high sampling rates like 96K is a bulwark against aliasing distortions. Its not about the bandwidth and how much of that which you can hear. But high sampling rates such as 96K pushing aliasing distortion out of the hearing range that would otherwise creep down to the audible range at 441 or 48K. It helps when using digital processing like saturation, and anything that adds harmonic content to the signal, it also gives some EQs the ability to not cramp. I heard a professor from ETSU's music school describe sampling rates in a very interesting way once. I wish I could animate it because its a great visual. He said imagine standing in front of a 10,000 feet tall concrete wall with a concrete floor and you are standing there with a tennis ball (aliasing distortion) in your hand. Imagine standing 21.5 feet away (44.1K) from the wall and throw the tennis ball as hard as you can and where ever it lands that is where aliasing would be heard in the hearing range. Now move back to 48 feet away (96K) and perform the same throw. Most likely at 21.5 feet the ball went over your head and landed behind you somewhere, but at 48 feet the ball is less likely to even make it back to the spot where you are standing. I thought this was a really great visual for describing aliasing.
@joesalyers replies to @joesalyers: @ViklasSvenske Oversampling in converters can help give the filter a more rounded shape, but the fact remains there is more to a session than just recording the audio through the converter. the antialiasing filter is still in the same place. So no matter how many times the chip will oversample x128 or x512 the filter at 48K will still be at 24 kilohertz on the output leaving the antialiasing filter in the same place same place but with a more favorable shape. Oversampling isn't a replacement for using better LP filter placement. The type of filter used as well is highly important. Lavry explained this in the sampling rate white paper that was published nearly 20 years ago when they were building their next generation conversion chips and they explain why converters with oversampling can mask many of the failures of digital conversion filters. Its one of those deep audio subjects you can't really have a good conversation about in the Youtube comments section. But saying there shouldn't be aliasing at any sampling rate for converters is true but we don't just record audio and thats it. We also use digital tools and effects that will work better at higher sampling rates. Time stretching, Tuning, Distortion, Saturation and Pitch algorithms all benefit from the highest sampling rate possible because this gives them more data points to work with. But in the end if it sounds good it is good so thats the great part about choice if a person wants to work a lower sampling rate they can make music at 44.1K and if other people want to work at 192K they can. I work at 96K because its the best spot for me but thats the great thing we can work how ever we choose. Cheers, Have a great day!
@MetalHead123345: And I don't care if you're on Amazon Music or anywhere else and yes, I have Amazon Music unlimited, you don't see a 192k that much you don't see 92K that much, but you see it a lot more than a 192I have. Seen the 88 thing or whatever. But usually you see 44.1. 2 48 16 or 24 bit. Seems. It can go either way, and sometimes the lower song sound, just as good h. D and ultrahd a lot of times sounds the same.
And yes, I've had the attack for my phone that runs all the frequencies. I also have a very high-end yamahal receiver in my home. I have equipment and I don't can hear s*** if it's there. And I've been doing music a very, very long time, and I grew up around a lot of bands. I am not influenced by a bunch of hearsay in numbers. I am big on the try it yourself. Because I keep finding out a lot of s*** on YouTube's a lie. A lot of people's opinions are c*** For something that's a fact you can't hear this. There's no way. Yeah, I'm a firm believer. And there are limitations and I don't think people hear everything they think they hear.It's called placebo effect because someone told them about it before they listened to it.Therefore, their brain perceived it, even though it was not there
@MetalHead123345 replies to @MetalHead123345: And I'm using my microphone and I apologize for misspelling or anything that's confusing. I am sorry for that the.M too lazy to edit incorrect.This right now
@GerhardBothaWFF: Anything over 40khz enables one to perfectly reconstruct signals below 20kHz. The problem is impact type sounds ( like percussion) , for one, exceed that 20khz bandwidth and even though you cant hear a 20khz sine wave, you can hear the sharp rising waveform of a drum thump as well as “distortion” on real instruments. Then also, the way the signal is reconstructed- rarely do the compute inverse laplace transforms etc - they just interpolate. Hence the divergence of opinions
@2232-q1l: In binary, 44.100 is a round number.
@FBAV: 96 khz barely makes sense
@johnnewcomb5162: 24 bit sounds the best on my $19.99 Sony headphones.
I have noticed that.
@charley2070: I think the whole discussion about sampling rate and audio quality is kind of pointless. Maybe we should Analyse the playback situation. What people listen to your music on what devices? If it is AirPod or a club mono setup you don’t need to worry about 44,1 or 48khz.
@shadowside8433: I love 1Khz as a tone. Smashing.
@nickmaddalena985: For me, it is probably the quality of the sound engineering.
If it is recorded in 24 bit 44 or above, it means they more than likely care and it has gone through a more modern process and mastering.
The difference in mastering quality is massive between, say, an 80s song and a song in 2020s.
A lot of the cd era 90s pop was mastered so poorly.
The cds sound metallic and harsh, unlike the older tape process.
The tape process was warm but compressed.
As an audiophile, it is literary hard to listen to older cd mastered music anymore, most likely due to the bad mastering processes used.
If you find them in 24 stream or flac and it is remastered, it sounds very different, not because of the bits but the process of mastering with modern day tools.
Would love to see a video on this... ❤️
@StupidityCensorV2: One word… Transients. Music is not a steady state phenomenon. Treating music as purely a function of frequency and amplitude completely ignores the transient attack of a note or sound.
@fabianoberlim9636: We all can perceive above 20khz
There’s a new reasearch that shows activity in the brain. The activity is where the brain perceives localization
The test was conducted with 20khz to 100khz and also most of the 66 participants preferred the music with ultrasonic content
@fabianoberlim9636 replies to @fabianoberlim9636: Passion for Sound made a video about the research
https://youtu.be/q_HfTgN1kB4?si=mix6t-F2HRfEy_SU
@Tortuosit: 24 vs. 16 bps guarantees 2-3 times as big FLAC files. What I often do is bringing the bits per sample down to 20-22 (so there's zeroing the LSB, technically still 24 bps), via dithering. So a lot of my audio ends up as 48/20.
Storage is cheap, sure, but with high sample rates and high bps space goes out of hand.
Converting FLAC source to target devices, I always go for 48/24 (even upsample 44.1/16 with SoX), because most devices handle it natively - no resampling on the fly.
@David-p6e: in my setup i run 192k 32bit umc , and i use a anlouge connection from tv , and stream dolby atmos codec to the inputs , the sound is beyond everything ive heard so far , now im gonna test even higher res and dsd when get this soundevice , in my opinion the higher sampling the better sound since my vst plugins rerender the 48k stream and oversampled it sound ALOT better than the original
@David-p6e replies to @David-p6e: and i use solfeggio scale aswell
@pietermol8508: Unfortunately, I don't have the time to write down the complicated rationales why 96kHz - IN VERY VERY FEW CASES - might be a better option than 44.1 or 48 kHz, so let me suffice to quote Nobel Pize winner Georg von Bekesy, the guy who basically was responsible for the still universally believed idea that the human hearing apparatus works as a Fourier-analysis machine, and that human hearing is limited to 20kHz (which is true, but lacks nuance):
"In time, I came to the conclusion that the dehydrated cats and the application of Fourier analysis to hearing problems became more and more a handicap for research in hearing."
If you start to investigate why Von Bekesy said this, you will start to uderstand that most of our understanding about how human hearing works is a social construct (look that one up on wikipedia). Here's a clue: not all waves are harmonic. You might object and say that by means of Fourier analysis non-harmonic waves can be broken down into a number of harmonic waves, but then: read the quote by Von Bekesy!
Ultimately, it all comes down to transient response. Cutting off audio at 20kHz affects transient response to a level that affects how we interpret sound (not just music, but all sound). If we capture at double the sampling frequency, i.e. 96kHz, than the problems of transient response are solved. But capturing at 96kHz also requires everything else in the recording chain to be geared towards capturing those transients, that in the practice of real life music production there is no justification for 96kHz sampling.
A final note on different sampling ferquencies. I always have a lot of fun asking 96/192 kHz advocates what sampling frequencies their music software, operating system and sound card are configured to. Not a single time I have met a person that got beyond the level of sheer ignorance.
@coreymoyers5771: I came upon this video looking for a way to explain a DAC to my daughter, who has fallen in love with high-end audio for the first time. I think I may have found the best example of nonsense on the internet.
It is easier to mock what you don't understand than take the time to understand it. You have simplified everything by assuming your explanation functions in a perfect world when it isn't. What happens to the sound when your 96kHz drifts to 95kHz one second and jumps to 97kHz the next? What happens when your DAC wants to hit 16 bits of dynamic range, but the current draw from the bass hit lowers the processor's 5V down to 4.95V?
You've tried to simplify a complex process without understanding how electronics operate. What's sad is that you and everyone else can clearly hear the difference when you listen to setups you would claim are snake oil. I pray you take the time to listen to a system with zero jitter and a clean signal on speakers that can start and stop instantly. Of course, you are already convinced you are correct, so you won't hear any difference. It's your loss.
For those who hear a difference but don't understand why, it's because no system is perfect. If it were, then just enough would be plenty. The slightest variation in sound changes the soundstage. Instead of sound coming from two boxes, it comes out of nowhere, yet you can point to it. Drums in the rear, guitar over there, sax in the corner. Have you ever heard a sound come from beyond a point past the outside of a speaker? How is that possible? It is either snake oil, or you raised the quality to a point where the mistakes are more diminutive.
In HiFI, timing is everything. Two DACs with the same chip will sound superior with a more significant dedicated power source and femto clock. It is the same math but with more precision. That is why Wilson Audio aligns its drivers like it does. Sure, 16 bits at 44 kHz is all we need, but the timing will be less precise, and the sound will be less enjoyable. Don't take my word for it; go listen for yourself.
@AudioMasterclass replies to @coreymoyers5771: Hey, I've made the best example of nonsense on the internet apparently! Is this your first time online?
@KMASCII: Just an FYI Audio Masterclass, musical instruments (from the human voice to stringed to wind to percussion etc) create a complex sound made of a fundamental tone along with various overtones, undertones, harmonics, etc. A digital representation of it would be akin to a spectra of frequencies. Similar to the this image. https://drive.google.com/file/d/1EYubbXHrYpqZ3mk-AtDvPbXZRnw6eksU/view?usp=drive_link. Suffice it to say it would be represented by something more complex that a single tone as so many like to imply. Sound is very complex.
When a microphone (transducer) captures that complex sound-wave it converts is to an electrical signal using its coil and magnet. The frequencies are captured as varying frequencies in a variable electrical signal. The volume, or amplitude, of the audio is handled the varying voltages the mic (transducer) creates. The electrical signal has frequency and amplitude modulation. Kind of like this image: https://drive.google.com/file/d/1JzvxprL4fKpju2ji8HuJFh14mxj2ZgSH/view?usp=drive_link.
When the audio signal (again, which is actually an electrical signal) gets converted to digital, what the digital converter does is it captures a snapshot of the electrical signal at specific points in time at a specific rate (the frequency of sampling). The "capture" is of the electrical signal and not the actually audio or sound-waves as many people believe. The electrical signal is captured at the specific sample rate (frequency) and bit depth (resolution). See this image: https://drive.google.com/file/d/1PVEtxcY2Z9_-F47j1qyDlW3w1VmdnQQP/view?usp=drive_link. This image is just a rough example of the rate of 44,100 samples per second (which is what 44.1 Kilohertz is) and the bit depth (a depth of 16 bits will allow for a resolution of up to 65,536 points of amplitude to measure) of the electrical signal.
Your explanation was great Audio Masterclass, but I do worry about always hearing how sound is visualized as what is being digitized when what is actually being digitized is the much simpler electrical signal. Feel free to correct me on anything, I'd appreciate the correction.
EDIT: I neglected to note that the capturing image/chart was intended to reflect a line signal capture. Hence the higher voltage than a mic's signal, and apologies for the long post.
@rajendrabareto8065: 16 bit input multiply it by 16-bit filter coefficients will give 32 bits. Is that what is required 😂. The amplitude should be within the dynamic range , hence there is no clipping or saturation to be exact. Low amplitude on other hand will be collected better in the higher order bits. But that also will include thermal noise. Now noise can be filtered out though to an extent if it is above the desired frequency range. Now the question is is the amplitude of some of the instrument sounds is too low that more than 16 bits are required? One can also normalize the individual instruments mic input and record so that they are well within the adc dynamic range. So it is not only the sac that is important. It is very much the adc and analog inputs while recording.
@veloxime: Please, a video on Cd vs. SACD, in understandable language
@InfamousInternetVillainJackSix: I really need you to get over yourself and stop being so pretentious. It sounds like you have good information, it's just shrouded by all of your narrow-minded assumptions, about your perceived viewers. Yet another dead-end video, where the content creator makes the false assumption that human hearing is applicable to everyone wanting to learn about this subject. I think you should drop the "C" and the "L" in "Masterclass", so that your channel name will better represent your content. It's ironic that you spend half your video talking down to audiophiles while simultaneously behaving just like one. Those pricks irritate me just as much as this video does.
@elephantaviator8342: my tinnitis having ass doesn't need anything higher than 26khz.
@ziofrenko: Science says we perceive above 20khz, and I can personally confirm this. There are linear microphones up to 100khz that are used both in the studio and live, There are tweeters that easily reach 40/50 kHz and obviously a dac working at 96khz had low pass at about 45khz audio... But above all: a crash cymbals had 40% of energy above 20khz, a rimshot 6%, a trumpet 2% that our perceptive system feels and is used to hearing live (obviously if it frequents musicians).
After saying this, listen to R.A.M. at 88.2 and then at 44.1 and tell me if you don't hear differences, they are macroscopic!!!
Not to mention mixing at 48 vs 96: it is precisely in digital mixing that aliasing acoustically invades the music because it is when you manipulate the sound that aliasing is created! And in fact all the best plugins include the oversample function to limit aliasing because it is when mixing that it comes out: if you record at 96 and without manipulating you pass it to 48 there will be no aliasing!!!!!
@ziofrenko replies to @ziofrenko: @nicksterj science prove it, this is the last but with other methods we have known this for at least 40 years.
Good studying...
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5285336
@ziofrenko replies to @ziofrenko: @nicksterj science prove it, this is the last but with other methods we have known this for at least 40 years.
Good studying...
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5285336
@ziofrenko replies to @ziofrenko: @nicksterj science prove it, this is the last but with other methods we have known this for at least 40 years.
Can give you a link in private because Delete comment with links.
@ziofrenko replies to @ziofrenko: @nicksterj science prove it, this is the last but with other methods we have known this for at least 40 years.
Can share a link in DM... Links block comments.
@nwimpney: Why waste space with unnecessary bit depth or sample rates. It makes sense to do mixing and processing at higher resolutions, but why waste disk space recording things you can’t hear, and which may even cause audible artifacts in some cases.
Almost no audio gear has a sufficiently quiet noise floor for the extra bits to matter, so bit depth is kind of pointless, too.
If you’re processing the audio, work in high resolutions to avoid quantization errors accumulating, but you may as well resample to 44.1 or 48k at 16 bits for distribution.
@Xogroroth666: 8 Hz for me.
And YES, I know people generally can only hear to 20 Hz.
However, I can hear 8 down to 7 Hz.
This was tested in a local university.
@Xogroroth666 replies to @Xogroroth666: 24bits, 192'000Hz, this is what my machine puts out.
Or, so it tells me. :S
@AudioMasterclass replies to @Xogroroth666: It's rumoured that a frequency of 7 Hz, if loud enough, will scramble your brain and other internal organs.
@Xogroroth666 replies to @Xogroroth666: @AudioMasterclass
It is also rumored that there are aliens among us, sir.
And that 5G created SARS-CoV2.
Or, that the sound of electro-windmills create cancer (ask Trump).
High frequencies can boil a brain, or burn skin.
The army experimented with such apparatus, like the Crowd-Control system.
Low frequencies can mess with hard materials, as seen in Earthquakes, when buildings fall apart due to resonance.
Which is why earthquakes create a rumbling sound.
Instead of listening to people, try and find scientific proof.
Never believe people, not even me, double check what you hear.
Then triple check it again.
Hope this helps?
@Xogroroth666 replies to @Xogroroth666: Also:
All sound, "loud enough", will cause severe pain.
1KHz will do plenty pain if loud enough. :)
So, "loud enough" is no excuse, is it, sir?
@Xogroroth666 replies to @Xogroroth666: @nicksterj
Not "headphones", sir.
These were special "earplugs", though I do not know how they work.
There was no brand on them, either.
Hope this helps?
@JamesBrown-ny2dw: Sampling rate is very tricky. The final quality depends on the DAC, which might reconstruct analog signals differently based on the sampling rate. Who knows how many tricks the engineers added to the DAC algorithm lol. I don't think there is a definite answer to this.
@JamesBrown-ny2dw replies to @JamesBrown-ny2dw: @nicksterj I had a fiio k9,500$, that sounds differently for different sampling rate settings. I'm now using a gustard r26, 1500$, the difference is much smaller, but still audible for some sampling rates. My conclusion is don't expect anything close to perfection for the dac is below 1000.
@JamesBrown-ny2dw replies to @JamesBrown-ny2dw: @nicksterj The two I mentioned are of good value. I ordered them because of that. Still some are more sensitive to stuff like these.
@JamesBrown-ny2dw replies to @JamesBrown-ny2dw: @nicksterj its more like its interaction with the PC upsampling setting. I guess it has a target upsampling rate and if the PC does more, it will do less. And this change will cause some difference in sound, which is reasonable tbh.
@mabehall7667: Considering the color of your hair, 22Khz is probably sufficient.😅
@jarosawmanicki3549: Both! When I convert my vinyl collection to a digital (wave) format, 24 bits and 96 kHz sampling allow me to fully preserve their beautiful analog sound!
@djangofett4879 replies to @jarosawmanicki3549: you're taking an album that was most likely recorded pr mastered digitally (or both) then transferred to vinyl and now you're transferring it back to digital.... seems pretty pointless.
vinyl collectors really like sipping the snake oil
@uglydoor1: “Audiophiles don’t use their gear to listen to your music, they use your music to listen to their gear“ Frank Zappa I think said this
@uglydoor1: I’ve heard of a panel test where listeners got the sample rates wrong half the time in a blind test. Again this is hearsay if anyone has the reference that would be useful for this discussion.
@uglydoor1: 48 is best if you’re doing some Wawa guitars for a new porno because Video runs at 48 and you sidestep the SRC which
also negates the advantage of higher sample rates in general because they need to be SRC at the back end anyway canceling out any imaginary improvements in “Fidelity”
@uglydoor1: Totally on board. I am a Audio engineering professor emeritus and a lifelong engineer and producer. Here’s my way of dealing with the elitists. Of course you can hear the difference with the higher sample rate. Let’s do the electrode test!(attached to your bullocks that is)
we’re gonna blindfold you and play you something at 44.1 also again but a higher sample rate..
and of course you will be able to tell which and your pride will be intact.
However, if you get it wrong, we will give you the electricity. what’s that??
There areeeee times where it’s hard to tell the difference? you don’t say.
Now take your Ludacris, hard drive clogging sample rates and be on your way you knucklehead.
@rickmilam413: Interesting. I thought the 44.1 was a result of the Nyquist theorem. 48KHz came from the DVD format. That's all there is to it. 49K is NOT hires. Unless your DAC has a 48K clock as well as a 44.1 (most do these days) it's detrimental. The streaming service I use calls 48.44.1 hires. It isn't, not to me. 20 bit is about all we can hear or our systems can deliver. Higher sampling rate does sound better to me, especially on classical, where low level detail and ambience is lurking in the mix. Digital has come a long ways. I could barely stand to listen to it in the mid 80's. I still listen to a lot of vinyl, mostly classical, recorded analog. But some 16/44 fare is damn good. Often I do find that digital releases of older analog classical fare sounds a bit... sterile? Ragged? Something about where the sound drops into the noise floor just doesn't seem right. But some sound great. So many variables...
@AudioMasterclass replies to @rickmilam413: 44.1 is from U-matic, which you can look up. 48 comes from broadcasting prior to DVD, which you can also look up. As for your comment, "That's all there is to it". Debatable. Which is why we have YouTube.
@4050Sixty: Ive spent a few years trying to figure this out,
I work at 24 bit 44.1, would there actually be an advantage to bump up to 32 bit 88.2?
Would bouncing it back down to 24 bit 44.1 after end up introducing more issues?
When I ask this question I seem to get answers all over the place
If someone with a deep knowledge of this could explain this to me it would be greatly appreciated
@johnchildress8707: I recently purchased the S.M.S.L. su-9 pro, because I wanted to be able to decode DSD to DSD512, and 24bit/192KHz for playback with my nearfield listening system. Trying not to be the audiophile that Alan Parsons refers to as listening to the equipment, and not the music, the main difference I think I hear, is not audio quality but time correction, and that probably has nothing to do with the sampling rate at all. My playback is though a 1975 Pioneer SX434 into Wharfdale Diamond 9.1's which probably produces the quality of sound somewhat equivalent to my aged hearing
@doktabob328: Only extremely high sampling rates can preserve the detail of very high frequencies.
A frequency which is the Nyquist frequency (half the sample rate) is represented by only two samples. A square wave. There is no way around this.
There are interactions of high frequencies which affect the overall signal.
It may still ‘sound nice’, but the fact is that typical sampling rates butcher the harmonics.
I am not such a purist that I can’t enjoy music at CD quality.
However, this fact of butchering high frequencies (even if they are smoothed by intentional or unintentional filtering) is simply a fact to which many ‘experts’ seem oblivious, and many even claim is untrue.
@bartnettle: DSD is 1 bit but it is 1 bit difference from previous bit. It is linear in that regard. So too is analogue linear. It is ideal for Mastering to and as an archival medium and for the audiophile to have exactly what came off the console; it doesnt get any better. It is a different method but has not caught on and is goin the way of the betamax.
@MikeleKonstantyFiedorowiczIV: i dont care what this guy say, it should be standard many years ago, now we should care about 32 bit music
@MaxCarola: Thank you for the ever exilarating but serious presentation. I use, if I can choose, 48KHz sampling rate and 32bits floating point resolution (mostly to avoid internal clipping). And I export my masters in 24bit 48KHz AND 16bit 44.1 properly dithered. I have tried 96 KHz and I DON'T like it unless it is for Jazz or Classical recordings. For Pop and Rock I feel is losing some grit (It might be some sort of psich-acoustic impression on me knowing the sampling rate. In general I am fully satisfied with 48KHz and if I can use oversampling in som plugins for extra resolution in some cases I am fine. The most important fact is that if it sounds good (to me) then IT IS good and that's it.. And if in 30 years from now they are still reprinting my recordings, then, good luck. I don't think any sample rate difference will make a difference. But who knows... When I built my first home studio around an 3340S Teac in 1975 I was dreaming to have a way to have a full 24 track studio in my bedroom. And in 1992 I was dreaming about a software I named "de-blender" that was capable of separate the individual instruments from a store recording... And here we are... both are a reality now. How impressive. The next step is a software that mix the tracks with A.I. (hahah... ) But I hope not as good and as artful as myself. Thank you for your contribution to the demistification of sound engineering.
@acecomet: The information in this video is wrong…
@AudioMasterclass replies to @acecomet: The information in @acecomet's comment is wrong.
@DwightPotvin: I do not do recording, but I can hear a clear difference between 16bit and 24bit and between 48khz and 96khz. I prefer 96khz. It's just more clean sounding.
@memcdm: Sadly, the vast majority of reviews are of little value to the potential buyer and in some cases, are a disservice. In all cases reviews cannot determine the long term reliability of a product. Reviews of ultra high end products are the worst. The law of dimishing returns sets in early these days. Many modestly priced audio pieces sound very good. Paying thousands of dollars more often gains the buyer very little in actual listening enjoyment beyond praging rights. In some cases, for example, very large and expensive speakers, the sound will be worse because they overload the room in which they are installed. Hugh powe amps are often a waste of money unless you have a near impossible to drive speaker and play at levels that destroy one's hearing. There will always be folks who want to buy exotic, insanely expensive audio products just like folks who buy exotic cars whose capabilities cannot be utilized on public highways without putting others at risk. That's fine but the sound may not be better and in some worse than a less expensive system with well designed and built products. One final comment. If a system that costs in excess of say... $10,000 or $100,000 , and it sounds very good I'm not impressed. If a system sounds very good at say . . . . $1,000 to $2,000 .... I'M IMPRESSED! I've heard a couple of systems at around $5,000 (very modest in today's audiophile world) set up properly that sounded amazing and could play over 100 db with very low distortion and by the way, looked wonderful as well. So pay your money and make your choice.
@themastersofreel2reel321: 96 FLAC is your limit and its does sound good from HD tracks i have Tori Amos albums are so clean
With it in my opinion
@thomaslutro5560: Why not just 42kHz? I wouldn't panic about that last halftone or so.
@themattprofessor: Well the argument for using higher sampling rates has to include aliasing distortion within plugins, which means realistically 384Khz and above is needed unless all the plugins you use, use over sampling (obviously potential issues with that, apart from processor power. So although even in heavy saturation plugins those aliasing distortion products are quite a number of dB’s down, if you use channel and tape modelling plugins, they will all add up, and although not necessarily sound bad, make somewhat of a mockery of all this analogue modelled claims, so there in lies the problem, all this under the hood stuff needs to remain under the hood for most people. Oversample plugins as a default, 48Khz sample rate is more than fine for a recording rate. I could argue 32bit float recording format might be a good idea, given the fact that whenever I get sent mixes to master or stems to mix undoubtedly stuff has been over recorded!
@AudioMasterclass replies to @themattprofessor: Your comments are related to comments I have received from producers who have said that higher sampling rates are better for time-stretch effects. I have yet to confirm this but it's on my list.
@j7ndominica051: The "algorithm" governs us. It's not exactly easy to browe the site to find something specific. Few search results, slow JavaScript. Hmm, American video runs at 30000/1001 fps unless it is old black and white, so that yields 44.056 kHz. 96 kHz would also buy you half a bit of "vertical" resolution, and another 3 dB for every doubling.
@meredithharvan5632: all wanted to know is what is the best digital sampling rate and bit depth that will give me a digital platform that sounds as good as the (horribly flawed) vinyl records I play
@AudioMasterclass replies to @meredithharvan5632: In theory, something around 60 kHz would be necessary to match the HF possible from a half-speed master. As for signal-to-noise I reckon 10 bits would easily do it.
@meredithharvan5632 replies to @meredithharvan5632: @AudioMasterclass Thanks! I will look for that when shopping for a DAC/Streamer
@AudioMasterclass replies to @meredithharvan5632: @meredithharvan5632 That was the technically correct answer. In practical terms 44.1 kHz / 16-bit, CD quality, would be perfectly ok for most people. Anything better than that would give you a safety margin of quality.