What Are Audio Bit Depth & Sample Rates, and What do the Numbers Actually Mean?

[@EdThorne]

💽 Track Mixing: edthorne.com/mixing 🎬 Can A DI Box REALLY Make A Difference? kzbin.info/www/bejne/fpaoZmqbjr2dfJo 🎬 Lewitt CONNECT 6 Audio Interface with DSP kzbin.info/www/bejne/qYa4fX9pfpWla7s 🎬 Re-Amping & Analog Mixing with the Volt 476P Audio Interface kzbin.info/www/bejne/e5nIlIujmN6FZ6M

[@BrodieTV]

Ed, love the informative video! Background music is a little loud though so it might make it hard to focus on the video for some viewers. Either way love your stuff my man keep it up!

[@EdThorne]

@@BrodieTV Thanks for watching and the feedback. This was an early video, I believe I have the music balance better now 🙂

[@BrodieTV]

@@EdThorne LEGEND

[@azuresonic69]

Appreciate your explanation! 👍

[@johnycash8266]

First guy who explain audio under 5 minutes..and everything what You need to know in that time..kudos.

[@sanders858]

Yes and no. When he compares 16 to 24 Bits, he's leaving out a crucial detail. His numbers are right, but he's not mentioning the recording process. If you're recording lower volume sound (say a guitar amp at 90db at the mic, which is below 96db) you can get by with 16 bit, but the closer to 96db you record at, the lower the headroom for processing you're leaving yourself. If you were to record the same sound in 24 bit, you'd be left with about 45db of headroom. 24 Bit is for recording louder signals to minimize noise floors, which means more headroom. That's where the difference really happens. Your ears can't hear the difference between bit rates, that value is pretty much for the recording engineer only.

[@shkedov.b]

The need of higher frequencies is when editing and slowing down tracks you will be able to retain higher frequencies after being lowered to the hearing spectrum.

[@dkm1211]

Exactly, mostly common in sound design for movies to create new fx

[@RonnieVaiArovo]

And what happens to frequencies that morph below 0 Hz ??

[@beautifulretrotimes9256]

@@RonnieVaiArovo distortion i think

[@יהודוןמגנובי]

תודה עזרתה לי😂

[@TonyMacaroni69_]

@Gus Burdeshaw stfu

[@danielprose]

With a High Sample rate like 96 you get less latency. 32-bit recording gives you recoverable audio if you did not have a ideal gain setting. For example, a gain setting may have been too high or too low. For the most part, this doesn’t really matter. however, if you were working a slightly chaotic environment with a lot of people and a complicated set up, these are nice settings to use as a safety net just in case a mistake is made. Unfortunately the file sizes are larger.

[@christianratajczak3884]

If a human had only one ear, I agree, recording at 44.1 or 48 kHz would be sufficient. However, human brain is able to resolve sound resolution for directionality, location like distance and sound character differentiation of 7 microseconds. This is why recording audio in stereo at lower sample rate take that data and lowers the resolution by 4. As the amount of that information encoded in a stereo stream recorded at 44.1 kHz is at 20.8 microseconds. It is much harder to precisely pinpoint the location of a given object in a stereo recording. You can also play with phase shift and other aspects to emulate a 360 degree sound scape using two only speakers, and you'll be able to do so if you record at 196 kHz sampling rate as the positional resolution is now at 5.2 microseconds. All that you said is correct, as a physicist, I agree with your argument, but that argument is only valid if you have a single source of sound and one ear. If you have two sources of sound and two ears, which also compute the location of the sounds encoded within the stereo stream to locate them, the spatial data of what our ears and brain can decipher needs to be contained in the recording. it's not about the pitch, or frequencies we are trying to create. It's about timing, of when sound is produced at the two sources and it arrives at our ears. When you hit the piano key and the two speakers produce the sound, but having one speaker produce that sound at 5 microseconds sooner than the other speaker to help us locate the location of the piano in the sounds stage, you can only get such small time intervals when you record at 192 kHz sampling rate. I hope this helps bit. This is why some people say that 192 kHz sounds crisper or cleaner or clearer. It's not that they can hear better reproductions of the frequency or pitch or better sound. It's because they can resolve the timing better, which in many times is translated as crisper cleaner sound. And if one covers one ear, the difference between 44.1 or 48 or 96 or 192 can no longer be differentiated. Our eyes do something similar when we look at people while they speak. As we get older we hear less, and looking at someone speak, helps our brain to fill in the sounds we may not hear anymore. Spatial positioning has a similar effect on the sound, this is why some claim they can hear better soundstage, or crisper cleaner better sound. Here's good video on this. Cheers, kzbin.info/www/bejne/qJDaqYV9nqeigMk

[@EdThorne]

This is fantastic information, and as you saw in the video, something I did not consider. Thank you for filling in the gaps and broadening the explanation. 🙂

[@christianratajczak3884]

You welcome. Our human mind and the sensors connected to it are amazing. If you have a chance check out how the mind handles imaging and how it uses simple tricks so that we can see color, motion, detail and so on. It’s amazing. Many ideas from our bodies are taken and utilized in the digital camera design. And still our bodies beat it in many ways. Cheers.

[@jonah1976]

To reduce production costs, early CD players used a single D/A converter which interleaved between channels to produce stereo sound. That did cause a tiny delay between the left and right channels, but modern D/As don't have this problem. The sampling rate ONLY determines the maximum frequency that can be represented in a digitized signal. Sampling rates cannot change the latency or timing of that signal, nor do they cause any change in the speed at which humans can perceive sound nor the speed at which those sounds travel through any given medium.

[@christianratajczak3884]

@@jonah1976 I think you did not understand what I was trying to state. When you encode a sound from natural instruments or nature sounds into two microphones for stereo recording these sounds hit the microphones at different times. The human mind can differentiate that difference to 7 microseconds. So when you encode this recording at 44.1 kHz that timing is limited to around 20 microseconds. If you do the math you’ll see that each sample the converters take is spaced out at 20 microseconds. Yes you’ll still hear the stereo. You will still hear the direction. But the resolution of that directionality is about 1/4th of what your brain is capable of resolving. Furthermore, our brain is pretty clever with audio and vision. When you have directionality that is of higher resolution the brain can actually use that data and make you feel like the sound is crisper and more detailed. That goes away if you only use one ear, or record electronic synthesizers directly into the interface as the source itself does not have such timing characteristics in the signal. There are great lectures in vision and sound and what happens when you play the same sound through headphones and introduce a tiny delay of few microseconds, you will actually think the sound is cleaner. This is one of the techniques used by our brain to get more information out of the sound and help up to decipher what we are hearing. Bose uses that trick among others to fool the listener that their speakers are better and the sound stage is larger, where in fact they heavily process and color the sound with timing phase shift snd equalization. Our eyes do the same thing. They micro vibrate or shimmer to get us higher resolution and so call ‘refresh’ the image we are looking at. And when you use a pair of eyes you get even more detail. I hope this helps a bit. Cheers.

[@christianratajczak3884]

@ReaktorLeak thank you for the link. I watched the entire video and it is informative. While he addresses the timing issue almost at the end of the video pushing through a bandwidth limited sample of the square wave, he makes an assumption that moving the sample back and forth doesn’t affect the timing as all points can be reflected. While this is true to a certain degree, it very much depends on the DA and AD converters. Each converter uses specialized circuitry to reconstruct the digital to analog by connecting the sample in time points to form the waveform that we can hear. However that circuitry is limited by its design. He addresses that when he shows the picture of alternate ways to connect these points that make no sense as going backwards in time in order to form a complex waveform is wrong. I give that to him without a question. But anything else I would argue. The reality is that while we humans are limited by 20KHz hearing range, there is a finer resolution that is part of our hearing which differentiates between the left and right which does goes beyond the Nyquist theorem. And he does say that the timing is limited by the Nyquist number. So the small timing differences are not in the resolution of as you state nanoseconds but milliseconds, otherwise we would be able to store much more data on an audio CD than the regular 700MB. We wouldn’t need Blueray if we could achieve nanosecond resolution as we could not only have a carrier signal encoded but also the modulation signal on top giving us the nanosecond resolution. His moving waveform back and forth just shows how the small details are resolved by the converters and he argues that no information is lost as timing is contained. Then my response to him would be that if he takes a decoder from today and one from 20 years ago which has the same specifications he would notice the difference using the same demo as in his video. Why? Because the function translation is much better these days on modern decoders than those from the past. Still even the best decoders and encoders can’t address all possibilities thus higher data sample rate yields better results. It would be interesting to see him comparing the 192 to 44.1 sample rate experiments and see which one yields better data beyond the Nyquist limit as that is what I talk about that it is needed to resolve timing information between left and right. I’m an expert in this as I design SerDes at rates of 25GHz for chipsets, so I know a thing or two and I would easily show the limitation of his demo as it dances around common situations of waveforms and that is what codecs and AD and DA are designed towards. But the limitations do exist. Let me give you two examples how encoders can encode analog to digital. When samples are taken they are taken at a time and recorded, or two are averaged at twice a resolution and stamped to be encoded as one average sample of half the resolution. There are many ways that encoding and decoding is done. These are just two examples I gave you. And as you see the same signal can be recorded two different ways. Which way is better? That’s a long debate. But this shows you how you can lose or average information. Do we hear that? That’s another debate as demonstrating something with scopes is limited as our brain does everything by differentiation. And Nyquist doesn’t apply in that sense. Let me give you an example I’m working on currently. I’m in Berkeley on UC Campus, from my office window I can see the Golden Gate Bridge. I take the 65 mega pixels camera and take a picture without using the zoom lens of the golden gate bridge. I can see the suspension cables of the bridge that look like the wave. But I can’t see the support cables from that main cable connected to the deck. But I can see those with my eyes. How can that be since the camera is about 4 x my eye resolution? Well, because our eye just like ears use differential calculations to form an image. Furthermore our eyes have micro movements and when our eye ‘pixels’ vibrate they see the transition of those small cables when each of our pixels goes dark and light and our brain draws that cable. But the camera of 4x the resolution didn’t see these cables as these happen to fall right between the pixels or are 1/4 the pixels. You would need a 250 megapixel camera to do what our eyes do with 1/8th of resolution. This is my friend why the gentleman who did the video should take some classes in human brain and sensory system to see that our hearing is more than the 20-20KHz sensor. Our brain uses our sensors and does many things to extract more information which his demo fails to address. We are still discovering more each day, and are wowed by the works of our brain and how it makes our limited sensors as super sensors blowing away the things we design with better specifications. Our brain can do more with limited means putting us to shame as it outperforms our designs and computers. Don’t let machines tell you they are better than you. They may calculate better and be faster, but we humans will always have tricks we are better at. Evolution made sure of that. Check out owls, how their ear offset in the height on their heads, allows them to not only differentiate left and right but also up and down much better than we humans can. It will blow your mind as it did mine. We got a lot to learn from us and nature. Engineering has yet to catch up. That’s why I became a physicist so I can poke fun of engineers. Just kidding. Cheers and love to hear from you.

[@shkedov.b]

Hey man, you're incredible! Explaining great and in the most interesting way possible. Hope you get the massive exposure soon! Thank you!

[@EdThorne]

Thank you, you're very kid :-)

[@mikeharwoodmusic]

Great advice Ed, I have only ever recorded at 48Khz and bounce down to 44,1Khz having said that, I'm currently using 44.1 and it makes no audible difference to me ha ha but a lot of difference to my cpu/dsp usage.

[@EdThorne]

Mike Harwood Thanks Mike. 48 provides approx an extra 2kHz of ‘hypersonic’ frequencies which although not necessarily ‘audible’, apparently has been shown to trigger alpha brainwaves, hence the ‘feeling’ of its presence. If you’re not short of storage, mate as well use the tools at your disposal 😀

[@teemuvesala9575]

@@EdThorne for recording 48, 88, 96 and even 192khz might make sense but for actual distribution you're likely going to use 44.1khz as that is the standard in music business to this day. all youtube videos are rendered to 44.1khz audio, as well spotify and all other music streaming services... unless you're using amazon ultra hd music or tidal hifi subscription, as those do provide up to 24 bit 192khz... however even vast majority of their lossless catalogue is still going to be 16bit 44.1khz. for movies and tv shows the standard is 24bit 48khz I think today.

[@IndraBahiaMusic]

Cool! I appreciate your honesty and simplified info on this! Love your teeth and your outro as well! :D Love from Berlin

[@EdThorne]

😊🙏 Thanks Indra

[@promiseesimorp2519]

I'm very happy I saw this. I didn't know how best to explain this to a friend and it turns out I needed the knowledge too... Thank you so much. You teach well! I'm now a subscriber

[@EdThorne]

Thanks. This is a simplified explanation but I think it’s useful to understand the processes involved. Glad you found it useful, and thanks for subscribing 🙂🙏

[@TrojanRabbit-sg9bl]

This is the best easy to understand explanation I've read and it's from 2009..."So, 24bit does add more 'resolution' compared to 16bit but this added resolution doesn't mean higher quality, it just means we can encode a larger dynamic range. This is the misunderstanding made by many. There are no extra magical properties, nothing which the science does not understand or cannot measure. The only difference between 16bit and 24bit is 48dB of dynamic range (8bits x 6dB = 48dB) and nothing else. This is not a question for interpretation or opinion, it is the provable, undisputed logical mathematics which underpins the very existence of digital audio." Taken from Head-Fi "24 bit vs 16 bit, the myth exploded."

[@EdThorne]

Very well said.

[@bigkeptsecret]

Great clear, concise and straight to the point video! Thanks

[@EdThorne]

Thank you. 🙏

[@sczealot2436]

That was wonderfully clear, thank you.

[@glevy17]

The higher the sample rate the closer we get digitally to the original analog wave form. HD 192 is technically the full value of an analog waveform. I have recorded at 192 a time or two because I could lol but I wouldn’t recommend it for daily use unless you transfer vinyl all day. I’m usually at 44.1 16 or 24 and have regularly recorded at 96 24 which is a much louder and much easier to apply plugins with. Just make sure you have a really powerful computer.

[@axilleas]

Not really, although that is a common misconception based on people trying to find an analogy between sound and image. And while in the image world the better the resolution the closer you get to “analog”, in the sound world that’s not the case. The reason is that a wave, any wave, can be represented by just two points. Imagine a sound wave described by two points, one at positive peak and one at negative peak. All you would be achieving by doubling the sampling frequency would be information redundancy, basically more points that provide nothing more than you had with the two points. The reason is that when you hit a DAC the points I mentioned are going to be converted back to a continuous wave. This stands true for any frequency till you hit the Nyquist frequency of your sample rate, i.e. half the sampling rate. There are some uses for higher rates than 48Khz but they are rather limited, namely some plugins work better at higher rates (although pretty much every modern plug in that needs it does the upsampling internally) as well as animal research (dolphins, bats, etc).

[@lHUSAMl]

Literally the most useful comment i have found I have the same opinion

[@glevy17]

@@axilleas yea and this is where dithering comes into play. I will not entertain dithering here as it hurts my head lol but yes 44.1 is the cd and otherwise accepted standard for high quality digital audio and 48k is the film standard.

[@Scripture-Man]

While it's true that any SINE wave can be perfectly represented by just 2 points, most real waveforms are far more complicated than just a single sinewave, plus most of what we hear is the combination of multiple complicated waveforms all interacting with one another to create something very complicated. So it is true that when you start reducing the sample rate, you do start to lose clarity of high frequencies and noise. But 44.1 kHz is generally accepted as being high enough for LISTENING (not necessarily for recording though).

[@savant8455]

That was some dope, forward explanation 🎉 you deserve a subscribe and a like👍

[@EdThorne]

I'm glad it was helpful 🙂

[@motuknight5569]

Having recently upgraded to high end audiophile, revealing equipment I find myself mostly “analyzing” the sound of the equipment itself. Using Tidal there does seem to be a difference in clarity, blacker background, dynamic range, imaging on the higher bit rate/kHz tracks. This shouldn’t be the case, perhaps the tracks I’ve heard are mastered better? The most profound thing, however, is that sometimes I get lost in the music and stop analyzing and just listen to the music. These are the times when the system is best enjoyed. A track played at a low resolution while enjoying the music is infinitely more enjoyable than a track played at optimum resolution while analyzing my equipment. 👍 peace and love!

[@maronius]

Well explained, accurate and concise, thanks!

[@EdThorne]

Thanks for watching 🙂

[@tommypartin6431]

Excellent video. Very well made, and to the point. Subscribed. 🤘

[@EdThorne]

Thanks, Tommy. It’s a very simplified explanation but hopefully it helps. 🙂

[@SinnerSince1962]

All of that was like an f15 Eagle.... right over my head.

[@EdThorne]

Basically 24-bit/48kHz is sufficient for everything 🙂 Maybe 96k if you want to get into over sampling with plugins

[@vmmd8229]

great info at 5 minutes no less!

[@dansellcreative]

Super helpful! Thank you sir!

[@MrOnedayamilent]

Very awesome man. Very inspirational to creating a music school.

[@DJFriendZone]

Great video, however I think with your discussion of Bit Rates you are confusing dB (signal to noise ratio) and dBSPL (decibels sound pressure level). When using a 16 bit ADC, a dynamic range of 96 dB refers to the fact that quantization error will affect the signal, but do so with a signal to noise ratio of 96dB which is represented as 10*log10(original signal/quantization error). This error is incurred by the process of quantizing the amplitudes when sampling a signal. Both a 16-bit signal and a 24-bit signal can have peaks as loud as 140dBSPL or more when played on a loud enough system. Interestingly enough, researchers generally found that it takes a dynamic range of 118dB for the effects of quantization noise to be inaudible. However, with that being said, a 16 bit ADC, if dithered correctly, can have a perceived dynamic range of 120dB, so the benefits of using a 24 bit may not be as pronounced

[@EdThorne]

Hi there. Thanks for the positive feedback and lengthy explanation. My video is definitely a basic summary of the topic without going into the advanced science of it. Hopefully people who want to understand it more thoroughly will see your comment, or access the further reading literature in the description. 🙂

[@HarvinderSingh-yy8th]

@@EdThorne Sir, is HR versions of TASCAM US 16*8 launched by the TASCAM CO.

[@rogerwilco6355]

10/10 A +

[@johnaweiss]

Will a 16-bit 140dBSPL signal sound different from a 24-bit 140dBSPL signal? Even if there's more quantization error at 16 bits, will we hear that error at 140dBSPL?

[@khnimusic]

Thank you for the clear descriptions!

[@EdThorne]

You're welcome, I'm glad it was helpful.

[@sevenstringsyndicate6250]

This video was done so well omg

[@IzzyThomasOfficial]

Found this really helpful, cheers! A producer i work with tells me to set the sample rate to 48khz, do you know why he would suggest that?

[@EdThorne]

Hey Izzy, Either because he thinks more is better and hasn’t thought about it any further (understandable, most people are like this) OR because he knows 48kHz sample rate in theory captures up to 22kHz frequency and the extra 2k in theory creates more presence in the sound. The reality is it’s hypersonic audio which we can’t hear but research suggests it triggers alpha brain waves which may explain why some people claim they can ‘feel’ the higher frequencies. At least, that’s what my research has lead me to believe. Also, broadcasting audio is recorded at 48kHz, so maybe he wants your recordings to be inline with that, without needing to convert files which may jeopardise the quality of the audio.

[@babagandu]

Because 48 Khz is half of 96 Khz

[@SuperIZL]

48khz let's the cutoff filter work at 22.1 khz instead of at 20.5 khz Wich allows for the cutoff filter to be less steep. Resulting in an opener bit more natural sound. It's not voodoo.

[@babagandu]

@@SuperIZL voodoo !! I use 768 sample rate

[@Wildeedge]

If you use your machine for more than music it should be at 48. It's the standard for the video industry.

[@RonnieVaiArovo]

Thanks Ed man, just discovered your videos as I was doing deep research into the Apollo twin. You are succint and your tutorials are very well explained ( and sometimes humerous!) and as far as I can tell, your teachings are accurate without being overly technical.

[@EdThorne]

Hi Ron, That's great to hear, and exactly what I'm aiming for, thank you :-)

[@arturzaduryan6108]

>deep research >discovers youtube videos way to go xd

[@rubengarcia6394]

If I’ve been recording the majority of the track in 16 bit, and I switch to 24.. would there be any changes to the track?

[@EdThorne]

Your computer will have to re-sample all the tracks, but this should only be once. There’s no benefit to doing that now if you’ve recorded at 16. It won’t improve the quality. 16 is also plenty of dynamic range. Most people’s tracks realistically only cover about 2 bit of DR (12dB).

[@Techiemotorist]

Thanks for such video, it totally cleared everything

[@smaganas]

Well done!!

[@pineapple_1066]

Super super clear!!

[@NipponKiwi]

Great Video ⛩ Mr Ed 🌍! I Learnt Heaps💡🧠! Love 💓 24 Bit 🎶!

[@mannradin4837]

Great Explanation and great Video keep it up

[@EdThorne]

Thank you 🙏

[@Anothercore]

What should I choose for youtube?

[@EdThorne]

KZbin recommends 24bit/44.1kHz. Higher sample rates are acceptable: support.google.com/youtube/answer/6039860?hl=en-GB

[@jonah1976]

Nice video. More people need to understand why formats beyond 44.1/16-bit don't really offer any benefits for the end user. However, the reason the 44.1khz sampling rate was chosen for CDs has nothing to do with FM broadcasting. FM pretty much tops out at 15khz, which is why TV and radio stations ended up using 32khz sampling rates for their in-house digital recordings (see Digicart). Rather, back in 1979, videotape was the only storage medium with enough bandwidth and storage space to make digital recording and playback feasible. With PAL video for example, 294 lines per field, 50 fields per second, with 3 samples stored on each line. 294 × 50 × 3 = 44,100.

[@LucaPalomo909]

Thank you, kind sir.

[@oliversaxon8656]

Nice, you’re like the Gerald Undone of the music world!

[@EdThorne]

His channel is great, I’ll take that as a compliment 😀🙏

[@kitegames3351]

awesome explanation

[@EdThorne]

Thanks🙂

[@chandra16108]

Thanks, nice to know all this!

[@ivaniasalasfranco4286]

very very very helpful, love your channel!

[@EdThorne]

Thank you 🙏 It’s a simplified explanation but I’m glad it helps 🙂

[@peaceprosperity547]

Thanks mate!

[@JameyTiffee]

Higher sample rates are only effective for instruments that produce harmonic overtones but if the mix is crowded, don't bother.

[@EdThorne]

Pretty much 🙂

[@raymondlooi]

Wow... This is fun & indeed it's a good lecture! 🤩👍🏻

[@EdThorne]

I can’t tell if you’re being sarcastic. I’m kinda hoping you are cos that’s hilarious 🤣🙏😀

[@raymondlooi]

@@EdThorne Honestly, you are AWESOME! 👍🏻

[@EdThorne]

Raymond Looi 雷聖雄 Ah you’re very kind, thank you 🙏

[@anmolhans1288]

really appreciate your work brother

[@EdThorne]

Thanks Anmol. I hope it’s helpful 🙂🙏

[@anmolhans1288]

@@EdThorne love from india bro i wanna follow you on instagram are you in it?

[@EdThorne]

@@anmolhans1288 Yes 🙂 @TheEdThorne

[@tandcatthemovies9818]

Great video!

[@EdThorne]

T and C at the Movies Thanks for the support

[@markushoffsten]

So what is best for me on a Computer PC. i have a sound card that can play 16-bits or 24-bits and betwen 44.1 Khz - 48 Khz - 88.2 Khz - 96 Khz - 176.4 Khz - 192 Khz ... Witch one is best do you think? Thx for a good video. / Sweden!

[@americanswan]

I have a question. Choose one song encoded in wav, 16bit and 32bit. Open the wave file. Find the third sample. What's the binary number stored and way? Open the 16bit file. Find the third sample. 16bit. What is the binary number and why? Open the 32bit file. Find the third sample. What's the 32bit binary number and why? After you have the binary numbers, then explain how the two or three binary numbers store your favorite vocab like the dynamic range, resolution, bit depth, db, loudness etc. And how the two binary numbers, one 16 and the other 32 are both lossless flac. Is 24 really better than 16 bit? Both use lossless flac encoding? Not one youtube video explains any of this, so it seems they don't know what they're talking about. I have yet to get an answer from anyone to my logical question.

[@sytiravajous]

Best way to test the application of sample/bit rates is in editing context. You can test if aliasing occurs with high pitched voices or added saturation, sine wave is good for testing because it has only one harmonic - which is also the fundamental frequency (the note you're pressing). You will see additional frequency response BELOW fundamental frequency if the audio bounces back from nyquist threshold. Another way to test is by slowing audio down, with higher bit/sample rate you have more frequencies above 22khz to shift back down to audible range. Third way to test is by clipping recorded audio past 0db on purpose. With higher bitrate, the audio is still recoverable (to a degree) if you turn the volume down after the fact. With lower bitrate you will see the peaks flattened out and the damage is irreversable.

[@chrisfeatherstone9691]

Great job! Keep up the great work!

[@EdThorne]

Chris Featherstone Thanks Chris 🙂

[@shaw-krowdashsabe2302]

An interesting perspective but when multiple sounds are playing at once you need a higher sample rate to acuratly reproduce the multiple perspectives as for each aditional sound more of the profile is used at once to maintain the multiple of resonate freq you need a higher tone (above human hearing) to realign the tonal values harmoniously with higher accuracy allowing more distinct differences between them as sample rate is increased

[@silas1414]

Interesting 🤔

[@Eyoldaith]

Supposedly higher sample rates do have an effect, not because the human ear can hear the difference in sample rates itself, but rather due to DA converters causing distortion at lower sample rates which doesn't occur at high sample rates. Whether this will have an audible effect outside of audiophile gear however, is a different question.

[@ihsanmoh]

So, I have question I have in windows speaker properties there is a "default format" I have 24 and 32 bit , 48kHz (studio quality) and also I have 32 bit 192kHZ (studio quality) etc, which one should I put and it will be any difference ?

[@EdThorne]

You won’t notice a difference but the matter requires more CPU power. 24/48 is plenty.

[@DroseMr]

As a new guy would it be safe to say recording 32 bit float i would set sample rate to 64? Does it mess quality up if sample rate is set at 96?. Thank you

[@EdThorne]

Bit depth and Sample rates are different things and not linked. I suggest 24/48. Bigger than that and your files sizes are unnecessarily large and your plugins (CPU) will struggle.

[@SportMinsitry]

Normally I use 44.1 /24. But my project sound gets out of tune sometimes, why?

[@EdThorne]

Check the sample rate of your samples if you’re using sample libraries. Also check the sample rate of your interface is aligned with your DAW.

[@darrentantoco]

Thank you! I've no problem now with the FiiO BTR3K I bought not giving a sample rate of 384KHz over USB-C DAC functionality.

[@EdThorne]

I've never even heard of 384kHz devices existing, wow.

[@darrentantoco]

@@EdThorne the FiiO BTR3K has it over Bluetooth but only 44.1KHz over USB-C, apparently. The FiiO BTR5 does 384KHz over both.

[@karoleq86]

Its actually impossible..384khz through bluetooth.That is Bs

[@rocioelys]

great video! out of curiosity, do you know if Spotify allows uploading songs in 24bit 48kHz? or they still using 16bit 44 kHz?

[@EdThorne]

Thank you. You’d have to check their website. I upload everything to DistroKid at 24bit/48kHz. Spotify may convert it, I don’t know

[@danieltill.marketing]

I'm trying to find this info - Spotify support is hopeless and there is nothing online...

[@dordasjaireartates8213]

If I'm having a 44.1 khz sample rate, what should I'd be having as rendering rate for better sound quality?

[@nathangabriel8888]

this was super helpful thanks!! so much marketing BS

[@EdThorne]

You’re welcome, Nathan. Thanks for watching 🙂

[@LoisLovesFitness]

when exporting my podcast I export at mono and it automatically has 192kpbs is this okay??

[@EdThorne]

Conversion sample rates (as discussed in this video) and Mp3 sample qualities are different things. 192kpbs is good for a podcast 👌

[@JohnDoe-dj3lw]

I didn't care a bit (lmao) about sound quality and kinda still don't (not exactly my thing), but I liked this video and your presentation. You've got me curious what can I say lol

[@EdThorne]

Haha. Thanks 🙂🙏

[@SuperIZL]

You made a mistake in your maths. The filter is at 22.05 khz translating to 44.1 khz when applying the Nyquist rule. 2x 20.05 is 40.1 in my brains. For the rest awesome vid one of the best on the subject! Furthermore 24bit is used in pro audio mainly because of the lower noisefloor so very handy. Higher samplerates in pro audio are mainly used for soundesign and to combat alliasing in specific plugins not to reproduce inaudible frequencies to the human ear.

[@EdThorne]

Well spotted. I’ve probably just said the wrong number, it’s easy to do when doing multiple takes mentioning multiple different numbers. Your other points are spot on too, useful to consider when choosing your settings.

[@SuperIZL]

@@EdThorne :)

[@accentontheoff]

Hi, would you advise I get a multi effects guitar pedal that outputs at 44.1 kHz or one that outputs at 48 kHz. As I understand it, audio typically requires 44.1 khz and video requires 48 khz. Since I work on both, would I need to upsample or downsample the guitar parts, or some such, or maybe just pick the multi effects unit for one of the two types of work. Thanks in advance.

[@sebastianvega4576]

it is so funny that a guy that admits at the beginning of the video he had no idea what sample rate and bit depth exactly is, gives the most accurate and short answer on the topic. shows how important it is for the average bed room producer: not at all! use 44,1 kHz at 16 bit until you actually have a reason to use something else and freaking concentrate on your music!

[@EdThorne]

Ha. Thanks. Hope it was helpful.

[@gibson2623]

For recording you do need 24 bits, that extra headrooom is very welcome :)

[@audio_buck]

Great explanation, although it was a 'Bit' over the top. Ha, just playin. It was super helpful. Thanks!

[@EdThorne]

Lol. It’s a very simplified explanation to what is very complicated science but hopefully it helps people understand a little better. 🙂

[@LaminarSound]

I was waiting for the “but.....”. And it never came. Lol great presentation of the truths but there’s def more to it than just marketing huff. For capturing audible frequency range maybe not but when working in post, higher frequencies most definitely effect the audible range. Now, to what extent is debatable. Lol still, great video mate.

[@EdThorne]

Sean Hebert Haha, thanks. This is definitely a simplified summary of the relatively complex science behind it, and I agree about the higher frequencies in post.

[@ejctechandhobbies]

Great video sir! but for my question, what do u recommend for my microphone setting for streaming, making youtube videos (song covers/vlogs/reviews), gaming, and for meeting calls? I only have 16 bit in setting but i cant choose between 16bit 44.1khz or 48khz (48 is the default on my headset's microphone) and for listening thru my headphone's speakers I choose between 24bit 44.2khz or 48khz ( the default was 24bit 48khz) which one's the best too? thank u in advance for the help guys! :))

[@EdThorne]

For streaming it may be worth keeping the audio quality to 16/44.1, but streaming audio is more about the Audio Bitrate (128/256/320 etc). That depends on what your streaming service (OBS/Streamyard) performs best at. Mine is currently set to 48kHz and 256AB

[@ejctechandhobbies]

@@EdThorne tysm! My pc is not upgraded yet so i might stick to the smoothest one 🤍

[@cagkansavk]

Which sample rate should I choose to streaming with obs ? 44.1k or 48 k ?

[@EdThorne]

To be safe, I would personally use 44.1 to minimise any streaming/bandwidth problems.

[@Duke49th]

Thanks a lot. And as most people want MP3, it is even more useless to have something above 48khz (or even 44.1khz). And even with FLAC or any other "losless" codecs, who has headphones with more than 20 or 24khz?! First time now I really understand what that number is lol. I only knew, the more the better (in theory).

[@EdThorne]

Glad you found it useful. This is definitely a simplified explanation but one that at least makes sense. :-)

[@antoinedevldn]

Thanks that was very usefull

[@EdThorne]

You’re welcome. Glad it was helpful

[@ZkReAr-fs5do]

What about in streaming ???? There is a 120 160 320kb/s for audio..... do u think it makes a difference in streaming/recording??? Talking about quality ....

[@EdThorne]

Streaming rates are different to bit-depth and sample rates. Go with the highest your internet can handle (obviously factoring in screen capture video quality, which will consume the bulk of your band-width)

[@raulgomez8523]

My microphone records at 16bit, is it advisable to record at 24bit or 32bit in my DAW?

[@EdThorne]

I’d always suggest recording at 24. 32 is unnecessary.

[@raulgomez8523]

@@EdThorne Thank you very much!!!!

[@GenilsonOficial]

I believe that those who will simulate analog equipment with plugins at higher resolution will bring better results.

[@bitonic589]

I've made my own file format, modified flac 64 bit depth, amazing quality

[@EdThorne]

EPIC! Have you got audio examples?

[@bitonic589]

@@EdThorne idk how to make a download link

[@EdThorne]

@@bitonic589 Dropbox/Google Drive maybe? No worries if not, I’m just curious 🙂

[@hedgeearthridge6807]

I took a 24 Bit 1700 kbps FLAC of Metallica's S&M2 "Ecstasy of Gold" and converted it to 16 bit, and then compared. Wearing Samson SR850 headphones and with my questionable hearing, I could not tell a difference. There is a placebo effect im sure, but with random tests I cannot tell. Im sure with thousand dollar equipment you can, but with basic stuff it's just not there. Bitrate is what makes the biggest difference. Comparing an MP3 to a FLAC is like comparing a 360p video to a 4K video. 96 kbps and lower is unlistenable most of the time. 192 kbps is really good actually, not easily noticeable from the FLAC. Of course you should use OGG Vorbus or Opus instead of MP3, but I wanted to try the worst option for the biggest effect.

[@malibustacy3606]

I'd like to see you make a much bigger fuss about not wanting to be referred to as "Eddy Baby," and perhaps go as far as to describe in some detail which names you don't mind being referred to as, such as "Ted" or "Edward" for example.

[@EdThorne]

Lol…

[@motech6432]

nice video

[@EdThorne]

Thanks, Muhammad!

[@thethunderbulll]

I just record podcast and apply basic fix like compression eq . I use Zoom H4n pro . Will there be significant difference between 48 khz 16 bit vs 96khz 24bit?

[@EdThorne]

The dynamic range of a podcast will be very little so 48&16 will be plenty. Try it and see if you can hear a difference, it’s unlikely though

[@thethunderbulll]

@@EdThorne thanks for the support

[@mustakim2144]

Thank you

[@EdThorne]

You’re welcome 😀

[@llamasarus1]

If each sample is equidistant and each bit depth value represents a volume, then does the jump between bit depth values determine the pitch as the computer "understands" it?

[@EdThorne]

The computer understands the frequency (which we hear as pitch), yes.

[@llamasarus1]

@@EdThorne My question is on HOW the computer understands the pitch. And I'm asking if the interval between bit depth value's is what determines pitch and if not what does?

[@djkriswhiz254]

So between“16-bit 44.1kHz” and “24-bit 192kHz”which one is the best?

[@EdThorne]

I’d always suggest 24bit depth. If you don’t plan on slowing down or speeding up the audio I’d suggest 48kHz for file size and system speed. Or 96Hz if you feel you notice a difference but this puts more stress on plugins and cpu load. I record everything at 24bit/48kHz.

[@djkriswhiz254]

@@EdThorne Thank you so so much Ed,,😊🥳🙏💯

[@therealrussellsmyth]

@@EdThorne cool. Should I upload 44.1 or 48khz for a KZbin music video? I always upload 24 bit but unsure re the sample rate

[@EdThorne]

@@therealrussellsmyth I’d suggest uploading at whatever you recorded at. Converting from 44.1 up to 48 isn’t worth doing.

[@therealrussellsmyth]

@@EdThorne thanks man. From researching it seems KZbin accepts 48 but automatically converts to 44.1. Prob best to perform the conversion myself!

[@joelpoitras6226]

16 bits multitrack recorders vs 24 bits multitrack recorders by Tascam and Zoom. Is the 16 bit(CD quantity) record a lesser quality recorder than a 24bit Besides editing or and using pluging. Is there any audiofiles outhere that claim they can distinguish 16 bit quality to 24 bit? Explain

[@heyguyslolGAMING]

If I identify as a bat would that allow me to hear above 44.1khz?

[@EdThorne]

Ha, I recon so for sure. Try it and let me know 😜

[@emilsvlog1977]

Microphone with sample rate 192khz 48 bit. Can we record 48 khz 16 bit with this mic? Is it adjustable?

[@EdThorne]

A standard microphone doesn’t convert audio. You may have a USB/interface microphone that does, and yes, these bit depth and samples rates will be fine.

[@skeptick_souls4903]

thanks bro

[@EdThorne]

You’re welcome. Thanks for watching 🙂

[@HarvinderSingh-yy8th]

Sir, is TASCAM 4*4 good for recording song.

[@EdThorne]

The Tascam interfaces are great, they sound lovely. It was probably the best sounding interface in this Top 10: kzbin.info/www/bejne/rmaTlaqVd8uomrc

[@HarvinderSingh-yy8th]

Sir, I also observed all 10 tracks of these interfaces. I also found that TASCAM sounds amazing. My second question:- is HR serial best for song recordings or I should go with old non- HR interface. I sing all Old bollywood songs. I want to setup my home studio. So pl. guide me correctly.

[@EdThorne]

@@HarvinderSingh-yy8th May as well go for the newest model (HR)

[@HarvinderSingh-yy8th]

@@EdThorne HR is not available in TASCAM US 16*8 interface.

[@AmarachiDesi]

Is an audio bit of 126 kbit/s the same as 16 bit?

[@EdThorne]

No, 16/24 bit-depth refers to the dynamic range of the audio at the conversion stage. kbit/s is the resolution of the mP3 file.

[@WoWu-d5x]

The human ear can resolve 7 microseconds. Everyone always talks about 20kHz .... But that is something completely different. It's like comparing the frame rate of an image with the image resolution, for example. 44.1 only fulfills 1/3 of the human resolution. Each sample is therefore 3 times as long as in the natural signal. In this respect, the question of HIRes in relation to the temporal resolution can be answered with "yes". The lower the resolution, the "rougher" the recording appears. To justify 44.1 with Nyquist does not take into account the filters and their quality - in 1930, when the theorem was calculated, there was neither digital technology nor filters.

[@SirWickMusic]

VERY ACCURATE!!! The pronunciation of "Aliasing" had me at first, but that was only my very American ears :) LOL. Still, great video and one of the best I have seen since learning this in 1995!

[@EdThorne]

Yeh, it took me a few goes to pronounce it to 🤣 That’s a strong testimonial, thank you very much 😀

[@SirWickMusic]

@@EdThorne no problem! I subscribed also!

[@jayceelambii2073]

Just because you can't hear certain frequencies doesn't mean you can't physically feel them! Don't forget this.

[@EdThorne]

I have read that frequencies above 20k trigger alpha brainwaves giving us the impression of feeling them. I believe the same article also said there’s no evidence of frequencies above 25kHz triggering these brainwaves.

[@SinnerSince1962]

Okaaaaay.....

[@thanglongliving]

I often use 16 bit / 48 Khz for recording live stream, 24/48 or 24/96 for music.

[@MrHiglon]

So I got my hands on some good headphones but don't really know anything about audio. I use it right now with a Fiio k5 pro and encountered some cracking noises while playing games. (Dead Space). They happened after I set the audio resolution (I hope thats the term) to 32 bit; 384000 hz. Now I figuere that setting might be overkill and leading to the cracking artefacts but I'm not sure which setting to pick here. If I put it at 16 bit; 44.1 hz wouldn't I bottleneck the headphones? Would they still sound somewhat premium? Or is there a known sweetspot? Maybe is it always dependable on the audio source itself? I hope someone can explain this to me and that there is a setting I can leave it on while still having the quality of the hardware shine through.

[@EdThorne]

The crackling is likely digital clipping from overloading your output. Try turning the game volume down and computer/interface output up.

[@phillipallen5564]

192khz is great for surround sound bur stereo it makes the highs and treble cleaner

[@No_Limits_411]

What I've learned is that 48 khz is used mainly so the sound could be better linked up with the video.

[@ewitte12]

Computer audio codecs are mostly compressed the higher bitrate files lose less. There are lossless varient but not as common.

[@waedi73]

How many steps is then 16 bit ?

[@Licaonte]

Hey man, very thankful for the explain, but I still have a ask, do I lose everything if put my headset in maximum Hz and bit rate on the windows? 🤔

[@EdThorne]

Your headset?

[@Licaonte]

@@EdThorne seinnheiser gsp 600

[@siceastwood2714]

The bit rate you're referring to is actually the bit depth

[@EdThorne]

Point me to a time stamp?

[@cole1]

Cool video, but kind of misleading! For vocal manipulation, such as time stretching and melodyne / auto tune, higher frequency ranges make things much more forgiving. I do greatly appreciate your charts and figure for things such as size, though. Looking forward to more!

[@EdThorne]

Cole Hi Cole. Thanks for commenting. I understand how time stretching may require a higher sample rate but what makes you think tuning would as well?

[@cole1]

@@EdThorne I was just referring to the time chopping abilities of both of them

[@EdThorne]

Cole Ah ok, yeh, that makes sense.

[@neerajkr123]

Yeah, but what is the best for listening purpose- 16 bit 44.1KHz or 24 bit 48kHz / 192 KHz ??

[@jonah1976]

44.1Khz/16-bit. Search KZbin for "D/A and A/D | Digital Show and Tell (Monty Montgomery @ xiph.org)" if you wonder why.

[@neerajkr123]

@@jonah1976 okay, thankyou 😊

[@olorb2618]

196khz 32bit of course

[@olorb2618]

@@jonah1976 not true 48khz 24bit sounds much better, especially bass guitar

[@zemi2750]

For me it's 48K for games & movies. KZbin and browsing random web videos: 44.1k (you can leave it at 48k if you want, downsampling shouldn't be an issue for any medium overall). For recordings: 96k/192K if you need the resolution or want to do more on your mix, depends on your goals, if not.. just straight record to 44.1 since this is the standard. Keep in mind the buffer size when recording to avoid delay. Depends on what your PC can handle, the lower the better. I know you're asking for "listening" but just wanted to simplify for others.

[@sheffoooproductions9738]

192k is not a marketing tool or smth audio interfaces with higher like 48k 96k 192k they capture better high frequencies on human hearing

[@EdThorne]

But humans can’t hear above 20kHz so technically anything above 44.1k (22.05kHz) is wasted (unless you’re slowing down Audio. Now, there have been studies that suggest frequencies above 20kHz, although not audible, do trigger alpha brainwaves giving us the sensation of ‘feeling’ those higher frequencies, but even then these didn’t go past 24kHz. What I didn’t find out for this video was whether or not there are companies producing devices that sample more than two samples per cycle, 2 being the minimum to avoid aliaising. For example an interface that could capture 4 samples per cycle could easily use 88.2/96kHz and sound amazing due to the resolution so at that point it wouldn’t be marketing fluff at all, just a killer product.

[@sheffoooproductions9738]

@@EdThorne yes i understand you very well in explanation , what i said was audio interfaces that gives just 44.1khz are not good at high frequencies at 20khz on human hearing but in 48khz interfaces they produce better on high frequencies on human hearing betwen 15khz to 20khz

[@EdThorne]

@@sheffoooproductions9738 Yes, this makes sense 🙂 I think after 48kHz (which is what I record at) it’s a question of file size and processing speed. 96kHz is a LOT of data A) to store and B) for your plugins to process, and therefore CPU/DSP load.

[@sheffoooproductions9738]

@@EdThorne yes youre correct , i have interface with 192khz i use it on 48khz so interface with 192khz has more capable amplifiers but i set it on 48khz everytime

[@Islamicknowledge-yx8rt]

So all I'll get is more bats outside?

[@EdThorne]

Learn about Islam Haha. Best stay at 48k, for the rest of 2020 at least 🤣

[@Islamicknowledge-yx8rt]

@@EdThorne 😂😂😂