Why Hi res - an updated view
Вставка
- Опубліковано 2 чер 2024
- 96 kHz, 192, kHz, 384 kHz or higher, or DSD256? Does it really sound better? Over the years I have learned that things are not all what they seem. How about hi-res music?
Contents of this video
00:00 - Introduction
00:27 - Start of program
02:04 - Hi-res ’standard’
03:31 - 24 bit resolution
05:02 - Do we want 24 bit?
07:45 - Do we want 192 kHz?
10:19 - Oversampling, the solution?
11:31 - Is Hi-res the solution?
12:41 - The wrap
Relevant videos:
About our hearing • About our hearing
About my reference setups, July 2022: • About my reference set...
Relevant link:
www.jeita.or.jp
If you like my work, support it through:
My Patreon page: / thehbchannel
Paypal: www.paypal.me/theHBchannel
My book: • File Based Audio aka S...
My links:
My site: www.theHBproject.com/en
My channel: / thehansbeekhuyzen
My Facebook page: / hansbeekhuyzen
My Google+ page: www.google.com/+TheHansBeekhuyzen
My Twitter: / hansbeekhuyzen
My reference sets:
About Questions: • About questions
What DAC is the best: • What DAC is the best? - Наука та технологія
Absolutely detailed explanation of Hi-Res complexity
Thanks
I'm sure I speak for quite a few others as well, when I say that we really appreciate your technical breakdowns. They are informative, entertaining, and helpful in providing an understanding of what and how we're hearing. They also help when my friends have questions about how my various purchases aid in my love and appreciation for music, and movies for that matter. Thank you for all you do, and I pray you keep doing it for a long while. Bless you and stay safe. 🙏🏿
🙏🏻🙏🏻
I second this comment, and will add that your intelligent and open minded approach to the black hole that is audiophile equipment is most appreciated. For that reason, I have tremendous respect for your opinion. Your occasional dry humor cracks me up. I could have a couple of beers with you, and know in advance that it was going to be a great and thought provoking evening. Thanks for the knowledge, Sparks in Daytona
Your technical explanations are always so clear, thanks for this.
🙏🏻
Thanks for another great video Hans! I have recently come to the realization after many years of trying fancy oversampling and upsampling DAC’s, MQA decoders, 192khz etc that I actually prefer 16 bit 44.1 kHz especially if the music was originally mastered to sound good on CD. Just sounds more natural to my ears for long listening sessions.
I do like well recorded hi-res music, but most of the music that I enjoy was created before “hi res” was a thing.
Enjoy the music.
I have a Marantz CD recorder and it sounds the most believable at CD quality, smooth and lovely.
Its always pleasure to watch your content. Thank you :D
Happy to hear that!
This was truly excellent, Hans. Thank you.
Many thanks!
Another great review Mr. Hans.
🙏🏻
Hello Hans.
Thank you very much for this detailed information.
You are very welcome
This is the best explanation I've heard. Thank you.
Wow, thank you!
Many thanks once again, Hans. As always with you, there is much food for thought.
You are very welcome
Thank you for everything you did for us
🙏🏻👍🏼
Great stuff as always 👍🏻
Thanks again!
Thank you Hans...i was always confussed about all the topics you have covered in this video.....made everything clear to me..cant thank you enough
👍🏻
Very well explained. Thank you for making this video, you are making us all smarter with each video.
Glad to help
Excellent video. Thanks a lot!
🙏
F
Thank you, sir. This kind of honest, intelligent conversation is the reason I subscribe to your channel.
🙏🏻
Dank voor je uitleg. Niet alleen leerzaam, maar ook gewoon leuk om te horen. Zo begrijp ik meer en meer wat, en vooral waarom ik het op een bepaalde manier hoor mijn installatie.
🙏🏻
Your conclusion make so much sense 🎉
🙏🏻
Outstanding as usual.
Wow, thanks!
Hey Hans, it is always a pleasure to watch your videos, I addressed the same topic a few days ago on my channel (in German only) and came to a very similar result but it is always very nice how you are deconstructing the technology behind it. Thumbs up!
👍🏻
Awesome Video, thank you Hans!
Many things have been said before but you have also made some new points which are very interesting. I was longing for such a Video where “the state of the art” digital processing and everything what has to be considered in digital audio in summarized for quite a long time. I really think it helps some of the people who still have an open mind about HiRes and Digital audio in general to get a better understanding (at least I do) and maybe some even overcome their dogmas (bits are bits and more than 20kHz is useless).👏
👍🏼
Hans, I know this video will help me really understand the matter in hires audio, but I need to digest it for a moment :-). I wonder what would happen if you were consulted for a DAC chip development. Your knowledge is practical, theoretical, scientifically sound (pun intended), drowned in first-person experience and therefore priceless. Keep on debunking the marketing crap thrown at us which I took for granted, despite my own electronics and software background. Thank you (again)!
If I would get involved in product deleveopment, I would loose my independence, meaning I had to stop reviewing. And reviewing is my life, my joy.
Hello Hans, very interesting!
It's an intriguing subject to me because I've always struggled to hear the difference, even between a good bitrate AAC/Vorbis/Opus and a PCM file (though it is audible). I take it neither my ears or gear are good enough for these subtle details. I'm also aware of how psychologic factors can affect our hearing. Things tend to sound better when we think they will. That's why a lot of people prefer the sound of vinyl: even though it's objectively worse than even lossy formats used for streaming, a physical piece of music has stronger emotional affect on people (and we tend to get used to the nice distortion and frequency response it has).
Not to say I don't believe your experience, as your ears, knowledge and gear are obviously far superior to mine. But I'm curious if there has been tests conducted in a "blind" environment proving that Hi-res actually sounds better for some people.
Something else I'm thinking a lot about is the inefficiency of 24 bit when it comes to lossless compression. More and more streaming subscriptions give lossless as an option, and I think it's dumb to give 24 bit quality as a default. Lossless encoding struggles a lot more with 24 bit, resulting in often twice the bandwidth for no audible benefit (in my experience). Apple Music doesn't have an option to stream 16 bit and will always use the highest available bit depth, and I think it's just bad for accessibility, optimisation and carbon emissions.
I'd be very interested to know your thoughts about all of this!
Concise explanation as to the issues of chasing 'higher and bigger (#s) is better.' Hifi audio has been mincing engineering principles for quite some time now.
Great insight as to the steepness of the LPF as to why a better DAC will tend to 'resolve' better regardless of the source content! This was a great revelation for me.
👍🏼
Thank you Doctor Audio
👍🏼
I had been upsampling to 192kHz in Roon and felt I could hear a very small improvement in separation and detail retrieval. It was so small as to make me think I might be imagining it. Then I tried upsampling to DSD512 and heard the soundstage, layering and refinement make a major leap forward. My take away is that pushing the compute-intensive delta sigma modulation to the PC and bypassing it in a cheap chip DAC (with native DSD) may be a good strategy to get much better performance out of inexpensive hardware.
That plus the fact that your DAC handles DSD better than PCM, obviously.
Imagine that source was DsD not CD.
My source is Qobuz so mostly 44.1 kHz 16-bit FLAC upsampled in Roon to DSD512 and sent to my ESS9038-based DAC as native DSD (over USB). Looking at the 9038 datasheet, it appears to bypass the chip’s limited delta sigma modulation and more directly convert a native DSD stream. I don’t know the specifics beyond that but it is exciting to me to get such a bump in performance for free (well, equipment I already own). It does place a heavy load on my iMac’s Intel i7 but as long as I am not doing other things on the machine it doesn’t seem to be a problem.
Thank you very much
You are welcome
Thank You Sir!
Most welcome!
Thanks Hans for a very clear explanation of the differences. I'm more than happy with 16 / 44.1 - it meets the good enough bar in my view.
And - from my standpoint - 16/44.1 plays in nearly everything without issue.
👍🏼
Well said, Hans. You know your stuff. :)
🙏🏽
You're 100% right! There's nothing to add.
🙏🏻
Thanks for your explanation. Very informative. I have always thought higher sample rates in music studios are used for capturing the best possible recording and for mixing/monitoring. I have also heard it helps combat aliasing introduced by digital processors/plugins in the mixing stage. Although I really have no idea.
We all learn over time, me included.
good video Hans - but since nearly all pop and rock 24/96+ HR on sale all use 'remasters' , remasters which have been dynamically crushed since the loundess wars since 1993 ish - the point of HiRes fed in to expensive equipment is almost moot.
There are notable exceptions. Pink Floyd HR are not dynamically crushed. Green Day took out the brickwalling of American Idiot for the HR 2012 release.
But just about everything else is - the latest Bowie , Queen, Prince etc etc etc HiRes are brickwalled horror stories in the wave form. They are fatiguing beyond belief to listen to.
They are LOUD-IN-ALL-PLACES-ALL-THE-TIME. They are Sound Formely Known As Music. Much of the vinyl revivial has surely been a search for music with dynamic range...
So Hi-Res technicals are almost meaningless when the vast majority of source material is more juiced than the East German Olympics team.
Original CD pressings before 1993 are going to be the Rosetta Stones of digital music (no vinyl snap/crackle/pop and pink noise friction soothing swhoosh for me) .....
pre 93 CDs are going to go (in a small way) the way of current vinyl mania - because they are the clearest representation of the mastertape that you can get (short of reel copies). And are ever likely to get. Because gatekeeping music streaming services use remasters, and music rights are being bought in bulk by hedge funds, and the brain is evolutionarily wired to like louder.
Going back is not an option. Now "even quieter and louder than before" aint going to sell...
It's the only case where technology has made playback fidelity worse. 1080p & 4k allow you a better experience than the whole cinema going public pre 2010, when the film has been restored, on reasonable priced kit.
But the loudness wars have destroyed much of music reproduction.
And the "ONE LOUDER THAN 10 "shift has impacted directly on the downgrading of chord complexity in music - with a brickwalled backing track only being able to be ridden against by a nursery rhyme like melody. Cue just about anything after 2002.
Check out Vinyl guru Michael Fremer YT video about testifying at the Quincy Jones / Jackson royalties trial. And having to explain to the jury that every CD since the 1987 original of Bad had been dynamically 'crushed' leading to the 25th anniversary brickwalled debacle.
The loudeness wars were a first order of magnitude crime against human culture. Ongoing....
Argreed
Brilliant comment, you have hit the nail directly on the head. I think “mastering” engineers have a lot to answer for.
@@bobbradley3866 And producers and artists, etc.
I thing your logo what is on display behind you is brilliant.People can see HiFi or HB and that everything in EQ look 👍 Good Job.Thank you for explanation in this video.What about MQA ? Maybe we get also updated video about that?
I'm trying to get my grips on MQA in practice. It seems that MQA releases sound better on myy setups 1a and 1b. 1b uses the Mytek Brooklyn MQA DAC but 1a uses the first unfold on Roon and then the upsampling by Grimm MU1 to feed the Chord DAC, which is not an MQA DAC.
@@TheHansBeekhuyzenChannel I have only one device what can decode and render MQA files (Arcam Solo Uno) and from my experience MQA is a little bit quieter, specially with 192 kHz files but its sounds like it should in my opinion. I feel like everything is more separated inside and i like this
Sir you are the real deal 👍
I try
Good video. Undoubtedly the big difference is in the way the sound was recorded. Then there's the capacity of each hifi system and the acoustics in the room where that system is.
True high dynamics must be very complicated, but recordings should have the best, most suitable dynamics possible for each format. High dynamics helps in the perception of sharpness and 3D. (left right front and back)
I dare to say that 80% of my music collection sounds fine on my setup 1a. Perhaps there are more good recordings than good stereo's in the world.
Even if they sound exactly the same on your cheap mp3 player. If you have a 1tb sd card. Or for the car 1tb usb flash drive. And your music 16/44 takes 200gb. You have 800gb free anyway so why not just put the hires files? And drive will get cheaper every year. In justa few years you will buy 1tb sd card for 50 euros. Against a 250gb sd for 30 euro.
Does that 20 euros really make a difference in peoples lives?
@@vitorfernandes651 Sorry but I didn't understand your answer, I don't even know what the reason for it.
Thanks!
Welcome!
Thanks much for explaining something that i was struggling with the past year. Not understanding way 192 or 96 did not always sounder better than 44 KHz. Had already came to a conclusion 44 KHz sounded more balanced and had my preference ween recorded well
👍🏻
Köszönjük!
Örömömre
Great explanation of the sense and non-sense of Hi-Res!
🙏🏽
Thanks for the lucid explanations. The mechanism you explained in the latter part of the video on how a higher res music file can sound better on a DAC should be objectively testable since it's not a subjective perception but rather about a measurable loss in sound quality, right? There are many tests measuring how transparent are different DACs, and some of the cheap modern DACs do score very well. I also don't know how much of these differences are audible to an average listener.
It's about time they realise time smearing can not be measured with available measuring equipment. People say that they can, don't understand what's involved.
Thank you Hans!
I know you like the MScaler and Roon. I would be very interested to hear your view on Roon and HQ-player.
As you say, many Dacs lack refined upsampling due to Hw limitations. A modern PC is very powerful. If you use this to upsample and stream through a quality streamer I wonder if you could get really good upsampling for less.
Is this something you are interested in reviewing? It could bring good quality for the masses.
I have bought a licence a year ago. But setting up when combining with Roon was not something I would advice my average viewer. You need extra hardware and thus extra connections. Even I like to avoid too much complexity But, provided you experiment with the settings to find the perfect setting for your DAC, you can gain audio quality.
👍 Thumbs up for HQplayer upsampling. Big improvement I think
Awesome.
Thank you! Cheers!
Hi Hans, awesome Video as always.
There are two things I find hard to follow:
1. You speak about the bad sound of steep filters, but then mostly about playback. But isn't the damage done on the recording side? As the steep filter has to be applied before the D-A process to avoid aliasing in the sampling process?
So how can oversampling in any (mega-ecpensive) dac fix that issue? Since the damage has already been done.
2. How does it take a super expensive dac to do oversampling properly? It should be a mathematically a rather simple process, that a modern CPU should be able to handle below 1% of load.
When I playback a CD file from my PC and the audio-link to the USB dac is set to 192kb/s isn't windows upsampling on the fly anyways?🤔
According to Shannon/Nyquist the Waveform of a CD file should be perfectly defined and adding in 3 more samples should really be a pretty trivial task for a computer in my humble opinion🤔. As far as I know, most DACs have a native Sampling rate anyways and all incoming signals are internally upsampled to that rate anyways, no?🤔
Just my two cents
1) A/D converters all are of the delta/sigma type. The output is 1 bit at a very high sampling frequency. That data then is decimated to PCM where the filtering is done in the digital domain.
2) Doing oversampling really good is rather difficult. I simplified it in the drawing. The result would be straight lines between the sampling point as where you want a true reconstruction of the wave form. That involves a lot of calculating work and then needs to be followed up by proper dithering. Just listen to a good scaler on a good stereo set and you will herar why.
I no longer worry about hi res. The quality of the recording makes the difference. Plus Dac are delta sigma decoders, far from PCM whatever.
So I stopped paying more for so called hi res.
Whatever pleases you
I concen myself with what I can hear, not what others say they can hear, or what machines say I could possibly hear. More than anything I enjoy music.
So true
Hey Hans, what's your take on HQPlayer?
Is it okay to use Volume Leveling in Roon, will it affect the dynamic range or sound quality in any way?
Volume leveling in Roon wrecks the sq and dynamics. But it all depends on the quality of your system. The more sensitive and hi end a system is, enjoy the music without digital manipulations. In my experience the only files that truly benefitted from hqplayer are mp3's. Full res files again feel Artificially processed. Just my two cents worth.
I use volume levelling on my setup 1a. It will reduce the level on loud tracks and thus adapts the recording to your system settings. I currently hear no difference in sound quality.
@hans, I would think 16 bits is 15*6dB dynamic range. 15 bits for the positive amplitude, 15 for the negative.
In a balanced system you could say that.
Great explanation, though I may have not understood everything you said. Does that mean that if I rip a cd and the software would oversample properly lets say to 96KHz that the result should be superior to a CD on a normal CD player? Software doesn’t require an expensive chip, it may require computing power. Thats no more a problem these days
Oversampling does not increase the sound quality in the file but it makes it easier for the digital to analog converter to do the conversion.
For editing, 24bit or higher is likely important.
16bit 44.1khz FLAC for consumers is fine.
I have a question, though.
Imagine we have one song encoded in wav, 16bit and 32bit maybe flac.
Open the wave file. Find the third sample, for example. What's the number stored and way?
Open the 16bit file. Find the third sample. 16bit. What is the number and why?
Open the 32bit file. Find the third sample. What's the 32bit number and why?
After you have the binary numbers, then what is dynamic range, bit depth etc.
And how the two binary numbers, one 16 and the other 32, are both lossless.
Not one youtube video explains any of this, so I am still confused.
That number depends on the music
I’m confused by the offerings on Apple Music. The most common format available above CD is 24/44.1. Why is this? It’s almost a settled standard for some reason.
The most unusual thing I’ve seen lately is the sampling frequencies on the latest Madonna releases. 176.4kHz. Why would they choose this?
176.4 is four time 44.1. Producing in 176.4 makes it easier to downsample it to 44.1 kHz for CD release
Agreed More you can do with the data. dose not sound better though to my ears. I personaly Dont care about super res audio and dont plan on testing it outside of a large speaker setup not that I could hear a difference Im a little bit deaf at this point in my life. Love the topic of the video. Thank you for the content Best regards.
🙏🏻
Recording in Flac gives all the information. Like a camera uses Raw rather than Jpg or JPEG
And so does WAV and IAFF.
One thing rarely mentioned when discussing higher sampling rates is how the analog part of the reproducer system will handle the higher frequencies. Is it fair to say that most power amplifiers and loudspeakers are designed with a 20 Hz to 20 kHz range in mind? Then what happens with the higher frequency content potentially output by a 96 kHz or 192 kHz DAC? Will it be filtered out (lost), or worse, will it create all sorts of lower frequency harmonics due to unexpected vibration modes in the speaker or the enclosure etc.? HiFi speakers are not ultrasonic transducers so unless the whole system is designed to reproduce a larger frequency range, pushing the frequency response of the source might actually be detrimental to sound quality.
What happens with higher frequencies in the amp, depends on the amp. There are amps that will filter out anything above 20 kHz and there are those that will let it pass. There are also amps that will get unstable.
It has been a long time since I encountered 20 bit audio, but where it comes to 24 bit and DSD there does seem to be something that is still mysterious or unknown. I always notice that reverb tails are better resolved at 24 bit than 16 bit, and essentially perfectly resolved at DSD. Yet, objectively those reverb tails are well above even a 16 bit noise floor. Neil Young commented on this too. Somehow, it appears to me that more digital resolution is required to avoid masking well above the noise floor, and DSD manages to sidestep this problem altogether.
I will say critically, I use a high end DSD 16x upsampling DAC and I can still hear a difference between sample rates, though it is indeed much smaller. Any upsampler/oversampling also has to use steep DSP filters, which also cannot be perfect even if they avoid the problems of an analog brickwall filter.
Chances are that ADC's used for 24 bit recordings are of higher quality ....
I have a Qobuz subscription being used through a BluSound node. But I have streamed several newer albums from this service in quoted Hi Res audio. I admit my ears are probably not the best, but to me some of these didn’t sound as good/ great to similar at CD quality. I often put this down to way it was mixed. I have also tried so called Hi Res on albums originally recorded back in the 1970‘s and obviously to digitise it would have been re mastered definitely don’t sound great to my ears, rather just used CD’s quality on these…
I don't doubt what you have heard, but that can be the mastering or your hardware....
I was hoping for some thoughts on DSD, I use a March DAC, anything below 88Khz PCM, I convert to DSD256, using Roon. I am more than happy with my aural experiences. DSD is coming very close to my LP sound quality, if not bettering it in some cases.
I already published several videos on DSD: ua-cam.com/video/hXFIq11JAas/v-deo.html, ua-cam.com/video/VyrDXUkOybY/v-deo.html and ua-cam.com/video/5NREJTmt180/v-deo.html
5:40 I tried explaining this just a few days ago. Was told dynamic range is irrelevant to listening and background noise levels. Smh ☺
And Theye were kind of right. What I showed is the technical approach. In the end it is the dynamic behaviour of the music that does it.
Should you use an audio normalization or not? I'm genuinely curious as I'm just getting into audio and high res audio
I have it switched on on my Roon system. Unless I do review listening, just to be sure.
Great video. People are seduced by "newer, bigger & better". A similar comparison would be modern TVs that boast thousands of colours- far more than the human eye can actually see- what's the point? It's mostly advertising and marketing. Another is the 70 million tracks available on your streaming service.... I wonder how many tracks the average subscriber accesses?
We'll never know
If a human could be described as "reference", you Sir are it.
☺️🙏
Well these sampling rates have mostly sense in recording/mixing phase as you are in DAW processing whole project with various of effects, but final mixdown higher than 44.1khz will appreciate only people with DACs, amps and speakers of exceptional quality, and definitely not people listening music with Bluetooth wireless ear plugs with brutal lossy compression. Or, if you have one of that "fantastic" players with hardwired 48khz, then 96/2=48, so at least music will play somehow normal and frequencies will not be screwed up 😀
If we take the lowest use cases as a reference in everything we do, it's going to be a very sad life.....
Nice vid. I am stuck at 44.1 basically. None of my music will ever be recorded at hires. I have thought about upsampling the files in a computer and playing those files but what a time investment that would be for little gain. I have hardware that will do 192k but havent heard a big difference with upsampled files. Have an old dac that upsamples everything to 192k. I will hook it up and see if it makes a diff. But the new dac is probably better anyhow.
Hard to say for me. But DAC's have become a lot better over the years.
@@TheHansBeekhuyzenChannel well i hooked it up and did my best to level match. I can switch back and forth quickly. Couldnt tell a difference. Old dac upsamples 4 times and goes to 24bits. Latest dac is R2r non os. Latest is fed with spdif from the Apogee Big Ben and the new one is fed using AES. After the dacs my signal chain is purposefully distorted: tubes, bass boost, crossfeed, soundstage expander, etc. so its probably not the best for listening to differences. Fwiw the bass on the R2R sounded a bit more robust. Thats the only difference I heard really.
Hi Hans, in this video, you said you doubt that 24-bit and higher sampling frequencies contain more relevant information. I understand your explanation about bit depth but on the higher sampling frequency, you mentioned filters and how they are a problematic issue with lower sampling rates. In your earlier video "MQA part 1; Why 24 bit 192 kHz" audio? you have talked about how choosing a higher sampling rate does increase the resolution in time and that is correlated with our hearing system (since the 44.1 gives a time resolution of 22.7 microseconds and our auditory time resolution is about 7 microseconds.) And only 192khz offers sufficient time resolution for our auditory system. So my question here is, do you still consider this point relevant, and is that still one of the important factors in choosing 192khz over lower sampling rates in your opinion. Thank you
I know and that is a theory that makes sense. But in this video I just listened. Surprised by how well my setup 1 does 44.1 kHz tracks. So I compared it with Hi-res, the result is in the video. Still there is much to be discovered. How come that MQA files sound better over setup 1a (with the Grimm upsampling and the non-MQA Chord DAVE DAC) sound slightly better than non=MQA tracks at the same sampling frequency? There still is a lot to learn and I might have to come back on opinions all the time or at least question them.
It will always amaze me how far ahead and future proof the CD was in 1982. Not before the year 2000 did people have PCs in their home that could hold more than a handful of full quality CDs, and even then storage was at such a premium that we got this mess with mp3 that we are in. Meaning that most people make due now with quality audibly worse then CD, of not on their device, then what reaches their Bluetooth ear buds.
With SACD the industry completely botched it by pushing the hardware prices way too far out of consumers reach. You can't have the CD be the standard and then it's successor moves back to cutting edge prices that were asked in the early 80s. DVD and BluRay did it right, SA CD did not need to fail.
It's a real shame too, the mainstream consumer may never experience high res audio.
The majority of people don't care too much about sound quality. The CD became popular because of convenience rather than sound quality (initially it sounded awful). SACD offered the same convenience at a higher price so the mass market was not interested while for such a product the boutique market is too small to make it economically viable. The same can be seen with Tidal/Qobuz versus Spotify. C'est la vie.
Hello Hans, thank you very much!, I have a very quick question, my DAC only shows the sample rate without the bit depth of the file that it is processing, does this mean that it does not take into account if it is 16 or 24 bits?, or does it run with some calculation and modify the sample rate thats is showing?, thanks again.
I can’t say for sure but modern DACs do accept 24 bit signals.
Hello, a slimmer and fitter version of Hans! 🙂 Everything you've said seems to make sense, except that from what I know the most advanced filter that Chord uses (Dave + Chord MScaler) is pretty much a brick filter, i.e. a very intensive/steep attenuation at/near to the Nyquist frequency. If one of the main benefits of oversampling is that it allows for filters to be less steep, this is certainly not the case here, yet the Dave appears to be your choice of digital to audio conversion done right. What am I missing here?
Chord upsamples to 'DSD', then a sharp filter is not neccesary because of the high sampling rate.
@@TheHansBeekhuyzenChannel Thanks for the reply, but I think that you're not only missing the point but you're also wrong about Chord.
1. I meant the exact opposite - Chord's long 1M tap filter, used in the MScaler, is a very steep/sharp one, with a very aggressive attenuation of ultrasonic frequencies. Which I believe is what it should be like going "by the book", i.e. by the theory behind digital to analog reconstruction. According to many, including yourself, Chord Dave sounds fantastic. Yet this seems to contradict your claim in this video that great sound is a result of less sharp/steep filters being used after upsampling.
2. From the research that I've done, it appears clear that Chord MScaler upsamples to PCM, not DSD, and even if you feed it DSD it will be converted and downsampled to PCM.
So we can take DSD out of the equation here because it's got nothing to do with Chord Dave's great sound (I haven't heard it myself, but most people who have swear by it, including yourself).
And Chord does not use upsampling to create less sharp filters - precisely the opposite, they use a high number of taps to ensure that no ultrasonic artifacts bleed into the audible range.
This seems to challenge some of the claims you made in this video. I'd suggest that perhaps great sound has more to do with the quality of algorithms used in the upsampling process.
Well I thought the true benefit of higher sampling rate was the higher sampling rate and not the added frequency bandwidth. Added resolution, if you will. Since the analog to digital and digital to analog conversion would have smaller steps, the differences in the frequency changes would be more smooth. It's like skiing down a slope rather than down the stairs. I thought that was the true benefit. But since you didn't mention this at all, did I misunderstand something? Or did you?
Nyquist says that bandpasses at half the sampling frequency, ALL information within that bandwidth is kept. Therefore it must be the band limiting filtering that is the problem.
Good stuff. I run my RME at 192 works great because I can keep the daw at whatever, usually 44.1, yet my RME DAC stays at 192, so in effect it is upsampling before reaching my ears ;)
👍🏻
On the subject of bit depth and dynamic range, you mention about the extremities of the ranges for different bit depths, but I've often wondered if higher bit depths allow the capture of more detail in subtle sounds. Would higher bit depths give you more steps in the sound level resulting in higher accuracy to the subtle sounds. I'm thinking of quiet sounds such as musician breathing or subtle background noises. Any thoughts on this?.
in theory it does but in practice low amplitude levels are not used for anything else than noise.
High bit depth is often used to keep noise floor down for boosting very soft signals after the fact
It has more value in before mixing and mastering than it does after
Hi Hans, please forgive me but you seem to have lost lots of weight, I pray your health is good and this was a conscious decision
HiHi, I lost 17 kg by keeping to a strict diet and exercise. I never felt better. Thank you for caring though.
So, for me, what this means is that the AudioQuest Dragonfly Cobalt would be good enough for streaming from my PC, since I'm 67 years old and my upper limit hearing max's out at around 12000khz.... would that be correct? Greetings from Ontario Canada!
No, absolutely not!! Watch: Hearing loss at age, the good news
ua-cam.com/video/JxFCnZ5pBp4/v-deo.html
For equipment like headphones or all-analog receivers, does the hi-res logo matter?
Well, it. can provide higher sound quality....
I have done testing on my frequency response with my Topping E50 DAC and HifiMan Edition XS headphones that can reproduce sound in 8-50 000 Hz range with an online tone generator. I and my brother could hear around to 12 Hz and up to 23 KHz. We did repeated tests and same conclusion.
This made me look deeper into the topic, and I got confirmation of the 20 Hz - 20 KHz myth, and how it was made the standard based on limited research.
This myth cannot be the truth as I am not the only one that has experienced this. I often hear sounds that other people do not hear, especially high frequencies.
Havo you any idea how many octaves above 20 kHz your brother hears?
@@TheHansBeekhuyzenChannel Is the base of the octave in this case 20 Hz? In that case, he and I hear 1/5 of an octave above 20 KHz. I forgot to mention that I am 21 and he is 18, so we're quite young and less affected by hearing loss.
Totally agree with your story Hans, on a good rig a 44.1/16 (and well recorded) sounds wonderful, but DSD is glorious. When you listen on the player in the car, the kitchen or the bathroom, HiRes sounds better…
Again, it depends on the gear. The better the gear, the smaller the difference.
can someone list the points in sound quality that hi-res tracks sound better at?
Again, it's all about the properties of the gear.
Thank u 😊
Welcome 😊
Looking good hans, looks like u lost stone.
Actually, Iost 2.7 stone😀
So is it wrong to think that a high res file has more density of information in the audible spectrum ? Because some people say high frequencies can be felt but not listened to and to me this is no sense
That's correct. The density might be higher but there is no extra information in it. It's like fifty ten cent pieces are equal to a five euro bill: it looks like more but there is no more value.
In my setup, curiously 16/44 sounds almost always better than 24bit audio, which is softer, I would even say "blurrier" (we're speaking of small differences actually). 16bit is faster, more dynamic, clearer. Maybe what I listen to in 24bit is not native and is interpolated?
Anyway, cd playback in my various setups succeeded in the years has and still is a step way above in sound quality to playing files, no matter if standard or hires, so this erases any supposed advantage of hires files (again, this is my experience).
It's all about the gear...
I've heard God, no sarcasm. WOW!
😇
96 kHz is more than enough for me.
Enjoy the music
Hi Hans, as usual I like your video and topic. However I think I disagree with the physics - please correct me if I go wrong somewhere.
Sound is just air that is moved
Sample rate indicates the number of samples you take per second
Bit depth represents the amplitude of the air being moved.
So therefore both a higher bitrate and sample rate get you closer to the original.
Yes, you need a higher bit depth to get to a low noise floor, but you also benefit from smaller intervals at louder passages.
Similar with the sample rate - you take more measurements in a given time interval, so on average you will get closer to the original.
Looking forward to your reply.
@wouter kolkman, I think Hans addressed both your questions but I might be able to say something about bit depth as I have done a fair bit of data acquisition using different ADCs. Essentially Hans was explaining that bit depth, past a certain point, when converted back to a voltage either meant that the range was unusable (too high a volume), or in other videos , that the change in the LSBs (least significant bits) would be within the noise of the amp (inaudible). Where high bit rate was useful was in convertions so as to reduce rounding errors.
@@dowster64 thanks Steve:-)
A 44.1kHz sample rate 'perfectly' samples a signal which is band limited to 22.05kHz (just above the limit of human hearing). It is not an approximation so there is no need for higher sample rates. It is called the Shannon-Nyquist theorem.
Correct, provided the band limiting is done lossless and that is rather difficult.
@@TheHansBeekhuyzenChannel Is that just speculation or do you have evidence that this 'difficulty' is audible in any particular material?
i want to buy ifi zen hi res audio bluetooth reciever can i get the benefit of hi res audio using android if the music source is from youtube?
I haven’t reviewed the iFi so I can’t say.
I find that concerts and films that are too loud essentially lost their dynamic range and sound compressed.
I agree completely.
@@TheHansBeekhuyzenChannel Director Christopher Noland wants his films played exceedingly loud. At concerts, I believe that the artists are trying to excite the crowd. They bring their own mixing person to do this. I used to complain about venues' sound systems but found out they have perfectly good systems but the mixer f'ed up the sound on purpose. Some venues, like the Hollywood Bowl, allow the artists to bring in their own mixing console.
How do Radio Paradise manages to sound better then almost all streaming services ?
Do they?
@@TheHansBeekhuyzenChannel do give them a trial and you would be surprised by the soundstage and seperation offered by Radio Paradise flac stream and even 320kbps stream sounds good.
I remember reading that higher res files are better at reproducing the interaural time differences humans are so sensitive to.
In other words, humans are capable of discerning ITD as low as 7-10 micro seconds, but 44.1 is only capable of ITD of about 22 microseconds.
Therefore, higher res should be able to reproduce things like: soundstage, imaging, ambience, spatial cues better, since those things are so dependent on reproducing lower ITD, i.e., as low as we are capable of hearing.
I know, you might even heard it from me. What I did in this video is put all theory aside and just listen. It appears that true high end equipment, like my setup 1a, seems to not suffer from this or do all other things so well that it doesn't matter. I really can't explain the phenomenon.
This is a common misunderstanding of digital audio. Temporal resolution is not the same as the inverse of sampling frequency! It is determined by signal frequency/amplitude and bit depth. Signal can be shifted less than sample point distance by adjusting the sample points a little bit (like having taken the sample points earlier or later). In practise 44.1 kHz 16 bits digital audio has much more temporal resolution than needed.
In your video and even more in the comments I find a lot of simplifications, which are incorrect. E.g. The dynamic range you mention is measured broadband , maybe with A-Filter, but that does not tell us too much about the ability of hearing. In my advanced age ( I am V 7.0 already) , the limit for pure tones is lower than 10 kHz. BUT due to a life long experience at listening carefully, my ability of listening into details is still very good. AND to decipher room information we need much higher resolution than 48 kHz. As always people forget the time and concentrate on static frequency diagrams. Our ear/brain system is capable of so much more than the mentioned measurements tell us. Think about a very low violin tone and its overtones in the background of a room. To hear that including its correct position, needs a lot if HighRes.
You start questioning my statement on dynamic range and then try to argue against it by describing things in the frequency domain.....
Based on that reply I would assume your an MQA fan then !!
I prefer listening to 24/48 or 32bit DXD. Those two offer a good balance between the directional expanse and faded depth & cohesion of the sound image. 24/92 requires crossfeed to sound decent in my opinion, at least with headphones.
It all depends on the quality of your equipment.
I liked Depeche Mode, Peter Gabriel, Queen, Beck - Sea Change, and Blade Runner soundtrack in DSD. I've heard some samples of tracks recorded in actual dsd. Just not a big enough catalog in the music i like to care about it. Upsampling my entire library to high res pcm or dsd made no difference for me. Most of my 96k or 192k was simply not that special for me to care anymore so i let it all go. I'm not into classical or jazz and i'm not listening to "audiophile" music all day. CD redbook is enough for all the music i like.
👍🏼
I like my tda1541
Don't t hey say that love is blind?😂
I don’t understand why people don’t just get the 24/96. Example. I have 1 terabyte ssd dedicated to music. Even 2 terabyte hard drive are so cheap these days. I had my music in 16/44. Which meant I had 350gb of music. So 650gb of wasted space
Now with most music in 24/96 I still use the same drive. Didn’t spend a single dollar to upgrade and the driver still has 100gb free
Moral of the story. For no money at all I get to stop worrying about it. Doesn’t matter if I can hear the difference or not. All my devices play these files.
It would have made sense 20 years ago even drivers were expensive
One thing this gentleman forgets to mention. The biggest reason for hires is not if you can hear or not. It’s the mastering. Old albums get new mastering. Sure a lot of álbuns get remastered on cd too but most don’t.
I haven't mentioned the storage space as an argument.....
I don't know about all the technical stuff. And you do make a lot of sense when talking about dynamic range. But all I know is there are CLEAR differences with each quality level.
MP3's at 128 have become virtually unlistenable to me. MP3's at 320 is what I would call the bare minimum threshold for quality music. And when listening to 16bit or 24bit I can tell a bit of difference. I like to explain my experience as....the more bits there are the more "space" there is for the instruments to play in. There seems to be very little muddying of the instruments. The guitars have their range, the bass is in it's range and rarely do the 2 ever get muddied together.
Anyway, that's just been my experience with updating my music library to hi-res. And the better the quality the easier it is for me to separate the instruments from each other.
It's all equipment dependent.
@@TheHansBeekhuyzenChannel can say that about any medium really. If your on vinyl then a cheap needle can ruin the experience. If your using a CD walkman then your CD experience can get ruined.
long story short...if you are NOT an audio geek (an audiophile doesn't count as audio geek because an audio geek actually knows how science and physics works) that knows how every record that plays is recorded and know how the signal path of the recorded instrument is processed and affects the sound until the master...stick to 44100 or 48000, no point to go higher than that, specially with music done, recorded and mastered entirely in the digital domain this days. Only reason to go higher is if you work with audio or dj where further audio processing and high SPL is applied....BUT in the other side if you are an audio geek that knows how to hear and knows how the records are recorded, then will be easier to choose the right high sampling rate audio that will sound VERY good in the giving listening system and strike...AUDIO NIRVANA!!!
.
I find 24/96 more smoother but it didnt change my life.
As said, it depends on you hardware.
Is bitperfect real? I listen to you and I begin to discover filters and manipulations even below my bed
Even the first Philips CD player did upsampling so there is nothing new. The problems lie in computers always converting audio files to the sampling frequency they work on for audio. And they do it with 'sufficient quality' which in computer terms means ' you can still recognise the music'. That is why you need bit-perfect players on computers. When those programs do volume control, the impact of the sound can be lower than that of a potentiometer based volume control in your amp.
@@TheHansBeekhuyzenChannel thanks. So please correct me if I am wrong. I have always thought that you have to put volume in soft at maximum level and then control volume in the last step (the amp). Do you mean the opposite is true?
At 96KHz the difference is on treble frequencies above 10KHz (more detail and brilliance) - at 44.1KHz the treble frequencies are muffle (especially at 16KHz range)
I’m sure that’s what you hear on your system. But it’s different on every other system.
We love to pretend we can hear beyond 20/20 but the truth of the matter is we don't have bat's ears... ;-)
So true
‘The wrap’ confused me. You start by saying the mastering is more important, so a 44.1khz track may sound better than a 92khz track. But the best investment for audiophiles is high rez music. Are you saying that because hi rez tracks are likely to have undergone a better mastering process?
No. If the recording,, mastering and mixing is all done perfect, 44.1 kHz on a true high end system sounds about equal to a 176.4 or 192 kHz copy. But the latter two sill sound better on less-than-true-high-end equipment. If recording, (re)mastering or (re)mixing is done poorly - mostly advertised as Remasters - the original 44.1 kHz CD copy might sound better.
@@TheHansBeekhuyzenChannel thanks for the reply and please keep bringing us this great content!
@@TheHansBeekhuyzenChannel And I thought that Remasters meant that the recording was lovingly improved to the highest standards. Maybe it is good that most of my music comes from pre-1990 CDs.
The 44.1khz track says "thump up" but 192khz rate says "thumb up." 192khz is better.
That says all about your hardware, as I explained.