[mbirand]

16/44.1 covers the full range of human hearing, so you won’t be able to hear any difference between that and higher sample rates or bit depths

thank you very much for the kind explanation (probabaly for the thousandth time). please let me ask this; is there any use case (as a consumer) where higher bits/rates are useful and possibly could make a difference ? or there is absolutely none whatsoever..

can i conclude then the audible differences, clarity, detail i hear with hi-res music is due to its mastering, rendering it somewhat louder and applying certain filters to highlight some details therefore losing fidelity on the other hand ?

there are some companies for example recording in dsd’s or super hi frequencies and selling them as pure dsd etc… can we call this also some sort of milking the audiophool ?

when you’re shopping for music to download, and when you’re given choices from pcm/dsd, mp3/flac/wav/aiff, 16-44/24-96/24-192/1-64/1-128/1-256/1-512 with very small differences in their prices, and sometimes 24-96 being much cheaper 16-44 (i find this on qobuz), what do you choose ?

I think many audiophools really don't know which one it is! They probably can't tell “satisfying your psychology” apart from “to listening to the best possible music with the best possible tools.”

as an audiophool newbie, i totally agree. there’s not much clear info (that a person like me can easily comprehend) out there. so please bear with me without judgements. i truly want to understand this, but i also don’t want to miss out on anything (SQ)

 

[bigshot]

I think if you have an interest in a subject, you'd want to do a little research into it. (Perhaps not with ice cream, but that's different!) Digital audio isn't that hard to understand.

The audio business is even easier to understand. CDs are designed to contain all the sound you can hear. It's the format hifi nuts have been dreaming of. It comes out and *almost* everyone is happy. The ones who aren't are the ones who argue tiny exceptions to the rule that don't matter in practice, and come up with hair brained theories why CDs sound *may* not be enough *in theory*. An audio salesman comes along and sees an opportunity to sell them something with these particular I's dotted and T's crossed. Everyone's happy now, right? Nope. The process begins again with someone dreaming up another solution to a problem that doesn't exist. It repeats and repeats. Reams of sales pitch is generated defending the unnecessary stuff. It gets complicated.

But it isn't complicated. CD sound is all you need. Your DAC, DAP or player is producing all the sound your ears can hear. There's no reason to fuss.

The more you know about how digital audio works, the more you realize that 16/44.1 is all you need. In fact, high data rate lossy does the job perfectly for human ears too.

Listen to the music, not the machine.

 

[Vamp898]

I think if you have an interest in a subject, you'd want to do a little research into it. (Perhaps not with ice cream, but that's different!) Digital audio isn't that hard to understand.

The audio business is even easier to understand. CDs are designed to contain all the sound you can hear. It's the format hifi nuts have been dreaming of. It comes out and *almost* everyone is happy. The ones who aren't are the ones who argue tiny exceptions to the rule that don't matter in practice, and come up with hair brained theories why CDs sound *may* not be enough *in theory*. An audio salesman comes along and sees an opportunity to sell them something with these particular I's dotted and T's crossed. Everyone's happy now, right? Nope. The process begins again with someone dreaming up another solution to a problem that doesn't exist. It repeats and repeats. Reams of sales pitch is generated defending the unnecessary stuff. It gets complicated.

But it isn't complicated. CD sound is all you need. Your DAC, DAP or player is producing all the sound your ears can hear. There's no reason to fuss.

The more you know about how digital audio works, the more you realize that 16/44.1 is all you need. In fact, high data rate lossy does the job perfectly for human ears too.

Listen to the music, not the machine.

100% agree _but_, and i can't stress this enough, often, Hi-Res Audio Recordings are Mixed/Mastered different and do sound significantly better due to this.

Of course they would sound as good if you downmix them to 16/44.1, they don't sound better because they are Hi-Res Audio. But you have to buy the Hi-Res Audio Version to get this better Mixed/Mastered Version.

Sadly this is not constient. Sometimes the quality of Hi-Res is identical to CD Quality, sometimes not. There is no indication when buying (at least at the portals i know) if it sounds different or not.

But again, this is not due to the higher Quality of the file, this is due that the content is different. But sometimes when you buy the Hi-Res Version, you also get the AudioCD Quality/AAC Version and can compare then and there are sometimes drastic differences, especially in the bass.

The Bass of the non Hi-Res Version often is mixed much stronger and sounds muddy while it is, generally speaking, much faster and better controlled in the Hi-Res Audio version. Also those songs often have mauch stronger/better Stereo separation.

So there are reasons to buy Hi-Res Audio, but its unrelated to the theoretical higher sound quality of 96/24.

More and more Studios even use separate Mixes/Masters for Streaming Services. So somtimes there are 3 levels of quality.

Streaming Service <--> Audio CD <--> Hi-Res

Best example is chu tayousei from ano. YouTube Music Version, Audio CD and Hi-Res all sound different.

But the Streaming Services are not consistent either. YouTube Music might have a different Mix/Master of the Song than Qobuz (which sounds identical to the Audio CD Version) and so on.

So the only way to ensure you have the highest possible quality is to buy Hi-Res Audio, well knowing that you could get ripped off and there might be zero difference to the other versions.

Its a mess and its annoying... i've burned money buying Hi-Res Versions that sound identical, but also bought Hi-Res Audio Versions very late thinking there would be know difference anyway, ending up with a much better version of the Song years later wasting my precious life time listening to an inferior version.

Since then, i always buy Hi-Res Audio well knowing that there is an almost 50% chance that im just wasting money.

 

[Vamp898]

16/44.1 covers the full range of human hearing, so you won’t be able to hear any difference between that and higher sample rates or bit depths.

The danger with satisfying your psychology is that with all the audiophile marketing BS you are in many cases effectively in a contradiction. Do you want to “satisfy your psychology” OR do you want “to listen to the best possible music with the best possible tools”?

As with pretty much anything, there’s an optimal amount, beyond that there is no difference and way beyond that it’s probably doing more harm than good, although it probably won’t be audibly worse.

Oversampling had a couple of advantages, the main one being that it relaxed the requirements of analogue anti-alias and reconstruction filters. It is NOT something to be avoided, in fact the opposite. These days most professional ADCs (Analogue to Digital Converters) operate at 512x oversampling, there are some benefits as oversampling also reduces digital noise but these benefits are relatively tiny (and are well below audibility). The DACs to actually avoid are the NOS DACs, unless: You prefer more artefacts/less fidelity (which is audible) and/or you want to satisfy a (false marketing driven) psychology!

G

Actually it doesn't, but the difference is inaudible. The ear could in theory be able to percieve an dynamic that is larger than 96db, but you would have to listen, obivously, at louder volumes than 96db (which you should not do at all) and even if, you would need recordings that utalize that (pretty much impossible) and even if they would, just because your ear could percieve it, you would not at all enjoy it.

Just imagine an 100db loud sound and then an 0db loud sound in the background... who would ever notice that and what qualifying information would be in that?

So in theory, the human ear is better than an Audio CD, but in the real world, it never matters.

 

[Vamp898]

excellent initial post with great explanations. Now my simple question is if it makes sense using a DAC AMP Stack for the laptop when listening to CD-rips in 16bit? or is it enough to just plug in the headphone to the audio port of the laptop?

Electricity is very simple.

V = R * I

That means if the resistence is the same and the voltage is the same, the volume is the same. So in theory, if you have the same voltage, you should have the same sound.

But, and here comes the but, the output resistence of nothing is linear. Resistence is generally measured at 1kHz but grately varies and especially with digital amps, the output resistance of the capacitors play a big role.

The better the capacitor, the more linear is its output resistance, but no matter what you do, its never flat.

That means you need a lot of capacitors and good capacitors if you want an as flat as possible output resistance. That is why the second generation Walkman for example uses 21 wound capacitors that were developed over years.

There is an explanation from Panasonic in details how temperature and capacitor design effects its output resistance and so the sound

Here an small screenshot

You see, the type and design of the capacitor and especially how many you use and how you implement them does heavily influence the output resistance of the capacitor and so the sound of the overall amp. Capacitors are not(!) linear, even when they are marketed as such.

And you also see the mark "25°", that is because this value changes with temperatur. That is why some amps start to sound different when they get warm. Also this value changes with the amount of power you draw from them. So the higher your volume, the more does it change the frequency response.

More informations are here: https://industrial.panasonic.com/ww/ss/technical/b2 but this is just scratching the surface.

As capacitors are also used in LC-Filters, the size and implementation of LC Filters further change the sound.

As a rule of thumb (important, this is getting less relevant over the years as technology advances), if you want good sound, you need big components. You want big wound capacitors and a lot of them and large LCs. But as they will never fit into your Notebook, your Notebook will sound different.

That is why Digital Amps like the Walkman have so many big capacitors and large LCs, even in an portable player

But in theory, there is more space in an Notebook to implement these things than in an DAP, so an Notebook could sound better than an DAP, if the Amp section is shielded good enough, just for the sake of more space for good components.

The Mid-Level DAPs (ZX-Series) from Sony uses different capacitors for 3.5mm and 4.4mm (to cut cost) which is the reason why 3.5mm sounds different than 4.4mm on Sony Walkmans. Most people mix that up and think this is due to the advantage of an balanced connection while the WM1 Series, who uses the same capacitors on 3.5mm and 4.4mm, sounds pretty much identical. So its the choice of capacitors that causes the change from 3.5mm to 4.4mm in the ZX-Series which shows, how big the influence of the capacitors actually are.

A lot of people think that Desktop Amps sound better, because they have more power and the more power you have, the better you can drive an headphone. As you seen in the formula above, that is nonsense. Desktop Amps driver Headphones better because they have better components and more space to properly implement them and only for that reason.

And 3000mW Desktop Amp wouldn't sound any different if you would limit its power to 250mW. So the choice and implementation of the Amp and its capacitors defines how good the amp can drive an Headphone.

 

[bigshot]

100% agree _but_, and i can't stress this enough, often, Hi-Res Audio Recordings are Mixed/Mastered different and do sound significantly better due to this.

Mastering is completely independent of the data rate of the file. I've seen hires files that had tons of noise above the audible range that could cause distortion in the audible range with certain equipment. And I've seen lossy streaming copies that are identical in every way to the audiophile release of the same title. You can't judge audio quality by the data rate. The only way to tell is to listen to seasoned collectors who have heard multiple versions of the same album and follow their advice as to what the best one is. That may be a hires copy. It may be a Japanese or European release. It may be an earlier standard release. And it may be the currently available remaster on CD. It may even be that all of the various versions sound the same.

There are advantages to the various formats, but they are functional, having nothing to do with sound quality. SACDs are capable of multichannel audio. CDs and SACDs start playing immediately and don't require you to navigate menus to play the music. Audibly transparent lossy is smaller so it takes up less room on mobile devices or it can be streamed without transcoding. CDs come with booklets. Download files don't take up room on a shelf. Those things clearly are advantages. But there are instances where each one of these formats sounds fantastic. You can't judge by numbers alone.

 

[castleofargh]

hat said, looking at my dap sp3k which has quad totl dac chips onboard probably for the purpose of oversampling like crazy

DACs with multiple chips do it for 2 main reasons:
1/ if they use 2 different chips to handle DSD and PCM.
2/ if they just try to get that little extra increase in specs. Nowadays, unless your DAP relies on antiquated chips and not greatly popular kind of analog doubling of samples(one DAC delayed by half a period), it's almost certain that multiple chips are not for oversampling. Not even in the ultra loose sense I just suggested.

As for a DAP, the amp section is almost always the limiting element(limit in noise floor, limit in power, increase in distortion and crosstalk based on the IEM/headphone it has to drive). And then of course the headphone/IEM will usually bring the fidelity down the most, along with the surrounding noises.
It's too bad when considering how the human listener is absolutely perfect in all regards.

Once you start to look at how music is really made, and at your own listening conditions and abilities, you often change your views on fidelity, resolution and what matters. What's funny is that the people telling you to mostly forget about hires and DAC resolution, are the very people that everybody on Headfi treats as objectivists.

 

[Vamp898]

Mastering is completely independent of the data rate of the file.

100% agree and not arguing against that. But companies do use Hi-Res Audio as a seal of quality and use those Masters exclusively for Hi-Res Audio Versions to make you think, it sounds better to make you spend more money.

 

[bigshot]

But in theory, there is more space in an Notebook to implement these things than in an DAP, so an Notebook could sound better than an DAP, if the Amp section is shielded good enough, just for the sake of more space for good components.

I've directly compared the output of my Oppo PM-1s through an HA-1, an iPod Classic and through my iMac and they all sound identical. I haven't compared my MacBook Pro against the HA-1, but I would bet that it sounds the same because I compared that to my iMac and couldn't tell a difference. I think laptops have come stock with much better audio in the past few years. What you're talking about might not be much of an issue any more.

 

[Vamp898]

I've directly compared the output of my Oppo PM-1s through an HA-1 and through my iMac and they sound identical. I haven't compared against the HA-1, but I would bet that it sounds the same through my MacBook Pro because I compared that to my iMac and couldn't tell a difference. I think laptops have come stock with much better audio in the past few years. What you're talking about might not be much of an issue any more.

This is the latest state of technology, nothing old or oudated im talking about.

The question is not, is the difference there.

I can hear a difference between my Notebook and every single Amp and DAP i own/owned and i can pick them apart in blind tests. So there is an difference. It can also be measured.

The question is, why can you not percieve it. And this is a complex thing that is not easy to troubleshoot. There could be tons of reasons why you don't hear the difference.

The first question is, is your IEM good enough to make the difference audible in the first place. In adition to that, what songs did you use for testing.

Some people think its related to the hearing, but my girlfriend is, due to an desease, hard of hearing and she can pick apart amps in blindtests too as her hearing goes bad above 10kHz and the biggest difference is in the Bass and Mids.

But the information i posted is the current state of technology and absolutely current and relevant. This is not old/outdated information. There is not one single maker on the market who can produce flat measuring capacitors and the differences are sometimes very big.

 

[bigshot]

I was just letting you know that there are laptops that are audibly transparent, and there are a lot more of them now than there were five years ago or more. I can't speak to your particular laptop, but if you can hear a difference, that might be something you'd want to address when you're ready to pick a new laptop. It is possible to get one that is audibly transparent.

As for the quality of the cans and DAC/amp I used, they are first quality. You can google them if you aren't familiar with them.

 

[gregorio]

as far my view (currently) objectivly goes is this:

1. higher samplerates have relaxed requirements for reconstruction filters where with 44,1khz you have a very narrow window (20khz to 22,05khz) for (well implemented) reconstruction filter which is nearly impossible to have

That’s exactly the problem! It’s next to impossible to have an “objective view” as an audiophile because so much of the available audiophile information is false. It is based on marketing requirements NOT any requirement to be honest or factually accurate! For example:

1. A typical modern audiophile bit of marketing. It’s based on a lie, in this case a “lie of omission”! The statement is true of analogue filters, not of digital filters. Hence why we have oversampling, which solves the problem because the requirement for a filter with a transition band of 20kHz-22.05kHz is now in the digital domain which is pretty easy. In fact transition bands of only 500Hz are pretty easy to implement in the digital domain these days, although you do have to throw some computing power at it.

2. resampling itself can have effects on SQ, there is a good comparision website for resampling implementations but i dont have the site right now, where you can clearly see differences between them, and also subjectively there are alot of differences between them, i tried 3-5 methods so far and all sound somewhat different, tho the differences might be negligable -between the good ones-, windows resampling is one of the worst i tried

2. I presume you mean the infinitewave (http://src.infinitewave.ca/) site. That’s been going for many years. Yes, you can “clearly SEE” differences between resampling algorithms because they have deliberately emphasised the visual differences! The FAQ states:
“Are most SRCs really that bad?
No. If you look at the decibel scale to the right from the graphs, you can see that the range of these graphs is very wide: down to -180 dB. The distortions generated by most properly designed SRCs are below -100 dB and can hardly create audible artifacts. The bottom line is that most tested SRCs range from fairly good to excellent, but the graphs are very sensitive to emphasize the differences.” - Emphasis mine. Incidentally, the Windows 10 DS resampler (the only listing for Windows) is very good, artefacts are roughly -120dB to -140dB and way below audibility or even reproducibility!

3. resampling, good implemented, can have definitely a positive effect on SQ compared to 44,1khz, "probably" because the reconstruction filter messes less with the audible range, or atleast that might be one reason

3. Oversampling is always “good implemented” and has been for 2+ decades. Oversampling the DAC has been around since the release of the CD almost 40 years ago. There may have been some poor implementations in the 1980’s, although I’m not aware of any.

4. if we speak about what samplerate is actually needed, objectively speaking 48khz is already a whole lot better then 44,1khz, if we speak purely about the problem with reconstruction filter

Objectively speaking, 48kHz is virtually no different as far as an analogue reconstruction filter is concerned. The earliest consumer oversampling DACs (built into CD players) were 4x oversampling, IE. 176.4kHz. By the late 1980’s many/most DACs in consumer CD players were using 64x oversampling (2.8mHz sampling rate) and “the problem with the reconstruction filter” was trivial!

5. there are alot of different reconstruction filters, purely because of this i believe the heared differences between dacs, because reconstruction filter sound different

There are not a lot of different reconstruction filters as far as I’m aware and reconstruction filters have been trivial to implement for 30+ years. What has changed is the number of available anti-imaging filters. This is a relatively new development and is entirely driven by marketing! Before this there were no switchable anti-image filters, DAC chips all had a single linear or minimum phase filter, with typically a 2-3kHz transition band and NO audible artefacts. Then a few years ago audiophile manufacturers applied their typical scam of taking a non-issue and falsely claiming it was an audible issue, this time to anti-image filters. In response to manufacturers demands to provide a solution to this non-issue, the DAC chip manufacturers implemented switchable filters. The vast majority of which are still inaudible but there are some with transition bands starting far lower, which can be audible.

So... what options are there?

Objectively and if high fidelity is what you’re after, the answer is: Do nothing! Typically the default filter is the same sort of filter used in DACs a decade or more ago with completely inaudible artefacts. If you’re not sure, don’t choose the option usually called something like “slow”.

G

 

[gregorio]

please let me ask this; is there any use case (as a consumer) where higher bits/rates are useful and possibly could make a difference ? or there is absolutely none whatsoever..

The simple answer is: No, absolutely none whatsoever. A more complex answer is: There could be rare consumer scenarios where greater bit depths might be advantageous. However, this doesn’t apply to the bit depth of the audio file itself (16bit) but to the environment in which it’s processed, which is invisible and unalterable by the consumer. An example of such processing where a higher bit depth *could* be beneficial would be a digital volume control.

can i conclude then the audible differences, clarity, detail i hear with hi-res music is due to its mastering, rendering it somewhat louder and applying certain filters to highlight some details therefore losing fidelity on the other hand ?

There are two reasons why consumers hear differences: 1. Placebo Effect, there’s actually no audible difference but a consumer perceives a difference due to cognitive biases. 2. The master is different. Typically this is just some additional compression applied to the 16bit version. In fact, about 200 pages earlier in this thread, there are some posts from a representative of one of the “Hi-res” suppliers (I think it was HDtracks but I could be mistaken), where they stated they instruct their mastering engineers to add additional compression to all their 16bit versions!

there are some companies for example recording in dsd’s or super hi frequencies and selling them as pure dsd etc… can we call this also some sort of milking the audiophool ?

Absolutely!

when you’re shopping for music to download, and when you’re given choices from pcm/dsd, mp3/flac/wav/aiff, 16-44/24-96/24-192/1-64/1-128/1-256/1-512 with very small differences in their prices, and sometimes 24-96 being much cheaper 16-44 (i find this on qobuz), what do you choose ?

Impossible to say. Depends what, if any, additional processing has been applied to the different versions. They could all be audibly identical, the 16-44 and lossy versions might have additional processing (such as additional compression/limiting) and in some rare cases, the higher rate/res versions have processing which can make them sound inferior in some listening conditions.

So, it’s a bit of a lottery. This is a good example of where audiophile marketing BS affects even those of us who know it’s BS. Without the marketing BS we would only have a 16/44 and lossy versions and wouldn’t have to take pot-luck as to which version to buy!

G

 

[gregorio]

Hi-Res Audio Recordings are Mixed/Mastered different and do sound significantly better due to this.

Generally Hi-Res audio recordings are Mixed/Mastered identically because (although over-simplifying) there is only one final mix and one master. This master is “virtual” (it exists only in the DAW), at the highest sample rate that will be distributed and with a bit depth of 64bit. From this master all the distribution versions will be “bounced down”. However, some distribution versions may have additional compression applied before bounce down.

The Bass of the non Hi-Res Version often is mixed much stronger and sounds muddy while it is, generally speaking, much faster and better controlled in the Hi-Res Audio version. Also those songs often have mauch stronger/better Stereo separation.

Not typically, the bass would not be mixed much stronger/muddier, there would be just one “mix” as mentioned above. However, if additional compression is applied when bouncing a version (when the mastering is otherwise completed), a full range compressor/limiter is most likely to affect the mid bass, as that’s where the amplitude peak is most likely to be with the vast majority of music mixes and this could easily result in the symptoms you describe.

More and more Studios even use separate Mixes/Masters for Streaming Services.

Yes, that’s quite common. It’s been fairly common for quite a few years and is becoming more common. This is due to the loudness normalisation applied by most streaming services. Unfortunately, they mostly have somewhat different loudness normalisation levels, for example YouTube is about -13LKFS while Apple is about -16.5LKFS.

So the only way to ensure you have the highest possible quality is to buy Hi-Res Audio

Unfortunately, that doesn’t ENSURE the highest possible quality, occasionally the Hi-Res version is poorer quality, although this is quite rare AFAIK.

The ear could in theory be able to percieve a dynamic that is larger than 96db …

Not really! As far as I’m aware we (science) don’t yet know the dynamic range of the human ear. For many years, a figure of 120dB was quoted, then it was discovered that it only has a dynamic range of around 65dB but is able to dynamically adjust that range within a 120dB window (in a similar fashion to how the human eye can dynamically vary it’s brightness range). Some more recent research suggests a dynamic range of only about 35dB which can rapidly vary within the 65dB window, which in turn can accommodate a window up to around 120dB (but more slowly).

Just imagine an 100db loud sound and then an 0db loud sound in the background... who would ever notice that and what qualifying information would be in that?

No human could notice that, because at most we only have a dynamic range of about 65dB. However, given highly specific conditions it might just be audible, although such conditions would not apply to consumers/audiophiles. You would need young, perfect ears/hearing, along with an anechoic chamber and a very sort duration of 100dB sound or many minutes gap between the 100dB and 0dB sound in order for the ear to re-adjust within the “window”.

So in theory, the human ear is better than an Audio CD, but in the real world, it never matters.

No, in theory the human ear is not better than an audio CD. Even if we take the full “window” of human dynamic range hearing of 120dB then (not coincidentally!), the audio CD has the same 120dB dynamic range (with noise-shaped dither)! I agree that in the real world “it never matters”, not least because there aren’t any commercial music recordings with anywhere near the dynamic range of even CD without noise shaped dither.

G

 

[gregorio]

I’m not going to respond individually to all your assertions about capacitors etc., it’s largely off topic and really it boils down to these two points anyway:

The question is not, is the difference there.

Especially as far as analogue signals/circuits are concerned your statement is true. According to both the laws/theory of physics and our high precision measurements.

The question is, why can you not percieve it. And this is a complex thing that is not easy to troubleshoot. There could be tons of reasons why you don't hear the difference.

Yes, that is the question. And, there could indeed be tons of reasons but by far the most common reason is because it’s below (or way below) the threshold of audibility!

Capacitors are a red herring. Of course you *could* be completely incompetent and use the wrong capacitor but that’s a broken/faulty design. Size is obviously not a concern. An Apple dongle for example is tiny, its components are microscopic and yet its output has measurable artefacts ALL well below audibility (and it only costs about $10!).

Differences in power/impedance/voltage can of course be audible and obviously varies from unit to unit (DACs, amps and HPs/speakers). Hence the requirement for reliable comparative listening tests to quite precisely volume match. The wild swings in impedance throughout the spectrum found with some HPs/IEMs have no audible effect assuming the correct amp is used (with sufficient power and an appropriate output impedance). With some rare pathological exceptions, such as a few tube amps that deliberately apply massive amounts of distortion, we’re back to our “most common reason” above, as proven back in the 1980’s.

Computer/Laptop sound cards can have audible differences, for starters there’s the close proximity to relatively high levels of interference, a common requirement to be exceedingly cheap and particularly in the case of laptops, a requirement to limit/conserve power. As mentioned by others though, there does seem to be a trend to place a higher priority on the quality of audio output from computers and in more cases it’s impossible to differentiate. Really terrible (audible) implementations are rarer today than the laptops/computers of say 20 years ago.

G