What audio system do you have, or plan on getting?

[Harry]

Ah! Streamers: there's a topic. I'm not sure what to make of it. I'm not even sure whether I presently have a "streamer" or not.

Follow my "Feanor's Stereo Configuration" link and you can see what I've got.  Is my Windows 10 computer a streamer?  I can "stream" music from my 4000-odd ripped CD collection residing on my NAS device.  I also can stream via that computer CBC Music's large selection of preprogrammed playlists in many genres as I occasionally do.

Of course I could access Qobuz or the like using my Windows computer:  I don't see why or how an purpose-made stream would help. I don't presently use Qobuz, etc., mainly because I want to avoid monthly access fees but also because I find searching for Classical music I might want to listen to awkward using their search engines.

I don't know what if anything I should do improve my listening experience.  (BTW, I don't listen to music all day and in every room in my home, which is a thing some want to do.)

Sort of yes.

[71 dB]

I appreciate your "like" for my post about hi-rez audio Harry. 

[Harry]

I appreciate your "like" for my post about hi-rez audio Harry. 

Well I appreciate the info you provide Poju. I have a different opinion in de case of HR and what it means for my ears, and what should be implemented in my system, but I always enjoy your insightful views.

[Florestan]

Oh, not region, but resolution! Well, I don't know if I can convince you Harry, but you or anyone else doesn't need high resolution audio, because it doesn't offer any audible improvements. If a high resolution file sounds "better" than a CD-quality or high bitrate lossy file, it is because it is made of a different master, something that is done deliberably to create audible differences between file formats and justify charging more money for high resolution music. People fall into this "scam" easily because is it so intuitive. Of course more bits means more accuracy, right? Blu-rays look much better than DVDs, and 4K discs look a little better than Blu-rays, so obviously High resolution audio sounds better than CD-quality, right? No, it doesn't, because CD-quality is already as good as 8K video. It is all music consumers need and then some. Higher resolution formats are only needed/beneficial in studios in music production for technical reasons just like a car needs a big factory building to be manufactured.

Sample rate doesn't dictate the accuracy of signals. It dictates the highest frequencies you can have. At your age it is a miracle if you hear anything above 16 kHz. So, you "need" a samplerate of 32 kHz + some transition band for anti-alias and reconstruction filtering. Let's make it 3 kHz. That's 35 kHz. CD quality gives you 44.1 kHz. That is 26 % more than what you need, seriously.

Bit depth dictates the noise floor. Every bit lowers the noise floor by 6.0206 dB. Without dithering the noise floor correlates with the signal, because it is quantization distortion. With dithering the noise floor is a few decibels higher, but quantization distortion disappears completely! There is just completely distortion free original signal mixed with a noise floor that doesn't correlate with the signal at all. Signals BELOW the least significant bit can be recorded and they exist with the noise floor partially or completely masked. People say analog signals fade into the noise floor. Well, properly done digital audio does the exact same thing. All we need is to get the noise floor down enough so that people don't hear it. How down is that? A new good vinyl has a dynamic range of about 60 dB (10 bits worth). Human hearing has a dynamic range of about 120-130 dB (from the hearing threshold to the pain threshold), but we can't hear this range at the same time. Our hearing has a dynamic "window" of about 70 dB that moves slowly according to how loud the environment is just like our eyes adjust to bring light and darkness. So, music "needs" about 70 dB of dynamic range and that is why even vinyl has almost enough dynamic range. Let's ad some savety margin to that and we have about 80 dB or about 13 bits of dynamic range. That is what music consumers "need" and it covers the most demanding music in most demanding listening scenarios. CD-quality gives you 16 bits. That's 3 bits (18 dB) of overkill. That's why CD-quality is enough for music consumers and high resolution formats belong to the studios. In fact, AAC 256 VBR lossy format is already quite enough not to mention AAC 320 VBR or similar lossy formats, but if don't want to worry about lossy coding, CD quality FLAC is enough bitrate.

Sound quality doesn't come from high sample rates and bit depths. It comes from performing, recording, producing, mixing and mastering well. Those things are done these days with high resolution formats anyway (at least at 24 bit if not at higher samplerates).

These are not only my "opinions." These are facts said by those who really understand digital audio and aren't earning their living by scamming people with hi-rez audio.

Thanks for the detailed and clearly-written explanation, Poju.

I think this should settle the matter once and for all, coming as it does from a professional acoustics engineer whose record contains publications in peer-reviewed scientific journals.

[71 dB]

Thanks for the detailed and clearly-written explanation, Poju.

You are welcome! I have written these kinds of posts for years on various internet forums, so that might explain the "clearly-written" part.

I think this should settle the matter once and for all, coming as it does from a professional acoustics engineer whose record contains publications in peer-reviewed scientific journals.

Well, these kind of posts don't settle things for everybody. I was sceptical about being able to convince Harry, and I was right about that it seems. Some people are convinced by these posts and have realized how hi-rez is "waste of money/storage space" and have moved back to "CD-quality." It has taken years for myself too to fully understand these things and admit that CD-quality is kind of the end point for music consumers, and there is nothing "better" beyond that. On the other hand it is great that consumer audio has achieved perfection (music format-wise) in the 80's or maybe in the 90's if we consider the quality of DACs and so on. There is still room for improvement on other areas of audio such as speakers, headphones and room acoustics. Those are the main areas were perfection hasn't been achieved at all.

I don't expect to convert everybody, but some people here and there might convert. It is up to me how much I "hone" my message and present if respectfully without insulting other people turning them away. Also, I don't expert fast results. I hope I can insert a "seed" to the mind of people and in time that seed might grow and lead to change of mind. Maybe that happens to Harry, maybe it doesn't. I am just trying to help him save a little bit money and make life easier. I have zero financial reason myself to speak about this. I feel morally obligated to speak up instead of keeping my knowledge and understanding to myself.

[71 dB]

Well I appreciate the info you provide Poju. I have a different opinion in de case of HR and what it means for my ears, and what should be implemented in my system, but I always enjoy your insightful views.

Thanks Harry, and it is all fine. How you use your money is your own business.

[Florestan]

You are welcome! I have written these kinds of posts for years on various internet forums, so that might explain the "clearly-written" part.

Well, these kind of posts don't settle things for everybody. I was sceptical about being able to convince Harry, and I was right about that it seems. Some people are convinced by these posts and have realized how hi-rez is "waste of money/storage space" and have moved back to "CD-quality." It has taken years for myself too to fully understand these things and admit that CD-quality is kind of the end point for music consumers, and there is nothing "better" beyond that. On the other hand it is great that consumer audio has achieved perfection (music format-wise) in the 80's or maybe in the 90's if we consider the quality of DACs and so on. There is still room for improvement on other areas of audio such as speakers, headphones and room acoustics. Those are the main areas were perfection hasn't been achieved at all.

I don't expect to convert everybody, but some people here and there might convert. It is up to me how much I "hone" my message and present if respectfully without insulting other people turning them away. Also, I don't expert fast results. I hope I can insert a "seed" to the mind of people and in time that seed might grow and lead to change of mind. Maybe that happens to Harry, maybe it doesn't. I am just trying to help him save a little bit money and make life easier. I have zero financial reason myself to speak about this. I feel morally obligated to speak up instead of keeping my knowledge and understanding to myself.

Actually, one doesn't have to have a degree in acoustics engineering. Basic high-school biology and physics will do. Besides, I'm pretty sure that in a blind test nobody, and I mean nobody, would be able to tell the difference between CD-quality and hi-res quality, all other variables being equal. Heck, blind tests have proven beyond dispute that most people weren't even able to tell the difference between mp3 and FLAC, all other variables being equal.

[Todd]

Actually, one doesn't have to have a degree in acoustics engineering. Basic high-school biology and physics will do.

Audiophilia has nothing to do with science; audiophilia is about marketing.

[Harry]

Actually, one doesn't have to have a degree in acoustics engineering. Basic high-school biology and physics will do. Besides, I'm pretty sure that in a blind test nobody, and I mean nobody, would be able to tell the difference between CD-quality and hi-res quality, all other variables being equal. Heck, blind tests have proven beyond dispute that most people weren't even able to tell the difference between mp3 and FLAC, all other variables being equal.

Well I can!

[Spotted Horses]

I am not surprised that there is a small but perceptible difference between CD resolution and high resolution because your equipment will typically use a different signal chain when the resolution changes. When streaming FLAC files from my computer I first had a setting that by default converted everything to 96kHz/24bit for Digital audio output and when I switched to a setting that let the media player set the output rate to match the media being played it sounded slightly different, presumably because the computers algorithm for upsampling from 44.1kHz to 96kHz was not identical to the DAC's algorithm for upsampling 44.1kHz to its internal native resolution. If I paid $10k for the 96kHz/24bit setting I could probably convince myself that "different" was "better."

[71 dB]

Actually, one doesn't have to have a degree in acoustics engineering.

I don't recommend degree in acoustics engineering for anyone. There's very little job opportunities for acoustics engineers and other branches of engineering are quite different making it difficult for acoustic engineers to get other kind of engineering jobs.

I was an idiot getting myself a degree in acoustics engineering. Instead I should have gone to HVAC engineering or (game) coding or things like that offering tons of job opportunities, but it is what it is. I made these "career plans" in a time when university degree in engineering almost guaranteed a well-paying job. The world has changed dramatically since, and I am paying a big price for my mistakes decades ago when I knew nothing about anything.

Basic high-school biology and physics will do.

To some extent, except few pupils really learn this stuff in school or remember anything about it afterwards. Digital audio contains mathematical aspects that are beyond high-school science. Even many highly educated people struggle with these things. One example is the temporal resolution of digital audio. Laymen think the temporal resolution of CD-audio in 1/44100 s = 0.000022676 s. This is completely wrong, because we can move the wave in time less than the time period between sample points. This happens by changing the phase of the signal. This results in sample data that is different.



In this example we have a 1 kHz sine wave sampled at 44.1 kHz. The lower version is the upper delayed by 1/4 samples = 0.000005669 resulting in different values for the sample points, but still representing the same sine wave. I could have made the delay only 1/100 samples just as easily, but the difference would have been impossible to see. The equation to calculate temporal resolution 𝞓Tres of digital audio is

𝞓Tres = 1 / ( 2 * 𝞹 * f * N ),

where f is the signal frequency and N is the amount of possible sample values for the signal (e.g. N = 6554 for -20 dB signals at 16 bits). For example for -20 dB sine wave of 2205 Hz at 16 bit the temporal resolution is about 11 ns, which is more than 100 times better than human hearing. As you can see, surprisingly the sample rate itself is not in the equation!! It doesn't matter, if the sample rate was 44.1 kHz or 192 kHz! The resolution in this case is still 11 ns. This is one of the good examples of how unintuitive digital audio can be. bigger bit depth does give much better temporal resolution, but this equation works for digital audio WITHOUT dither and practically all CD-quality audio is dithered. Dither makes the temporal resolution practically infinite. It means that as long as we dither correctly, temporal resolution is not a concern in digital audio, no matter, hi-res or not. 

Besides, I'm pretty sure that in a blind test nobody, and I mean nobody, would be able to tell the difference between CD-quality and hi-res quality, all other variables being equal. Heck, blind tests have proven beyond dispute that most people weren't even able to tell the difference between mp3 and FLAC, all other variables being equal.

Yes, and there are listening tests done indicating this. There is no 100 % proof of it, because these listening tests are statistical in nature (how many time could people choose correctly), but there is near 100 % certainty of it. People who believe in the audible superiority of hi-rez keep defending their view by saying the listening tests cause "pressure" to the listeners and make listening differences harder etc., but the real reason why people claim to hear differences are:

Placebo effect. This is very strong when we listen to sounds. What we think we hear is affected a lot by what we think we should be hearing. All of us may have experienced this effect in interesting ways. For example, if you have a mono button in your audio gear and you have activated it for some reason, but you have forgotten it, you may listen to a stereo recording for minutes before getting a feeling that something is off. When you realize you have been listening to mono sound thinking it is stereo for minutes you laugh at yourself, but that's placebo effect. The belief of listening to stereo sound makes your brain accept mono sound as stereo for minutes before cumulating a feel that something is wrong. My "rule of thumb" is to assume 90 % of differences we hear in audio are due to placebo effect, imagined differences. This explains why differences tend to go away or become much smaller in proper blind listening tests.

Different masters. A high-res file might be masteres differently from a standard-res files. If so, of course people are hearing differences, but the reason is not the resolution. Downsampling the high-res version properly to CD-quality will result in the exact same sound quality.

"Surfing the wave." This is about taking a very quiet part of the signal and amplifying it massively (say 50 dB) to bring out the noise floor to clearly audible levels. A file with bigger bit depth most probably has lower noise floor and wins, but it is a victory that doesn't matter in practise. Who listens to music by turning the volume knob to maximum whenever the music fades into the noise floor? In any practical or sane listening scenario about 13 bits of dynamic range is enough.

Ultrasonic IM distortion Hi-res music may contain strong infrasonic noise not even related to the music that has gone unnoticed in the production due to not be audible, but with some audio gear not designed to handle strong ultrasonics may cause intermodulation distortion that generated distortion component to audible frequency range. CD quality audio on the other hand doesn't contain anything above 22 kHz (all of it is filtered to extremely low levels) and such IM distortion doesn't happen. In this sense CD quality audio can actually be purer than high-resolution audio, but this is theoretical rather than practical. 

[Fëanor]

Well, these kind of posts don't settle things for everybody. I was sceptical about being able to convince Harry, and I was right about that it seems. Some people are convinced by these posts and have realized how hi-rez is "waste of money/storage space" and have moved back to "CD-quality." It has taken years for myself too to fully understand these things and admit that CD-quality is kind of the end point for music consumers, and there is nothing "better" beyond that. On the other hand it is great that consumer audio has achieved perfection (music format-wise) in the 80's or maybe in the 90's if we consider the quality of DACs and so on. There is still room for improvement on other areas of audio such as speakers, headphones and room acoustics. Those are the main areas were perfection hasn't been achieved at all.

Nowadays I'm completely unconcerned with resolutions above 16 bit @ 44.1 kHz. If I could actually hear differences at higher rez, I don't care because any improvement would certainly be insignificant.

Perhaps the fact the I'm stone deaf about 9-10 kHz has something to do with how I feel, or maybe it doesn't.

Anyway, far more that resolution of the medium, the quality of the recording & mastering is the bigger determinant of the sound quality we perceive.

[71 dB]

I am not surprised that there is a small but perceptible difference between CD resolution and high resolution because your equipment will typically use a different signal chain when the resolution changes. When streaming FLAC files from my computer I first had a setting that by default converted everything to 96kHz/24bit for Digital audio output and when I switched to a setting that let the media player set the output rate to match the media being played it sounded slightly different, presumably because the computers algorithm for upsampling from 44.1kHz to 96kHz was not identical to the DAC's algorithm for upsampling 44.1kHz to its internal native resolution. If I paid $10k for the 96kHz/24bit setting I could probably convince myself that "different" was "better."

This is a good point. One has to be careful about what difference we are actually listening.

44.1 kHz to 96 kHz upsampling isn't "trivial."

[Spotted Horses]

Anyway, far more that resolution of the medium, the quality of the recording & mastering is the bigger determinant of the sound quality we perceive.

I would say the original manner of recording is the key. I find many of the old Mercury's more convincing than a lot of state of the art recordings because their very simple setup (three omnidirectional microphones suspended above the orchestra) allowed the sound the be mixed by the hall. Similar considerations for older Decca (tree) and RCA victor recordings. There are wonderful recordings being made today, but there are also recordings made with very refined equipment that sound bad to me because they over engineered (nearing the point of a microphone on every instrument and mixed on a digital console).

[Florestan]

I don't recommend degree in acoustics engineering for anyone. There's very little job opportunities for acoustics engineers and other branches of engineering are quite different making it difficult for acoustic engineers to get other kind of engineering jobs.

I was an idiot getting myself a degree in acoustics engineering. Instead I should have gone to HVAC engineering or (game) coding or things like that offering tons of job opportunities, but it is what it is. I made these "career plans" in a time when university degree in engineering almost guaranteed a well-paying job. The world has changed dramatically since, and I am paying a big price for my mistakes decades ago when I knew nothing about anything.

That's kind of funny, because I got myself a degree in HVAC engineering based on a "plan" which looked sound back then but quickly proved to be not so realistic.

To some extent, except few pupils really learn this stuff in school or remember anything about it afterwards. Digital audio contains mathematical aspects that are beyond high-school science.

I'm not talking about the mathematical aspects which are indeed difficult and require calculus. I'm talking about the basics of the human ear and the physical process of hearing: there is a minimal and a maximal threshold of frequencies and intensity, below which the ear perceives nothing and above which it perceives only noise, painful one even, and this physically limited range gets narrower with age. Any audio format which claims they extend this physical range and can make people hear things they cannot physically hear, especially elder people, is bogus.

Placebo effect. This is very strong when we listen to sounds. What we think we hear is affected a lot by what we think we should be hearing. All of us may have experienced this effect in interesting ways. For example, if you have a mono button in your audio gear and you have activated it for some reason, but you have forgotten it, you may listen to a stereo recording for minutes before getting a feeling that something is off. When you realize you have been listening to mono sound thinking it is stereo for minutes you laugh at yourself, but that's placebo effect. The belief of listening to stereo sound makes your brain accept mono sound as stereo for minutes before cumulating a feel that something is wrong. My "rule of thumb" is to assume 90 % of differences we hear in audio are due to placebo effect, imagined differences. This explains why differences tend to go away or become much smaller in proper blind listening tests.

Of course. Plus, when one has spent $50,000 on a new pair of headphones, the Placebo effect is all the stronger. I'm sure they are really convinced they hear things that hard science says they cannot hear.

[Florestan]

I would say the original manner of recording is the key. I find many of the old Mercury's more convincing than a lot of state of the art recordings because their very simple setup (three omnidirectional microphones suspended above the orchestra) allowed the sound the be mixed by the hall. Similar considerations for older Decca (tree) and RCA victor recordings. There are wonderful recordings being made today, but there are also recordings made with very refined equipment that sound bad to me because they over engineered (nearing the point of a microphone on every instrument and mixed on a digital console).

This is a good point as well.

[DavidW]

The 280s are seriously underrated. Despite only being 64ohms, a powerful amp brings out the best in them. They are used as monitors in a number of top recording studios and having owned a pair I can understand why. They have a very neutral sound signature and this is a bit off putting for many but for classical music it's a plus.

I also remember them having a great seal, which should help in Spotted Horse's case.

[DavidW]

I have a Bluos Node 2i streamer and made my first steps in streaming content. The experience was positive, and now I seek a streamer to insert in my HR system, and would very much like to invite suggestions which way to go. The Bluos Node is a fine streamer, but not fit for HR.
So.......? I myself was thinking of a NAIM UNITI ATOM HDMI, roon implemented.

The Node that I just bought streams hd audio.  Optical out can and does pass hd audio to the dac.  It doesn't have the bandwidth for multichannel hd audio though if that is what you're thinking.

[Harry]

The Node that I just bought streams hd audio.  Optical out can and does pass hd audio to the dac.  It doesn't have the bandwidth for multichannel hd audio though if that is what you're thinking.

What I meant was that I fitted Node into my HR system, and the quality did not match my expectations, so I am seeking for an upgrade in that respect.

[Fëanor]

I would say the original manner of recording is the key. I find many of the old Mercury's more convincing than a lot of state of the art recordings because their very simple setup (three omnidirectional microphones suspended above the orchestra) allowed the sound the be mixed by the hall. Similar considerations for older Decca (tree) and RCA victor recordings. There are wonderful recordings being made today, but there are also recordings made with very refined equipment that sound bad to me because they over engineered (nearing the point of a microphone on every instrument and mixed on a digital console).

Regarding the old Mercury Living Presence recordings, I agree that they are exceptional in terms of capturing the ambience of the actual venues where they were recorded.  I have quite a few of these recordings in CD format.

On the other hand and for various reasons I find the some more contemporary recordings to be at lease as good in most respects (-- I won't say they necessarily capture the actual ambience of their recording venues).  For instances you might check out Grammy Awards for Best Engineered Album, Classical, for recent years.

I recently purchased 2019 award winner, Boston Symphony Orchestra conducted by Andris Nelsons: Shostakovich, Symphonies 4 & 11.  This possibly the best orchestral sound that I've ever heard on record.