See Psychoacoustics. Yippie!
why? I dont understand
Yes, the audio chain.
To justify 192kHz files existence, there is needed:
- ignore the fact that fq over 20kHz nobody can hear
- 100kHz microphone with flat fq response used at studio (exists but quite new on market since 2019 I think)
- audio processor and amplifier with flat fq response up to 100kHz
- speakers with flat fq response up to 100kHz (I don’t think any exists on the market)
- speakers +/-3dB response down to 20Hz (nobody cares as standard music (except electronic or octobass instrument) doesn’t go so low)
- room EQ up to 100kHz (if somebody want ultrasonic fqs - he wants them precise, right?)
- ability to hear over 20kHz (to hear 100kHz
) in his age - animal rights activists audio system certification to avoid bats injury when listening with opened windows
Each hi-res enthusiast should check his hardware specs, request type of microphone used in studio to be reported on record label and he should check if audio engineer who was mixing track is capable to hear at least above 16kHz in best case over 20kHz and has ultrasonic speakers in studio so he could properly balance ultrasonic frequencies in his mix (and that the record was tested on dogs not them to freak out).
6 Likes
A higher bit rate in recording doesn’t mean its recorded below or above what human ears can hear 20hz to 20 khz. It means there is more information present in the file allowing for more nuances, detail to be heard. Depending on your system you might not hear the difference between MP3 and Flac. Could also be a limitation of your hearing. Personally MP3 or AAC sounds crushed compressed. Switching to CD or better quality FLAC or ALAC the sound stage opens up, there is depth, detail, nuances (such as fingers dragging on guitar strings, chimes that sound like chimes and not some cheap keyboard). Can you hear the difference between CD and DSD1x or 24/48hz mathematically you should be able to. Does your playback system have the resolution or are your older ears able to hear it well thats what high-end audio manufacturers make a living on.
1 Like
A higher bit rate in recording doesn’t mean its recorded below or above what human ears can hear 20hz to 20 khz. It means there is more information present in the file allowing for more nuances, detail to be heard.
Sorry to say, this is wrong. The audio technology you are reffering to is based on Shannon Sampling Theorem. It basically says, that eg with 44.1 KHz sampling rate, you can completly restore the original signal.
You can proove this easily if you compare the analouge with the digitalized signal. (Fast Fourier)
It is the same like you would transfer infra-red information within JPEGS. The image has no better quality for the human eyes…
3 Likes
HAHA thats funny my college professor filled two chalk board’s with mathematical proof when someone said CD was better than LP. This is why I specifically stated 24/48 and above. Thats is whats needed to record 20hz to 20 khz. Regardless Is "16 bit CD Quality" Good Enough?
This gentleman wrote an article about it. Sorry I was wrong more can be recorded in that space but who can hear above 20hz-20khz the cat and dog?
As for your analogy try looking at a 14bit raw image file and comparing it to 12 bit or lower jpg/jpeg on a high end monitor or TV there is much missing information as compression sets in. Yes I can hear the difference between MP3/AAC and Flac/Alac our audio club we love doing blind listening tests to see what we come up with. I am harder pressed to hear the difference between say 24/96 and DSD 1x. Can pick off well recorded 24/96 compared to 16/44 (we use Roon volume leveling and rip music with replay gain (or add it)). As I get older it becomes diminishing return with upgrades. My last one will be replacing three boxes with a Bryston BR-20. As for Apple Music and 24/192 Alac my new apple TV sounds better playing Queensryche OM, E and PL than the same remastered CD’s on Roon. Harder to do a blind listening test without volume leveling. (Do make sure you go to Music in app setup and enable Lossless over AAC).
… and regarding bit depth (in my understanding). It does not increase vertical resolution, just maximum loudness and therefore SNR. You cannot hear difference in quality between 16b v. 24b on the same loudness normalization level. It’s just about volume.
24b is used in studios to have safe headroom for recording and mixing to avoid clipping and separation of noise. Then consumer doesn’t need it. If extreme dynamic range is needed then track is mixed in -20 LUFS for example.
Bit depth is only and only about maximum volume posssible.
1 Like
…actually it is the dynamic, which marks the difference from the lowest to the highest level. Ironically we discuss hi-res, when the state of the art mastering approach for pop is the “low dynamic” most of the recent recordings you can encode with 10 bits, also the remasters. Ok nobody said, that a remaster needs to sound better…
Mr Martin_Kelly
As the old adage says, “Let’s compare oranges to oranges”. You need to look at the same recording produced from the same master on the various formats (CD, high-res, 48 - 192Khz, DSD) to see if there is a difference. And with a good master there definitely is. Going across the entire catalog to demonstrate the HI-RES quality doesn’t cut it. One thing you can be sure of is if the hi-res stinks than the so does the CD.
I agree that the average listener doesn’t care about sound quality, but this was always true. When stereos were first on the market there were those of us that liked transistor radios.
Any hobbyist that spends good money on a DAC, Preamp, Amplifier, Speakers and maybe power conditioner is going to hear a diffference on all the formats. 
1 Like
Always appreciate hearing Mr. Darko’s thoughts, which I think are spot on. Thank you for bringing this forward. Seems the thread has been hi-jacked by 'splainers who are reiterating nothing we haven’t heard before 
.
Will say that none of the professional and excellent amateur musicians I know give much of a rip about high-res music when they listen to streamed music. Perhaps because after mastering a musical instrument and creating/hearing the analog reality, reasonable approximations are AOK. Certainly for appreciating technique, which is often what musicians are listening to and audiophiles may or may not appreciate.
The more music I listen to, the more I am happy with pretty darned good SQ, rather than the exquisite best. Nothing against the exquisite best, but this gets into cost/benefit issues that are thin gruel for many folks that love live, streamed or digital/analog archived music. Also requires a level of attention that is rare to find sustained in any age cohort.
Vive la différence. No right answer, however 'splained it may seem to some.
A primary tone beyond the range of hearing can easily introduce noise within the audible range by harmonics. Which you as a physicist are thoroughly familiar with.
1 Like
That is incorrect. Harmonics are multiples of fundamental, so they are always above fundamental, not below. Frequencies that show up below fundamental are not harmonics, they are images caused by aliasing or intermodulation distortion. Those are obviously not desirable; you are correctly calling them ‘noise’. That’s one more reason to avoid ultrasonics in audio systems.
2 Likes
While you might think on paper this is good in reality using my ears it is not try turning on volume leveling and listen for micro detail. Then turn it off again, assuming of course the track is adjusted by an extreme amount.
Cheers
Please help me understand this Track Frequency Response. My understanding is although it’s marked as 24/96k but in reality it’s a 16/44.1k track.
If I take the Otis Taylor album I downloaded from Octave Records (Hey Joe Opus, superbly recorded in DSD, then just as superbly converted to 44.1/24, 94/24, 192/24 and DSD64) and make a playlist of one track, all four resolutions, then shuffle it over and over, I can easily tell when there is a change in resolution, and whether it increases or decreases. I know, because I have done it several times. And I’m 70 y/o, but have taken very good care of my ears.
1 Like
That’s still not valid. You keep avoiding doing an actual blind test for some reason.
All filters damage the sound to at least a tiny bit. The higher the frequency the filter starts working, the better, as they do less damage the higher the cut-off frequency. Even though the vast majority of music is under 5kHz (the highest fundamental frequency of instruments is just under 5kHz), music can have harmonic frequencies up to 40kHz, We don’t hear them, but all those frequencies interact with all other frequencies and can have a definite impact on the overall experience. And those fundamentals below 5kHz?? Those make the biggest impact on the experience, as the better they get reproduced (transients, imaging, ambiance, etc.) the better the harmonics get reproduced. At 70, with an upper hearing cut-off of 12.4kHz, I won’t get as much of an impact as I would have at age 23, when I could hear up to 24kHz, but I still get enough info to easily tell the benefit of higher resolutions.
This seems to be 16/96 (96 ksamples to reconstruct 48kHz).
Got it… It looks like there is some information between 24kHz to 48kHz (below 75db). In you opinion, can we really hear them?
No, you can’t.
- If you are superhuman then you can hear up to 24kHz but vast majority of mid-age don’t hear fqs over 15-17kHz
- even if you could then above 20kHz it’s too quiet (if you listen around 85 dB SPL then -75 dB is 10dB SPL - ambient noise in quiet room is around 30dB SPL)
- your audio gear is on 99.9pct not capable to play over 35kHz flat
- your blood pressure will most probably cause higher noise in ears than is volume of those fqs
- higher frequency higher directivity (move 20cm from calibrated sweet spot position, rotate your head slightly, move the curtain behind you… and it changes SQ perception way too much more than any difference in audio file quality)
https://www.chem.purdue.edu/chemsafety/Training/PPETrain/dblevels.htm
But people here will disagree with me