[whitedragem]

I don't have the link but testing was performed showing that humans could detect up to 500000hz and quite perceptibly to 200000hz.
 
perhaps it is smell.or some sort of synesia.
 
I don't care, but would agree that the 384khz upsampling dacs at least have a nice numberthat everything multiplies into neatly.
 
regarding bit rates, do electrical circuits generally resolve more than 21bits of info? Negatory.
 
Does 24bit 96khz sound better? Absolutely
 
and I am still happy playing back my CDs off a beaut transport, or soundfiles from a DAP,orheck my headphones from mywalkman, or even a minidisc player (yes I am aware of what it does to the sound file-but it has a great amp section and makes headphones sound good!).
 
actually tapes ifrecorded from good sources have a real lot to offer.
 
but lets argue FOR digital, presently, being good
 
if Microsoft hadn't killed off hdcd, I'd be very happy, asthe world would have gotten a cheap evolution of the CD.
 
so pundits go back to vinyl.....
 
 
edit....,oopps it was late, and I kinda just laugh at first line.
 
they were that sampling rates that people could detect variations at. Most people easily, under testing to 200, and extinction around 500khz sampling rate. Yes sound is listenable, sampling is listenable as having more samples and therefor a hopefully smoother sound, or closer to the origional, samples being the plotted points on the X axis, bit depth being the plotted points on the Y axis. part ofa DACs job to recreated those recorded points as accurately as possible. I find the improvement of even 4000 more per frame very noticable, and 48 multiplie into 192, doesn't create the artifacts that a 44khz sample will. So DVD rips when possible fill parts of my collection.

 

[ratinox]

Does 24bit 96khz sound better? Absolutely

Better than... what? A CD mastered from the same masters used for pressing vinyl? That's hardly fair; they're two different things. Redbook audio has so much more fidelity that the recording and mastering mistakes that are inaudible on vinyl become apparent on CD.

In actual double-blind, controlled testing, no human being has ever been able to reliably tell the difference between a 24/96 or 24/192 audio source and the same source downsampled to 16/44.1.

Edit: I don't know what sources you're referencing, but no reliable source that I've ever seen demonstrates human hearing outside of the Fletcher-Munson range. Dogs can hear up to around 60 kHz. Cats can hear up to around 80 kHz. But humans? Not even close. In fact, most adult humans with healthy hearing can't hear anything above about 16 kHz.

Edit 2: I found some references to people who claim to be able to hear things like ultrasonic rodent repellers. These sweep from about 30 kHz to about 60 kHz at high volume levels. Inaudible to humans. What they're hearing are lower frequency reflections from walls and furniture. In other words: comb filtering noise.

 

[ratinox]

double post

 

[bigshot]

  I don't have the link but testing was performed showing that humans could detect up to 500000hz and quite perceptibly to 200000hz.
 
Does 24bit 96khz sound better?

 
No and no.

 

[ag8908]

Sorry, dude, but "flat" has a well-defined meaning. That meaning is "level to within 1dB from 20Hz to 20kHz". Any deviation of more than 1dB across the audible frequency range is not flat. HD 800 has a measurable peak in excess of 1 dB at around 6.5 kHz. Therefore it is not flat (neutral).
Er. No. Math doesn't work like that. If you take X away then you cannot put X back unless you know what X is in which case X was never really taken away. No amount of computing power or clever programming can change this.

Tell that to autotune which creates voices which never existed in the first place. Also, if you have the mathematical characteristics of the thing you are trying to remove (e.g. an annoying hiss on a vintage recording), you can remove it. The DAC does that often on voice calls by removing background noise and things like that.

 

[ag8908]

Or the rest of us that like acurate playback of the best recordings we can find of our favorite music, maybe even without a CPU in our playback chain, might need our VR goggles checked and a ticket to the future. . . .
600mips? Call it Jack. It will even hear you!

 
To an extent there is no such thing as "accurate" playback, all DACs have to create a fantasy by converting data into sound. Their program can try to do it in a way that it sounds, in your headphones, like it sounded to someone standing in the recording studio, but that's really a huge effort which probably requires a customized dac for every headphone.

 

[ratinox]

Tell that to autotune which creates voices which never existed in the first place.

Ah, I understand now. Yes, Auto-Tune adds stuff that wasn't there. It doesn't recover stuff that was originally there and then discarded. Once discarded it is gone forever. The best that you can accomplish is a close approximation of what was originally there.

Also, if you have the mathematical characteristics of the thing you are trying to remove (e.g. an annoying hiss on a vintage recording), you can remove it. The DAC does that often on voice calls by removing background noise and things like that.

Point 1: removing something that's there is relatively easy.

Point 2: the DAC has nothing to do with either of these. "All" the DAC does is convert bits to electrical signals. It's a very specialized kind of signal processor. Performing noise reduction is a different kind of signal processing. Similarly, what Auto-Tune does is a different kind of signal processing.

 

[ag8908]

Regardless of what the ear can perceive, let's all agree that with huge hard drive spaces, memory capacity and processing power in the most basic device these days -- there is ABSOLUTELY no reason why song files should be compressed in any way. Sure in 1998 you needed to compress the MP3 so it was 4 megabytes, but today they could all be 100 megabytes each and I would have the space and processing power for their extra information.

 

[ratinox]

To an extent there is no such thing as "accurate" playback, all DACs have to create a fantasy by converting data into sound.

This is a fallacy. Analog to digital is a reversible process. That is, what comes out of the DAC is identical to what went into the ADC.

 

[ag8908]

This is a fallacy. Analog to digital is a reversible process. That is, what comes out of the DAC is identical to what went into the ADC.

 
That can't be true because my various DACs have different sound signatures. I can easily tell them apart. One thing I noticed about the objective 2 amp was that it reproduced the original DAC's sound signature. So if the three DACs (actually I have four now) are producing different sounds, then if you tried to convert the four sounds back into digits they couldn't result in the same product.

 

[ratinox]

That can't be true because my various DACs have different sound signatures.

The entire circuit, including the DAC (the IC, not the whole box), the op-amp, rectifiers, capacitors, resistors, transformers, power supply, etc., plus any signal processing performed before and after the DAC stage, all together have a signature. The DAC and op-amp actually have little to do with it, with the exception of some early CD players loaded with cheap DACs with poor anti-aliasing filters. As bigshot (I believe) pointed out: the implementation of the DAC is what's most important, not the brand name on the chip.

A distinctive sound signature is a lack of transparency. A transparent device will have no distinctive signature. Assuming that none of the devices in your collection adds distortion -- that is, adding sounds that aren't in the source -- then appropriate EQ settings should restore transparency making the conversion reversible.

 

[bigshot]

  Regardless of what the ear can perceive, let's all agree that with huge hard drive spaces, memory capacity and processing power in the most basic device these days -- there is ABSOLUTELY no reason why song files should be compressed in any way.

 
There are two very good reasons for compressed audio... Streaming and portable use. If your ear can't hear it, it's basically digital packing peanuts in the file.

 

[Xenophon]

Better than... what? A CD mastered from the same masters used for pressing vinyl? That's hardly fair; they're two different things. Redbook audio has so much more fidelity that the recording and mastering mistakes that are inaudible on vinyl become apparent on CD.

In actual double-blind, controlled testing, no human being has ever been able to reliably tell the difference between a 24/96 or 24/192 audio source and the same source downsampled to 16/44.1.

Edit: I don't know what sources you're referencing, but no reliable source that I've ever seen demonstrates human hearing outside of the Fletcher-Munson range. Dogs can hear up to around 60 kHz. Cats can hear up to around 80 kHz. But humans? Not even close. In fact, most adult humans with healthy hearing can't hear anything above about 16 kHz.

Edit 2: I found some references to people who claim to be able to hear things like ultrasonic rodent repellers. These sweep from about 30 kHz to about 60 kHz at high volume levels. Inaudible to humans. What they're hearing are lower frequency reflections from walls and furniture. In other words: comb filtering noise.

Not a scientific fact but an interesting piece of trivia:  my wife claimed that she could hear an ultrasonic dog repeller claimed to operate at 25-28 kHz.  At age 30 at the time.  This was outside so we can discount most bounce-back at lower frequencies.  I didn't hear a thing and my audiologist last week confirmed that my hearing still goes to 17 kHz at age 40 so I'm not doing too badly.  Obviously I didn't believe her so set up my own blind test where she couldn't see if I was pressing the button or not and had to indicate when the 'sound' started and stopped while standing outside with her back to me at a distance of 4 meter.  She never missed once.  And her name's not Tracy, she walks on 2 legs and talks.  I have no way to tell if the device actually emits at this frequency or a lower one but still.  Acute hearing has been a curse for her all her life, which is one of the reasons why I listen via headphones, not speakers.

 

[bigshot]

Sensing sound pressure isn't the same as hearing sounds. It's likely that the ultra high frequencies were at a very loud volume, and it was creating vibrations in the fluid inside her head. That can be uncomfortable. They use super audible loud speakers for riot control sometimes. Instant headaches for everyone in the path of the sound waves.
 
In recorded music, there is very little to hear above 16kHz. You could dial the top end off entirely using an equalizer, and it wouldn't make much of a difference. The only stuff up that high is upper harmonics on cymbal crashes and triangles.

 

[Xenophon]

  Sensing sound pressure isn't the same as hearing sounds. It's likely that the ultra high frequencies were at a very loud volume, and it was creating vibrations in the fluid inside her head. That can be uncomfortable. They use super audible loud speakers for riot control sometimes. Instant headaches for everyone in the path of the sound waves.

At any frequency, what else is sound other than a pressure wave propagating through air hitting the diaphragm in one's ear?  But I agree that it must have been 'loud', went inside, lit a candle and directed the device at the flame from a short distance, it was easy to see the displacement.