A Phoenix record store owner set the audiophile world on fire • Hacker News Board

Re: A Phoenix record store owner set the audiophile world on fire

#41

Post by kazinator » Fri, Aug 05, 2022, 7:40 PM UTC

doix wrote: post #33

jrajav wrote: post #12

Earlier quoted context omitted.

44100hz sample rate is enough to reproduce well above the highest frequency even young humans can hear, and 16 bits of depth is enough for 96db of dynamic range, enough to make the noise floor of any consumer system totally inaudible. Standard CDs might not be good enough for archival and further production work (where you also need wiggle room for further processing or format changes), but they have plenty of headro…

So it's been many years since I looked at this stuff, but the reason for higher sample rate isn't to do with humans being able to hear over 20khz. If you sample at 44khz you need a high pass filter at 22khz to avoid aliasing. Since that's pretty close to 20khz, it needs a pretty sharp drop off. Having a filter with a sharp drop off can introduce artifacts. With 96khz sampling rate, you don't need as a sharp drop off…

You are correct; with a 96 kHz sampling rate, you don't need such a brick wall filter. But then you dohn't have to actually keep the data at 96 kHz; it can be reduced to 44, this time using a digital filter.

A sharply cutting off analog filter is expensive to produce. It has multiple stages to create the multiple poles. High precision resistors and capacitors have to be used to get all those circuit stages to line up. The filter will have phase distortion.

That's the basis of "supersampling": sampling at a higher rate with a simpler filter with less of a cutoff, then completing the job with a digital filter to get to the target sample rate.

This can be done in reverse, in reproducion. Take, say, a 48 KHz signal, and digitally interpolate it to a higher sample rate like 96 KHz. That is fed to the DAC. The filter after the DAC then doesn't need such a steep cutoff after 20 KHz. A greater bit depth can be used; like 16 bit samples interpolated to 24 bit at a higher rate, fed to a 24 bit DAC.

The digital filter or interpolator doesn't care about accurate resistors, capacitors or drift in component values over time or due to heat; it does the same thing with the same data every time.

Digital filters can look at future values also. The state of an analog filter is determined by only the current and past values of the signal; but digital signal processing can delay the signal a little bit and look at a "box" around the current value. I think that is key to preserving phase relationships.

Re: A Phoenix record store owner set the audiophile world on fire

#42

Post by MivLives » Fri, Aug 05, 2022, 7:40 PM UTC

nerfbatplz wrote: post #28

gwern wrote: post #2

"The fallout of the MoFi revelation has thrown the audiophile community into something of an existential crisis. The quality of digitized music has long been criticized because of how much data was stripped out of files so MP3s could fit on mobile devices. But these days, with the right equipment, digital recordings can be so good they can fool even the best ears. Many of MoFi’s now-exposed records were on Fremer and…

Not only can you not tell the difference between analogue and digital, you also cannot tell the difference between lossless and properly transcoded MP3. Most audiophile mania is poorly informed at best and astrology at worst. http://abx.digitalfeed.net/

Badly transcoded mp3s are the worst. For most artists I couldn't tell the difference between a v2 and a flac. There were a few who you could but they're mostly not what people are listening to (intentionally very noisy artists). Realistically if I just heard them blind I don't think I would have cared.

Re: A Phoenix record store owner set the audiophile world on fire

#43

Post by kstrauser » Fri, Aug 05, 2022, 7:40 PM UTC

mod wrote: post #31

jrajav wrote: post #16

Earlier quoted context omitted.

Most of the time people will argue that vinyl has "warmth" and this makes it subjectively better. This is true, but it's actually because of the pleasing distortion record players add to the music. Being distortion, even if it improves your listening experience (and this depends on genre, though it generally does), it is certainly objectively worse in terms of accurate reproduction.

Can't the warmth just be reproduced in a digital player anyway?

Yup. Boost the mids on your EQ and you're halfway there.

Re: A Phoenix record store owner set the audiophile world on fire

#44

Post by masklinn » Fri, Aug 05, 2022, 7:41 PM UTC

mod wrote: post #31

jrajav wrote: post #16

Earlier quoted context omitted.

Most of the time people will argue that vinyl has "warmth" and this makes it subjectively better. This is true, but it's actually because of the pleasing distortion record players add to the music. Being distortion, even if it improves your listening experience (and this depends on genre, though it generally does), it is certainly objectively worse in terms of accurate reproduction.

Can't the warmth just be reproduced in a digital player anyway?

Well yes but that's not actually what the intent was.

Though I would expect it'd be more of an eq profile thing.

Re: A Phoenix record store owner set the audiophile world on fire

#45

Post by allears » Fri, Aug 05, 2022, 7:42 PM UTC

People who claim CDs have more dynamic range than analog are missing the point. Think of it in terms of graphics. 8 bit graphics can represent totally black, or totally white. That's a complete dynamic range. But it's the in-between shades that require more bits to specify.

It's the same with audio. It's not how soft or how loud a recording can get. It's the subtle shades of variation. 16 bits is pretty good, but not as good as 24 bits. You can tell by listening to the shimmer of a cymbal, or a plucked string fading to silence. You may not experience it as "stair-stepping," but there's an added degree of realism that more bits can confer.

And similarly about frequency range. Any audio system or media has more linear reproduction in the middle part of its range. So if you design a system that exceeds the range of human hearing, you'll get more linearity within the audible part of that range.

Re: A Phoenix record store owner set the audiophile world on fire

#46

Post by com2kid » Fri, Aug 05, 2022, 7:42 PM UTC

mod wrote: post #31

jrajav wrote: post #16

Earlier quoted context omitted.

Most of the time people will argue that vinyl has "warmth" and this makes it subjectively better. This is true, but it's actually because of the pleasing distortion record players add to the music. Being distortion, even if it improves your listening experience (and this depends on genre, though it generally does), it is certainly objectively worse in terms of accurate reproduction.

Can't the warmth just be reproduced in a digital player anyway?

Yes, the "warmth" generated by from literally any possible analog circuit can be reproduced digitally.

Physics doesn't care about people's beliefs, but sadly many people also don't care about physics.

Re: A Phoenix record store owner set the audiophile world on fire

#47

Post by Jaxan » Fri, Aug 05, 2022, 7:43 PM UTC

amelius wrote: post #39

kstrauser wrote: post #5

Earlier quoted context omitted.

That whole line of thinking just blows me away. It's possible to accurately sample and reproduce any vinyl recording in the digital realm. It's absolutely not possible to accurately record every digital recording onto vinyl. For instance, it's not hard at all to make bass so loud that it causes the needle to leap out of the track, or treble so high that the needle can't accurately track the high frequency vibrations.…

> On recording, you cut the bass and jack up the treble. On playback, you boost the bass and reduce the treble. Sounds like if you combine the two steps you end up with the identity operator :)

That’s the point!

Re: A Phoenix record store owner set the audiophile world on fire

#48

Post by defterGoose » Fri, Aug 05, 2022, 7:43 PM UTC

iforgotpassword wrote: post #26

kstrauser wrote: post #5

Earlier quoted context omitted.

That whole line of thinking just blows me away. It's possible to accurately sample and reproduce any vinyl recording in the digital realm. It's absolutely not possible to accurately record every digital recording onto vinyl. For instance, it's not hard at all to make bass so loud that it causes the needle to leap out of the track, or treble so high that the needle can't accurately track the high frequency vibrations.…

That's the main reason I really like vinyl. Physics prevents the mastering engineers from going overly crazy. Vinyl often sounds more carefully mastered, even though nothing would technically prevent you from releasing the same thing in digital form. The other thing I like is that it really takes more effort to put on a record than start a track on Spotify. I really only do it when I can take the time to do so. Pick…

It's still true that it takes a skillful ear/hand to do that mastering. I've heard terribly mastered older (not abused) records and modern releases that are excellent. A good original copy of Rumours is about as good as it gets IME.

Vinyl _does_ also provide a nice soft upper limit to the loudness wars.

Re: A Phoenix record store owner set the audiophile world on fire

#49

Post by colejohnson66 » Fri, Aug 05, 2022, 7:44 PM UTC

amelius wrote: post #39

kstrauser wrote: post #5

Earlier quoted context omitted.

That whole line of thinking just blows me away. It's possible to accurately sample and reproduce any vinyl recording in the digital realm. It's absolutely not possible to accurately record every digital recording onto vinyl. For instance, it's not hard at all to make bass so loud that it causes the needle to leap out of the track, or treble so high that the needle can't accurately track the high frequency vibrations.…

> On recording, you cut the bass and jack up the treble. On playback, you boost the bass and reduce the treble. Sounds like if you combine the two steps you end up with the identity operator :)

That’s the idea, in theory. But because analog components such as the resistors, capacitors, and transistors will not be identical[0] between the master and player, it’s not perfect.

[0]: you can get very low tolerance resistors and capacitors, but they’ll cost you

Re: A Phoenix record store owner set the audiophile world on fire

#50

Post by kstrauser » Fri, Aug 05, 2022, 7:45 PM UTC

amelius wrote: post #39

kstrauser wrote: post #5

Earlier quoted context omitted.

That whole line of thinking just blows me away. It's possible to accurately sample and reproduce any vinyl recording in the digital realm. It's absolutely not possible to accurately record every digital recording onto vinyl. For instance, it's not hard at all to make bass so loud that it causes the needle to leap out of the track, or treble so high that the needle can't accurately track the high frequency vibrations.…

> On recording, you cut the bass and jack up the treble. On playback, you boost the bass and reduce the treble. Sounds like if you combine the two steps you end up with the identity operator :)

To a point, but remember that you're dealing with low-bit floating point math. That boosting stage creates information that wasn't originally there. That's not so bad in the treble case where original * boost / filter is pretty close to the original. But in the bass case, original / filter * boost might be significantly different than the original. That's usually not a huge problem because the low bass frequencies are pretty forgiving, but it's something to consider.