The comedy of some Hi res recordings

[Hammeredklavier]

I usually pass 'modern' ABX tests 320kbps or VBR vs RBCD but with some constraints

- on my "big" systems (I am not a headphone guy).
- actively paying attention to details.
- on the type of music I like.
- on good recordings.
- on recordings or configurations (say a chamber music quatuor) I am familiar with.

A lot of constraints indeed...

On the other hand, when I play random tracks from my NAS, where I have stored my music collection (mostly rips of my own CDs, a few hires downloads) I can always almost instantly recognize my old 320 kbps files from my lossless ones. My ripping evolved over time, I started when encoding to 128 kbps was actually taking much longer than listening to the CD and when storage was expensive, only digitized everything to 320 kpbs a bit later, then finally went for a lossless ripping marathon... And since I am a fundamentally lazy person, I never cleaned up my mess which means I have duplicate 320kpbs/RBCD of nearly my entire library.

It may be subjective, or they may be objective reasons I am not aware of in the encoding behavior of old encoders, but the giveaway is almost always the width and depth of the sound stage in the case of my library. That really jumps in my face. Sometimes artifacts I have trained myself to detect too, but that requires paying attention.

That leads me to think either the encoder or the parameters choice weren't optimal when I did the 320kbps pass. But that was, of course, a time where legit and non-legit encoders proliferated, where some commercial encoders simply stole encoding libraries from open source projects and vice-versa.

Ultimately, I don't really want to worry about how the 320 kbps stream was generated and stick to RBCD, which has the added advantage of having convinced me it was all I needed for biological and mathematical reasons. As far as the recording/mastering process is concerned, on the other hand, I am all for giving engineers all the data they need and more, if simply to avoid things like the dirty voice track introduced in Amir's example.

And, the consistent instinctive choice, in multi-channel situations, would be to give every channel RBCD bandwidth. One less thing to worry about even if I don't really know and haven't investigated if that is even necessary (probably not on the whole frequency range anyway).

If you can reliably ABX MP3 vs RBCD at 320 kbps then I'm impressed. My hearing's pretty good and I don't think I'd stand a chance, unless a carefully chosen killer sample was used. Certainly if recent versions of the LAME encoder or the one by Fraunhofer were used! Very early MP3 encoders could sound pretty rough at lower bit rates I'll grant you.

I'm not saying LAME is transparent at @120 kbps variable, but I'd have to be paying attention and I'd probably find the sound quality perfectly acceptable at that. With storage being so cheap I honestly haven't tried it for a bit.

 

[PierreV]

@Hammeredklavier Given the constraints listed above, yes. But definitely not on a 15 sec Suzanne Vega, Eagles or Daft Punk sample

It's not that I am trying to claim a Golden Ear (can't hear much above 14.5 kHz for example) and that is why I am perfectly willing to concede pre-emptively that my 320kbps rips were not optimally done.

That being said and, at the risk of sounding a bit pretentious, subjective or whatever, 192 kbps VBR is absolutely, definitely unpalatable on my large setups. I even stopped listening to Radio Paradise and only timidly came back when they introduced 320 kbps. That's mostly a question of speakers imho. The whole scene/impression is totally different with 192kbps. Note: except when I bring the Linn preamp/amp combo in for convenience/features, my "large" setups are usually 95% speakers, 5% rest in terms of cost.

BTW, I used one of those setups with Qobuz Hires streaming and the 2L sample files to convince a friend not to spend on 'Sublime'/HiRes/Cables and I discontinued my own hires streaming (kept CD quality). Couldn't hear benefits.

 

[Hammeredklavier]

That being said and, at the risk of sounding a bit pretentious, subjective or whatever, 192 kbps VBR is absolutely, definitely unpalatable on my large setups.

I haven't tried for years, I find the ABX tests irritating. Using recent versions of LAME I don't think I'd stand a cat in hell's chance at 320 kbps though!

It doesn't give me too much trouble to believe that the differences between the original and the MP3 are easier to hear in very quiet rooms and at very high volumes. I normally used headphones so had to keep the levels sensible. That said, it's one thing being able to detect a difference and another to find the MP3 unpleasant! Using a decent, modern encoder, anything over 120 kbps VBR sounds perfectly nice enough for me to enjoy my music.

 

[svart-hvitt]

FWIW, I never use low resolution...except when listening to radio, www.nrk.no, @192 kb/s. The sound (of music) is clearly below 16/44. Flat and lacking. Could be due to radio remastering and compression? But sound is undoubtedly second rate and much better when streamed through Tidal/Qobus in 16/44 or higher.

 

[RayDunzl]

It may be subjective, or they may be objective reasons I am not aware of in the encoding behavior of old encoders, but the giveaway is almost always the width and depth of the sound stage in the case of my library. That really jumps in my face

That's what bothered me too, back whenever it was.

It was enough to keep me from "collecting" lossy material.

Now, don't know. I haven't paid attention in that way to known lossy sources in that way for a long while, but haven't "noticed" it, either.

I casually listen to HDRadio, which, as implemented, has varying bit rates among the channels a station may enable/allocate.

---

Wow. 48kb/s. I hereby declare myself deaf. The local transmitter farm is very close to me - about 4 miles.

To: me
Sent: Friday, March 11, 2011 6:18 PM
Subject: RE: WUSF HD Radio - Broadcast Specs

Hi and thanks for your e-mail.

We use the 48kbps dynamic MP3 setting for the HD encoding. This is our best option because we need two streams of audio (one news and one classical) and bandwidth for Navteq traffic data. If you have a GPS unit with the traffic function built in, it looks to our frequency in this market to get that information. They pay to lease this space from us through a collaboration of NPR affiliates across the country.

We are currently transmitting the HD at 1% of our analog power which was the maximum allowed by the FCC until 2010 when they increased the injection level to 10%. We have received a grant to purchase a new FM/HD transmitter and once it’s installed, we will be able to go to 10%. I can’t tell you when it will be on the air, but we had projected to complete this project sometime in the second half of this year. Given the current budget negotiations at the state and federal level, this is all in flux at the moment.

To get more info on HD radio, look up IBOQ or IBiquity on the internet. You will get a complete rundown of the specs of the HD-Radio system in the US.

Mike

 

[GlennS]

Divide sampling frequeny by 2 and there is your upper frequency limit.
For dynamic range use this table...

I know these stats. I was meaning that it would help us if the site selling the digital music would give stats from the recording being sold. So if we saw that the recording had no content above 20kHz, then we would know not to buy the higher resolution recording (of course this would hurt the pocketbook of the seller so they would have no incentive to do this) unless the higher res version was re-worked to have other benefits.

Also to any other posts regarding the instructional video about digital audio:
Yes I'd seen it several years ago and I understand the science behind the sampling rate and the freq range constructing a waveform. But every explanation uses a simple single waveform. Compared to a waveform created by a tone generator, I always assumed that real world audio is much more complex with multiple waveforms overlapping each other. Therefore it made sense to me that recording sampling points more often would result in smoother replay. But you all have explained that 44.1kHz sampling records everything needed within 20kHz freq no matter how complex the audio.
Thank you RayDunzl for your explanation.

 

[RayDunzl]

Compared to a waveform created by a tone generator, I always assumed that real world audio is much more complex with multiple waveforms overlapping each other.

Five tones, a little pink noise, overlapped, mixed to a single track:

Spectrum of the mix:

 

[PierreV]

So if we saw that the recording had no content above 20kHz, then we would know not to buy the higher resolution recording..."

But you already know that, unless very young and in a special laboratory environment, you will not hear any content above 20kHz...

 

[andymok]

It's not about levels at which frequencies go to, it's about how far in the freqeuncy range frequency components of the original waveform reach to.

As we can hear up to 20kHz we don't need to record frequency components higher than those, and to do that 44100 samples per second is sufficient.

You should also notice that we, for the same reason, cannot distinguish difference between a violin and a guitar playing 20kHz tone as we cannot hear the harmonics (frequency components of the original tone) that differentiate those two waveforms.

Things become "toneless" beyond 5k, doesn't mean we can't tell between a cymbal and a triangle

 

[GlennS]

But you already know that, unless very young and in a special laboratory environment, you will not hear any content above 20kHz...

You can't hear below ~25hz, but you can still feel it which adds to the overall experience.
Do we know for a fact that content above 20kHz is completely imperceptible to our overall experience?
When available, I do buy higher res audio just in case higher frequencies can affect (effect?) our senses (or hopefully the higher res version contains other benefits such as remastering). If I know for a fact that higher res audio has no added benefits, then RBCD is fine for me.

 

[sergeauckland]

I know these stats. I was meaning that it would help us if the site selling the digital music would give stats from the recording being sold. So if we saw that the recording had no content above 20kHz, then we would know not to buy the higher resolution recording (of course this would hurt the pocketbook of the seller so they would have no incentive to do this) unless the higher res version was re-worked to have other benefits.

I think you've answered your own question. Of course the Record Company won't tell you their HiRes recording is a fake. As to reworking to have other benefits, that's called Remastering, or even in extreme cases Remixing and Remastering, and record companies do it all the time. Where it's a complete con is taking old analogue material from the 1970s and remastering it and issuing it as HiRes, when it can't possibly, under any circumstances, be HiRes.

S.

 

[RayDunzl]

Things become "toneless" beyond 5k, doesn't mean we can't tell between a cymbal and a triangle

Perhaps it is the lower frequency content that distinguishes them.

http://www.drummerworld.com/forums/...89b0f940d8cf89df8493ca20&p=745320&postcount=2

Cymbal crash (higher frequencies may not be included here - 100kHz or beyond I believe has been noted elsewhere)

Can't find a Triangle's spectrum, but, it won't look like that.

 

[PierreV]

You can't hear below ~25hz, but you can still feel it which adds to the overall experience.
Do we know for a fact that content above 20kHz is completely imperceptible to our overall experience?
When available, I do buy higher res audio just in case higher frequencies can affect (effect?) our senses (or hopefully the higher res version contains other benefits such as remastering). If I know for a fact that higher res audio has no added benefits, then RBCD is fine for me.

Bass: resonance and air wave pressure, which don't happen, at least in a way we can feel or hear, on the high end.

On the ultrasonic side, we are bathed in them for a bunch of reasons. The question is - do they hurt us covertly?

This paper generated a few headlines and some controversy a few years ago.
https://royalsocietypublishing.org/doi/full/10.1098/rspa.2015.0624

In the end, it always depends on total energy applied/absorbed.

Oh, and if it does, wireless charging devices are a major health issue and even short range ones should never be in the same house as your music

This, for example, is a bit worrying

uBeam has developed a high-powered air-coupled ultrasonic transducer to transmit and receive sound waves at a single frequency within the range of 45kHz to 75kHz with an output of 145dB to 155dB (or 316 W/m2 – 3kW/m2)

from here

https://techcrunch.com/2015/11/07/wireless-power-charger/#.v1vbjhj:Ckow

The science is very much open there ...

But even if we assume those ultrasonics do something (which isn't established at low power) to the sound, how do we know in doesn't degrade it?

It is always the same in the audio industry: expensive cables are described as "lifting the veil" but for some reasons, no cable ever "drops the curtain"

 

[restorer-john]

This, for example, is a bit worrying

uBeam has developed a high-powered air-coupled ultrasonic transducer to transmit and receive sound waves at a single frequency within the range of 45kHz to 75kHz with an output of 145dB to 155dB (or 316 W/m2 – 3kW/m2)

Domestic cats can hear up to 70KHz...

 

[Shadrach]

You can't hear below ~25hz, but you can still feel it which adds to the overall experience.
Do we know for a fact that content above 20kHz is completely imperceptible to our overall experience?
When available, I do buy higher res audio just in case higher frequencies can affect (effect?) our senses (or hopefully the higher res version contains other benefits such as remastering). If I know for a fact that higher res audio has no added benefits, then RBCD is fine for me.

I don't think that's right. I think some people can hear down to 18hz.
Does your stereo reproduce sounds above 20kHz?
I'm just curios.

 

[GlennS]

Does your stereo reproduce sounds above 20kHz?

Whatever the lowest freq that a person can hear, the point is that just because you can't hear it, doesn't mean that it doesn't have a noticeable affect (effect?).
According to the spec sheet, my Onkyo has a range of 5Hz-100kHz.
I assume my separate amp (ATI 1506) would be similar to my Onkyo.
My speakers... not sure how high they can reproduce.

 

[Blumlein 88]

Whatever the lowest freq that a person can hear, the point is that just because you can't hear it, doesn't mean that it doesn't have a noticeable affect (effect?).
According to the spec sheet, my Onkyo has a range of 5Hz-100kHz.
I assume my separate amp (ATI 1506) would be similar to my Onkyo.
My speakers... not sure how high they can reproduce.

Sound at 100 khz is absorbed 3.28 db per meter. So even if speakers make the sound not much will reach you.

 

[Frank Dernie]

Sound at 100 khz is absorbed 3.28 db per meter. So even if speakers make the sound not much will reach you.

Doesn’t the beam also get narrower as frequency increases too on most tweeter designs meaning unless you are precisely listening where both tweeter axes intersect you also won’t hear it?
I know the NXT tweeters aren’t supposed to beam but all others do don’t they?

 

[March Audio]

Doesn’t the beam also get narrower as frequency increases too on most tweeter designs meaning unless you are precisely listening where both tweeter axes intersect you also won’t hear it?
I know the NXT tweeters aren’t supposed to beam but all others do don’t they?

It does indeed.

Both of these issues were a major PITA when I was involved a few years back with some research into the effects of high frequency industrial noise on Orange Leaf nosed bats. !

https://environment.des.qld.gov.au/wildlife/animals-az/micro-bats/orange_leafnosedbat.html

 

[Krunok]

Things become "toneless" beyond 5k, doesn't mean we can't tell between a cymbal and a triangle

Cymbal plays in the 3-5kHz range. Triangle is a pitchless instrument as its base tone is masked with non-harmonic overtones.
That's why you can tell the difference..