advertisement


An awful lot of truth in this...

Yes I agree, the noise floor is lower meaning greater resolution of even lower level signals - that’s what dynamic range is. However, if we’re playing music at normal levels it’s not possible to hear any more of this resolution, as 16 bit audio already has far more dynamic range than we’d ever be able to use for music listening.
16 bits is not far more than we'd ever be able to use. It is about enough if things are just right but when we add in a few real world concerns involving processing, level and the like 24 bits starts to look preferable in order to remove any concerns. 32 bit floating point would be perhaps better again these days since this gives almost 24 bit precision while removing concerns about level. The hardware limitations that lead to 16/44 becoming a standard disappeared decades ago.

44 kHz sampling does ask a fair bit for the filtering to be clean upto 20kHz as mentioned earlier. 48 kHz would ease this while 96 kHz would allow concerns about multiple stages of processing to be dropped. 32f/48 playback (if it is expressed like that) or 32f/96 for playback and recording/processing looks about right to me.

The problem with high sampling rates is that the signal may contain energy at inaudible frequencies above 20 kHz which could then interact with nonlinearities in the playback chain and be aliased into the audible frequency range and heard. I suspect if these inappropriate sampling rates do become common the more informed folk are likely to start using 'scratch' filters again!

Out of curiosity do turntable enthusiasts these days tend to use 'rumble' filters or have they disappeared along with tone controls? There can be similar issues with nonlinearities aliasing infrasonic signals into the audible range.
 
16 bits is not far more than we'd ever be able to use. It is about enough if things are just right but when we add in a few real world concerns involving processing, level and the like 24 bits starts to look preferable in order to remove any concerns. 32 bit floating point would be perhaps better again these days since this gives almost 24 bit precision while removing concerns about level. The hardware limitations that lead to 16/44 becoming a standard disappeared decades ago.

44 kHz sampling does ask a fair bit for the filtering to be clean upto 20kHz as mentioned earlier. 48 kHz would ease this while 96 kHz would allow concerns about multiple stages of processing to be dropped. 32f/48 playback (if it is expressed like that) or 32f/96 for playback and recording/processing looks about right to me.

The problem with high sampling rates is that the signal may contain energy at inaudible frequencies above 20 kHz which could then interact with nonlinearities in the playback chain and be aliased into the audible frequency range and heard. I suspect if these inappropriate sampling rates do become common the more informed folk are likely to start using 'scratch' filters again!

Out of curiosity do turntable enthusiasts these days tend to use 'rumble' filters or have they disappeared along with tone controls? There can be similar issues with nonlinearities aliasing infrasonic signals into the audible range.

Firstly, I never mentioned sample rate! But good points ;)

The recording chain is kind of separate from home listening though, and is probably where all this demand for ‘hi-res’ came from. People thought they must be being cheated out of the full 24 bit files, when in fact the 24 bits were just used for the editing side of things.

I still maintain that 24 bit files are irrelevant at home as our equipment does not have a low enough noise floor to make use of the last few bits, even if we had music files with sufficient dynamic range. Once we start using valve amps and playing at low levels it’s even more irreverent.
 
The recording chain is kind of separate from home listening though, and is probably where all this demand for ‘hi-res’ came from. People thought they must be being cheated out of the full 24 bit files, when in fact the 24 bits were just used for the editing side of things.

I still maintain that 24 bit files are irrelevant at home as our equipment does not have a low enough noise floor to make use of the last few bits, even if we had music files with sufficient dynamic range. Once we start using valve amps and playing at low levels it’s even more irreverent.

With growing amounts of DSP processing downstream of the delivered digitized music for level adjustment, active crossovers, room correction, tone controls, etc... a few bits of information can be lost. This could become audible in some cases. 23 or 24 bits is plenty to avoid any of this and the cost to use 24 or 32 bits for audio is negligible these days. Of course one can perform DSP tricks to limit the loss of information like accurately resampling to 32 bit floating point, processing, accurately converting back but simply moving to a more appropriate standard of 32 bit floating point values and eliminating all the messing about seems far more sensible and rational. Which is probably a strong reason to reject it given the way the current audiophile industry makes much of it's money.
 
I still maintain that 24 bit files are irrelevant at home as our equipment does not have a low enough noise floor to make use of the last few bits, even if we had music files with sufficient dynamic range.....
DAC designer Rob Watts says in one of his lectures that a huge improvement in sound quality in his DAC came by lowering the noise floor. As an owner of one of his products I would agree that the sound quality is excellent and take his word that lowered noise floor for the upsampled music tracks is one of the reasons. So I would disagree with Nagraboy's assertion that our equipment doesn't have a sufficiently low noise floor. My replay equipment does.
 
DAC designer Rob Watts says in one of his lectures that a huge improvement in sound quality in his DAC came by lowering the noise floor. As an owner of one of his products I would agree that the sound quality is excellent and take his word that lowered noise floor for the upsampled music tracks is one of the reasons. So I would disagree with Nagraboy's assertion that our equipment doesn't have a sufficiently low noise floor. My replay equipment does.

Just to be cheeky, I’d say that the noise floor of hi-fi enthusiasts’ credulity can always be lowered! ;)
 
I like Hans channel. Hes' another Youtube voice who deserves a wider audience. I've followed him for a while. His humour doesn't always work in a second language.. but at least he makes the effort and he seems very thorough when he reviews something.
 
  • Like
Reactions: GT
Studios don't record at 192/24 to capture “more sound”. They do so in order to allow the many hundreds of processing steps that will to be performed on the audio to be carried out without introducing audible noise. Every time you process, mix or modify a signal—even a digital signal—you introduce noise. At low bit-depths and data rates, this can quickly raise the noise floor of your resulting signal into the audible range.

The main technical argument for high-bitrate audio in a domestic setting is similar: that it will allow for more processing by replay equipment (for example to correct for room effects) without introducing audible noise. Theoretically, you could just up-convert the 44.1 original to 192, do your processing and then down-convert to 44.1 again, but starting with 192 gives you a much bigger margin of error. Providing a margin-of-error for pre-reproduction processing is the reason why cinema audio is coded at 192kHz: a modern cinema will have around thirty or forty speakers, each of which will be fed its own signal synthesized from the original six to eight channels of the audio. We're a few years away from this becoming ubiquitous in home reproduction, but it's already there at the high-end with the likes of B&O's Beolab 50.

The other, non-technical, argument for high-bitrate audio in a domestic setting is that in the last 20 years or so, the final mastering of a studio recording down to a 44.1/16k CD image has been accompanied by aggressive loudness processing: this makes the recording sound “big” on cheap equipment, on FM radio and in cars, but when played through anything with good reproduction (i.e., your hi-fi) it sounds crushed, lacking in dynamics and tiring to listen to. The advantage of “hi-res” recordings is that for many titles, the higher bitrate file gets you the album as it sounded before it was subjected to that heavy-handed dynamic range compression. I'd say that that's where about 99% of the arguments for “hi-res is a night-and-day difference” comes from.
 
I am not sure that 'hi-res' versions of albums are necessarily mastered better than the 16/44 versions. If they were you would think that they would be marketed as such.

It is verging on criminal the lack of information included in recording details. Many remastered items sound terrible.

Hifi News does a service to is with their analysis of hi-res recordings and what they really contain. Loads of them are effectively fake.
 
@KrisW

I've never heard it said that a hi-res file does not have the crushing compression applied. Where did you hear that? I've got plenty of 24/96 or 24/192 files that are audibly compressed compared with the same album on CD from decades earlier.
 
Forgive me for not understanding this. Is it because a compressed master was used, unlike when the CD was first produced?

They could well have been mastered from the same files but whether the final master for release is 16/44 or 24/192, the amount of compression (or any other processing) applied is an entirely unrelated issue.

Eg. If R.E.M.’s remastered albums that appeared on 24 bit DVD-A some years ago were sourced from the original analogue tapes they could potentially be an improvement on the old CDs, which also used the same tapes. But if the mastering engineer was told to crush the dynamic range to just a few dBs so they sound better on cheap earphones and Bluetooth speakers, then the ‘24 bit hi-res’ label is not much more than a marketing angle.
 
The advantage of “hi-res” recordings is that for many titles, the higher bitrate file gets you the album as it sounded before it was subjected to that heavy-handed dynamic range compression. I'd say that that's where about 99% of the arguments for “hi-res is a night-and-day difference” comes from.

Many titles? More common for the hi-res to be just as brickwalled IME. I am thinking of the likes of HD Tracks. Specialist SACDs etc are generally better.

Tim
 
Many remastered items sound terrible..

Indeed. Back in more gullible times, I bought quite a few remastered albums only to find many of them appalling.

To my mind it's one of the arguments in favour of physical media, if I'm daft enough to buy a remastered copy of Martin Stephenson's first album then choose to throw it in the bin in disgust, that's my choice.

If a streaming service chooses to 'upgrade' to the remastered files, chances are you – or at least your ears – are screwed.
 
@onlyconnect As an example of what I'm talking about, you could look at a release that was the worst example of CD range-crushing: Metallica's Death Magnetic. The 24-bit release is quieter, with much more dynamic range than the CD version. Of course, if they down-sampled that hi-res version to 44k/16, it would be just as good, and you shouldn't have to pay extra just to get a properly mastered recording, but there you go...

They could well have been mastered from the same files but whether the final master for release is 16/44 or 24/192, the amount of compression (or any other processing) applied is an entirely unrelated issue.

Eg. If R.E.M.’s remastered albums that appeared on 24 bit DVD-A some years ago were sourced from the original analogue tapes they could potentially be an improvement on the old CDs, which also used the same tapes. But if the mastering engineer was told to crush the dynamic range to just a few dBs so they sound better on cheap earphones and Bluetooth speakers, then the ‘24 bit hi-res’ label is not much more than a marketing angle.
The “24 bit hi-res” label is of course a marketing ploy, but it's one that comes at a price premium, so there's less incentive to master it to make it sound loud on cheap equipment. Where a release was available in SACD and CD, the 24-bit file tends to follow the SACD.

You mention the R.E.M. remasters, and these are actually pretty good, but it has to be said that the original CD releases by REM were also pretty well mastered, so the improvement is marginal. Comparing my rip of the original release CD of Automatic For The People with the 24-bit version on Qobuz, the newer release is a little more open, but then so is the "CD quality" version of that remastered release.
 


advertisement


Back
Top