[interpolate]

High-fidelity audio seems to be the latest "fad" by the powers at be and whilst the advantages of 24/96 hi-fi sound is not obvious to the casual listener playing back on typical equipment, what about us lot with higher end equipment or studio quality reference gear? 
 
I downloaded the FLAC 24/96 version of Pink Floyd's last two albums although I already have them. The only way to tell a difference is by using a spectrogram image which represents different frequencies by colour mapping the sound. This technique gets used to edit unwanted sound like hiss, coughs, vinyl clicks etcetera. However back in the world of audiophile listener this information should be neither here or there.
 
Although in retrospect, there does seem to be a little je ne sais quoi of difference when listening to the Pink Floyd albums. There seems to be  a subtle difference in the way instruments sound. Possibly the direct high-resolution transfer without any dither or minimal dither allows for more sonic detail or this is a placebo effect.
 
As an experiment I created DSF files from certain tracks on the Pink Floyd albums and played it back on my FiiO X3 ii and AKG K702 using a High Gain Output. Personally I would say there is something there although sometimes I don't think it's enough to justify paying another 50% on the price of music.
 
Muse, Mumford & Sons and Florence & Machine have recently released high-definition 24/96 versions of their albums which is a turn up for the books. 
 
Now when you bring science into this and the benefits of anything over 44.1Khz being useful it becomes a large multi-page debate. Nyquist/Shannon listening curves aside I think if you enjoy it you should listen to whatever format you wish. If there is no real benefit to all of this, I can't help thinking we might be getting conned by clever marketing Like when all these teenagers suddenly discovered MP3 and were convinced Apple created it.
 
Are we kidding ourselves or should we embrace the high-resolution revolution?
 
I am thinking of repurchasing some The Doors albums in high-resolution FLAC although just wanted to see other people's viewpoint on this before committing to a buying decision.

 

[arnyk]

  High-fidelity audio seems to be the latest "fad" by the powers at be and whilst the advantages of 24/96 hi-fi sound is not obvious to the casual listener playing back on typical equipment, what about us lot with higher end equipment or studio quality reference gear? 
 
 

 
The current battle of high rez audio commenced near the end of the previous century with the advent of a DVD-A and SACD.
 
The industry itself arguably torpedoed the formats by glutting their new releases with material sourced from low resolution sources.
 
Unfortunately, neither the audiophiles nor the high end reviewers ever noticed this for at least 5 years until some skeptics used FFT analyzers to discern the provenance of all these allegedly high rez audio recordings.  
 
The percentage of material in high rez formats sourced from low rez sources was found to be about 50%.

 

[interpolate]

I had a couple of DVD-A before although I don't think the source is used for playback was really adequate to appreciate it. As for SACD, I just don't have the playback equipment for that either.
 
It's a technology I've been aware of although never took advantage of. Hybrid SACD disks are a bit of mystery also. What frustrates me is I can't get always get the music I want from bands I like in High Resolution format. As a hobbyist musician my main objective will be to make higher resolution formats available as well as the usual suspects.

 

[Music Alchemist]

High resolution is not high fidelity. As has been documented countless times, the only thing that matters in terms of playback is the recording, master, etc. itself. High resolution audio formats do have benefits for studio work that requires advanced computer processing, but not for listening to music. "HD" music downloads only sound different from some of their CD counterparts because they use a different master. (It's just a marketing trick.) If you convert those "HD" files to Red Book (16-bit / 44.1 kHz), they sound exactly the same. The reason for this is simple. All 24-bit does is add more dynamic range, but 16-bit already has more than enough dynamic range to handle all recordings in existence. 44.1 kHz is designed to play all the frequencies we can hear. (Roughly 20 Hz to 20 kHz.) Anything higher than that is inaudible. So don't worry about formats; worry about the recording and the master. (You also don't need to create multiple threads on this topic.)

 

[Steve Eddy]

"High fidelity" has two contexts.

The first is upstream, where the fidelity is ostensibly to the performance that is being recorded, mixed and mastered.

The second is downstream, where the fidelity is to the end product, the mastered recording.

High resolution digital is relevant to the upstream context, but not the downstream context. At least no one to date has shown this to not be the case.

se

 

[arnyk]

"High fidelity" has two contexts.

The first is upstream, where the fidelity is ostensibly to the performance that is being recorded, mixed and mastered.

The second is downstream, where the fidelity is to the end product, the mastered recording.

High resolution digital is relevant to the upstream context, but not the downstream context. At least no one to date has shown this to not be the case.

se

It's easy for a person who understands recording technology well to understand that high resolution audio is relevant to neither the upstream context or the downstream context because the limiting technologies are the ears of the listener, the dynamic range of the recording venue, and the dynamic range of the playback venue. That covers both upstream and downstream.

 

[Music Alchemist]

  It's easy for a person who understands recording technology well to understand that high resolution audio is relevant to neither the upstream context or the downstream context because the limiting technologies are the ears of the listener, the dynamic range of the recording venue, and the dynamic range of the playback venue. That covers both upstream and downstream.

 
Perhaps you never noticed that recording studios use 24-bit and higher when editing audio.

 

[arnyk]

  High resolution is not high fidelity. As has been documented countless times, the only thing that matters in terms of playback is the recording, master, etc. itself. High resolution audio formats do have benefits for studio work that requires advanced computer processing, but not for listening to music. "HD" music downloads only sound different from some of their CD counterparts because they use a different master. (It's just a marketing trick.) If you convert those "HD" files to Red Book (16-bit / 44.1 kHz), they sound exactly the same. The reason for this is simple. All 24-bit does is add more dynamic range, but 16-bit already has more than enough dynamic range to handle all recordings in existence. 44.1 kHz is designed to play all the frequencies we can hear. (Roughly 20 Hz to 20 kHz.) Anything higher than that is inaudible. So don't worry about formats; worry about the recording and the master. (You also don't need to create multiple threads on this topic.)

 
To be more specific there is one really clear case where high sample rates can affect sound quality, and that is where music is being generated by synthetic means and high sample rates help avoid audible aliasing when large amounts of nonlinear distortion is being added. An example of this would be a synthesized guitar note being processed by a synthesized fuzz box. 

 

[Music Alchemist]

  To be more specific there is one really clear case where high sample rates can affect sound quality, and that is where music is being generated by synthetic means and high sample rates help avoid audible aliasing when large amounts of nonlinear distortion is being added. An example of this would be a synthesized guitar note being processed by a synthesized fuzz box. 

 
There are no examples to support this in the context of playing back digital audio files, though.

 

[arnyk]

   
Perhaps you never noticed that recording studios use 24-bit and higher when editing audio.

 
I've noticed that it is not uncommon, but it is also not a general rule. 
 
But that's just custom, not a theoretical proof.
 
Audio editing is almost always just linear mixing, involves audio signals with more like 12-14 bit actual resolution, and so there is no audible benefit to even 16 bits.
 
I've done a great deal of live recording with professional grade CD burners because we delivered the finished recordings immediately at the end of the performance. Working with "just" 16 bit data was never a problem, but keeping the peaks within a few dB of FS with unrehearsed performances was. Of course more bits would have not helped that at all.

 

[interpolate]

Fidelity comes from word fidelis which is Latin for faithful. Just to clarify things I know what the two words mean.
 
Evidently if the master is bad, you can't polish a turd whether it be 32Khz or 96Khz.
 
If you downsample content to a lower sample-rate and it sounds the same, then the job is done correctly. I do know about the human hearing threshold and Nyquist perceived hearing of sounds. Although I do know some things, it's just not everything.
 
With that in mind, is there any need to patronise me like I just discovered water. 

 
 
Just messing

 

[interpolate]

Just for the record, I always use 24-bit (32-bit) for initial recording and then bounce down to 16-bit & dither if necessary although as a rule the average listener won't notice the difference.  

 

[Steve Eddy]

It's easy for a person who understands recording technology well to understand that high resolution audio is relevant to neither the upstream context or the downstream context because the limiting technologies are the ears of the listener, the dynamic range of the recording venue, and the dynamic range of the playback venue. That covers both upstream and downstream.

I prefer to record 24 bit because I'd rather take a clean 24 bits down to 16 bits (where I can use the whole 16 bits) than risk clipping when recording with 16 bits or making sure all of the peaks are kept well below 16 bits to avoid clipping.

Same reason I shoot full resolution RAW before editing and converting to a lower resolution JPG.

se

 

[interpolate]

There are better formats than JPEG like BMP or TIFF which in a way are like the FLAC & ALAC of the image world.  [I was making a comparison].

 

[cjl]

   
I've noticed that it is not uncommon, but it is also not a general rule. 
 
But that's just custom, not a theoretical proof.
 
Audio editing is almost always just linear mixing, involves audio signals with more like 12-14 bit actual resolution, and so there is no audible benefit to even 16 bits.
 
I've done a great deal of live recording with professional grade CD burners because we delivered the finished recordings immediately at the end of the performance. Working with "just" 16 bit data was never a problem, but keeping the peaks within a few dB of FS with unrehearsed performances was. Of course more bits would have not helped that at all.

There are genuine advantages to recording and editing in 24 bit, specifically due to available headroom and combined noise floor (if you mix a bunch of 16 bit dithered files together). For playback, 24 bit offers no benefit though.