The case for 96 kHz (and 88.2) vs Lower Resolutions (44.1 and 48)

QuadraphonicQuad

Help Support QuadraphonicQuad:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
OK.... Yet another thread of this...

I have always been an absolute defender of double blind tests. BUT you have to compare apples with apples.

The only comparison test that will give adequate results is:
- Have a source file Hi-res (24/96 or 24/192). This is the 'good' master produced.
- Downsample that file to CD resolution (16/44.1)
- Play alternatively both files with exactly the same equipment, same sound settings, same etc., same etc., same etc.
- I repeat same etc. again because a simple change even in just volume level may affect the non lineal frequency perception of the human hearing. And then you say... I like more this.

I guess many if not all would not notice any difference. And the statistics of good double blind tests would give nearly 50% due to random choices.

The master is what matters.

If I would have to sell Hi-res in all its forms, I would surely produce a master that sound better to audiophiles to be released in Hi-res formats. Am I the only one that would do that? Thats the way to have so many people that says the Hi-res format sound better.... I.e. that master in that Hi-res format sound better.

All this apply to comparison between Stereo CD and Hi-Res

Other considerations may apply to Atmos/TrueHD vs Atmos/DD+ where there are reduced bandwith to cope with the whole bunch of Atmos objects.
For the immersive case, I think streaming Atmos/DD+ works good for some electronic mixes with brief punctual sounds on each channel (less bandwith required). xPropaganda/Tidal, for instance.
But when there are more complex sounds/objects TrueHD vs DD+ has a quite perceptible difference for us with old ears.

Again, for the spatial audio, we have to wonder if Blu-ray/TrueHD and streaming DD+ come from the same master.

When pressing a vinyl there are several modifications that have to be made to the original master to "adapt" it to the vinyl format and avoid problems. High pass and Low pass filters, de-essing to cope with the RIAA eq, reduce stereo width for low frequencies. All that gives that "warmer" sound that we found on the vinyls. But the process has "destroyed" some original fidelity of the master. So the CD with the original master, perhaps untouched, will have more fidelity, but could sound harsher to someone. BUT hey... no hiss, no pops, no clicks. What a pitty... Lets simulate the "bad sound". Lets introduce hiss, lets create the "Lo-Fi" style.

Would something similar be requiered to transfer/render the original good spatial master to DD+? If the mix has many complexity with channels/objects, they would have to deal with reduced bandwith for all and compress lossy some sounds more than requiered. It's just a guess wether they follow that tunning process to adapt the master for streaming. In this case we are talking of an inferior format DD+ with respect to TrueHD, the same way vinyl is an inferior format with respect to CD.
 
OK.... Yet another thread of this...

I have always been an absolute defender of double blind tests. BUT you have to compare apples with apples.

The only comparison test that will give adequate results is:
- Have a source file Hi-res (24/96 or 24/192). This is the 'good' master produced.
- Downsample that file to CD resolution (16/44.1)
- Play alternatively both files with exactly the same equipment, same sound settings, same etc., same etc., same etc.
- I repeat same etc. again because a simple change even in just volume level may affect the non lineal frequency perception of the human hearing. And then you say... I like more this.
I agree that double-blind tests are the best (though not perfect) way to resolve such issues. The only additional feature I would add is that multiple sample sizes should be used and grouped according to how sensitive their hearing is, rather than testing one large random group. This helps to reduce another aspect of experimental bias: that a person's hearing generally declines with age. So an agreed preference for a source file amongst, say, three groups with the same hearing sensitivity yields a more robust result than a single random sample of individuals with a diverse age range (and hence, expected hearing sensitivity, which can easily be tested as well).
 
I record 96 or 192...it definitely helps when processing.

Once you actually bounce it down, I can't tell the difference between 48 and 96, and usually there aren't too many ultrasonics in my music, so I just bounce it to 48.

The difference between 16 and 24 bit is definitely there and 24 bit is a must always.

Whatever it is that you are doing, PM, your recordings sound superb.
 
I agree that double-blind tests are the best (though not perfect) way to resolve such issues. The only additional feature I would add is that multiple sample sizes should be used and grouped according to how sensitive their hearing is, rather than testing one large random group. This helps to reduce another aspect of experimental bias: that a person's hearing generally declines with age. So an agreed preference for a source file amongst, say, three groups with the same hearing sensitivity yields a more robust result than a single random sample of individuals with a diverse age range (and hence, expected hearing sensitivity, which can easily be tested as well).
Sure. The tests can be designed as much as rigurous as possible. But they will NOT contribute to clarify the fact that for the final listener the difference is minimal, if exists.
If there is interest in selling music in Hi-res format, it will be ensured that the masters are good.
And we, the consumers, will look for those editions not because of the format, but because they sound good, because it's a good master. Same thing would happen with MQA.

In the end, whoever puts the most money on the table prevails. Publications that push the audiophile criteria outweigh rigorous tests that may work against the "goodness" of Hi-res formats as a final product for listening. Mark Waldrep should have put all things in place, but there are other insterests that prevail.
 
Also, our ears do not hear the same all the time and adjust to the environment at hand. Take a look at this video from Paul at PS Audio on horn speakers starting at the 4:20 mark. He goes on to say that while the horn speakers initially sound "wrong," his ears adjust to them in about 15 minutes and then they sound natural.

 
Not another one of these threads! It should be obvious to everyone even the sceptics the the higher the resolution the better, even if you don't notice the difference. Overkill, so what with large hard drives high capacity medium and ultra fast download speeds why settle for less? there will always be those who think that mp3s are good enough and sound the same a high resolution audio. All the power to them if that makes them happy! Sound quality is cumulative, a reduction of quality in one area might go unnoticed but continue that through out the whole audio system and the negatives all add up.

I don't claim that I can hear a difference between lower resolution and higher resolution audio via my workbench Minimus speakers and seventies vintage power amplifier but on my main system I definitely do. Double blind tests don't necessarily prove anything either, if the system is veiled in some way i.e. speakers or electronics you won't hear any difference! Can you see through a blindfold, or hear clearly through ear muffs?

I record my vinyl at 192Khz and believe that the click repair and spectral repair functions work far better than if I was using a lower resolution. High resolution files can always be downsampled if need be for compatibility. Always start off with the highest resolution possible even if it seems like overkill and then downsample only if necessary.
 
This is one of those threads I shy away from as I am not experienced in this type of discussion, but I do have a history of buying HiRes down loads, etc.
I'll keep it simple.
1. I believe a early release RBCD, year 1990's can sound better than a new remastered CD with extra tracks even when released in a 24bit format.
2. When I listen to 24 bit, higher Khz, say 96 or 192, I do tend to hear a better mid range quality, a sense of softness for lack of a better explanation.
3. Today's example: Megadeth's famous album called Countdown To Extinction was released today for first time by HDTRacks in 192/24, I purchased it because for me it is an iconic album that I want. My other is an RBCD with Bonus Tracks released a few years ago. I sold my original RBCD years ago.
The Dr's for the RBCD are 6's, 7's, 8's, 9. The new DR's for todays release in 24 bit are one notch lower at only 6's and 7's.
I am aware of placebo, but as I am not a beginner I have the ability to trust my ears.
As of this writing I have not yet A/B the two, but I will.
I will come back with my experience of this one Metal album, by the way, a metal album probably not the best to experiment with but I'll give it my best shot.
4. Why do I choose 192 when there is a 96 or lower available, usually about a $2.00-$3.00 savings. Because, I suffer from a bit of tinnitus which effects the highs (treble) this has been documented by my hearing tests at doctors office. I believe, probably placebo, that if my ears are not the best than I should by the best, just in case. Placebo, actually has been documented that if you interpret it as being better through Placebo, then who is to care?

Let me be very clear: I am not saying anyone is wrong or right on this subject, I only stepped in to offer my experience, which can be or not be the same as others.
 
I enjoy SACD the most. :devilish:
17grams-shut-up.gif





:D Fight!
 
Interview: Jethro Tull's Ian Anderson on 2014 HD Album | Digital Trends
Digital Trends: Do you feel strongly about having your music made available to listeners as high-resolution, 96-kHz/24-bit files?

Ian Anderson:
Well, I’m very keen on the 24, which is absolutely necessary to get the best out of digital recording. 16-bit recording is alright — or it was, back in the ’80s. But 24-bit is not just 8 bits better — it’s a huge amount better.

“Anybody who pretends they can hear more than 48k audio is just bull$hitting. Even if it’s my dog.”

That probably represents the limit of our human physiology to appreciate any difference between 24-bit digital and the analog world in which we live, in terms of recording. 24 bits is all we need. There’d be no point in going to 32 bits. That would be beyond our ability to perceive any actual benefits.
But 24 bits is crucial. 96k, to me, is just double the file size. I’m personally a believer in 48k/24-bit recording, because with 48k, you’ve got ample headroom to more than exceed the limits of human hearing. 48k is going to give more than 20k of bandwidth, and anybody who pretends they can hear more than that is just bull$hitting. Even if it’s my dog.
 
Well, Mr. Ian Anderson, can you explain why ALL your fantastic sounding Steve Wilson 5.1 remixes with the exception of Aqualung [BD~A 5.1] have been released with LOSSY DTS 5.1 DVD~Vs and 96/24 LOSSLESS Stereo remasters?

Why not release BOTH in LOSSLESS formats?

[In REPLY to above Post #35]
 
Well, Mr. Ian Anderson, can you explain why ALL your fantastic sounding Steve Wilson 5.1 remixes with the exception of Aqualung [BD~A 5.1] have been released with LOSSY DTS 5.1 DVD~Vs and 96/24 LOSSLESS Stereo remasters?

Why not release BOTH in LOSSLESS formats?

[In REPLY to above Post #35]
Warner is responsible. I have requested Blu-ray. Problems with the Aqualunq Blu-ray not playing doesn't help the cause. I'm told it doesn't really matter with rock music.
 
Warner is responsible. I have requested Blu-ray. Problems with the Aqualunq Blu-ray not playing doesn't help the cause. I'm told it doesn't really matter with rock music.
What an 'insipid' response from Warner, John .... It's ROCK music ... doesn't matter! The Nerve. At one time Warner/Rhino was tops in releasing their music in hi res and their Chicago//Doobie QUADIO BD~As in 192/24 4.0 are SUPERB! Why they also released those two Fleetwood Mac box sets [Fleetwood Mac/TUSK] in LOSSY codecs after newly remixing them for 5.1 was ABSURD!
 
What an 'insipid' response from Warner, John .... It's ROCK music ... doesn't matter! The Nerve. At one time Warner/Rhino was tops in releasing their music in hi res and their Chicago//Doobie QUADIO BD~As in 192/24 4.0 are SUPERB! Why they also released those two Fleetwood Mac box sets [Fleetwood Mac/TUSK] in LOSSY codecs after newly remixing them for 5.1 was ABSURD!
It was Aqualung Adapted in particular. I wanted Blu-ray.
 
Well I for one am all for hi res throughout the process.
But you can't equate "hi res" e.g. 24 bit/192 kHz with quality sound, because if there's a way to screw something up, someone will often find it.
Take one company's hi res downloads....please! I'll never buy any more of them. It's like the music was intentionally screwed up.
IMO.

Not saying all hi res downloads are bad. But I've heard some doozies (highly technical word, meaning doozie)
 
Back
Top