Inside High-res audio: PCM vs MQA vs CD: 2L Sampler Comparison

[Herbert]

Interestingly 24bit is a standard in cinema distribution. I assume the Idea was to keep
dialogue level at least around -37dbfs while
maintaining -10db headroom and reserving the remaining levels in between for rare Sound FX, like hunder of lightning etc. Don‘t nail me on the numbers. But mixes are normally louder anyway, 16 bit would be more than sufficient.

 

[Herbert]

Isn't that what happened during the loudness war? Compressing and filling every spectrum making everything very loud while losing almost all dynamic range :'(

Which has to happen with vinyl anyway.
My introduction into digital was Joe Jackson‘s „Body and Soul“ from 1983 (I guess) The first
CD releases were very low in volume with high dynamics.

 

[Limopard]

Indeed you can measure a wine, subjectively of course. I believe the issue with taste is that there isn't an objective reference. Having said that, name two wines and I'll tell you which is better

A blind test would be interesting. I suppose, adding price tags to otherwise neutral wine glasses would change the results substantially. (given the wines are comparable in color, dryness etc...)

 

[danadam]

For pure comparison, speaking about the MQA side. Would it be worthwhile to do a test where the MQA file of the same sample is fed through a MQA capable DAC (e.g. the Topping D90), route the analog output to the ADC of the analyzer, record it at 24bit 96kHz WAV. And then, do the same but with a 24bit 96kHz source FLAC file, fed into the same DAC and route the output to the ADC of the analyzer, record it at the same sampling rate and bit depth. And then do a diff between the two recorded files after we compensated for where the starting points are?

The point I want to make is I would like to see how good/bad MQA fared with a lossless file format when played through the same DAC, with the MQA file properly unfolded and decoded. I know that there will be some random-ness in the DA and AD process... Could be an interesting test IMO.

I believe that's exactly what Archimago did a few years ago: https://archimago.blogspot.com/2017/02/comparison-hardware-decoded-mqa-using.html

 

[Frank Dernie]

Isn't that what happened during the loudness war? Compressing and filling every spectrum making everything very loud while losing almost all dynamic range :'(

No.
That was much more extreme.
With big ensemble classical music the dynamic range is more than domestic audio systems can handle so a little judicious use of peak limiting makes the recording much more universal. After all a lot of these big orchestral peaks only last a few seconds as a rule so only somebody used to live concerts will miss the difference since the rest of the music is unaltered.

The loudness wars is and was massively compressing pretty well all of it all the time and raising the average SPL, which is now possible without clipping because the dynamic range is small, which makes the recording sound loud without moving the volume control, so it often sounds better to those unaware of the effect.

Classical recordings for LP have to be compressed a bit more than they do for CD (the available dynamic range is more like 75db rather than over 90 for CD) but nothing like the record companies have been doing in the "loudness wars" for popular music which is a much bigger manipulation.

 

[LTig]

Which has to happen with vinyl anyway.
My introduction into digital was Joe Jackson‘s „Body and Soul“ from 1983 (I guess) The first
CD releases were very low in volume with high dynamics.

It's from 1984. My first rock CD I bought together with my first CD player in 1985. Despite its age one of the best sounding recordings ever made.

It shows that if the engineers know there job and are not hindered by producers perfect digital recordings were possible since the beginning of the age of digital audio.

Edit: I also own the vinyl and its sound is very close to the CD, the vinyl used is excellent and the groove noise is one of the lowest I've heared so far.

 

[AudioSceptic]

Thanks for the response. It's obvious theres a gap in my understanding of digital sound file encoding - off to the Internet I go. Hopefuly, on my return there will be fewer of these posts!

Make sure you include Monty's videos at <https://xiph.org/video/>

 

[Nathan Raymond]

The point I want to make is I would like to see how good/bad MQA fared with a lossless file format when played through the same DAC, with the MQA file properly unfolded and decoded. I know that there will be some random-ness in the DA and AD process... Could be an interesting test IMO.

The best comparison/analysis of MQA I've seen is this:

 

[AudioSceptic]

I have recorded, mainly classical, music for well over 50 years now.
I have never heard of any music with wide enough dynamic range to tax 16-bit. Back in the old reel-to-reel tape days it required quite a bit of skill to get the levels set so that the quiet bits were not in the noise and the loud bits were not overloaded too much on classical music. The first 16-bit recorder I bought setting the levels was easy so there was no clipping and the background noise is, in any case, so low you can't hear it at all at normal listening level. It was also the first time where I couldn't hear a difference between recorder and microphone feed.

The reality is that you almost certainly haven't ever had a recording with 16-bits of dynamic range, so I believe your premise is invalid.
I do have a 24-bit recorder now. 24-bit makes it totally idiot proof to set levels, but there isn't any music that needs it, it may help at the recording stage if something unexpected happens, that is all. All the music on a recording fits into a 16-bit window, usually very easily, so all the other bits are a pointless waste.

There would be no point in releasing recordings which had so much dynamic range that either the quiet bit was inaudible and the loud bits cause hearing damage! Even if a recording of a full 16-bit dynamic range existed if you set it so you could hear the quiet bits the loud bits would be painful.

Didn't Philips originally think that 14 bits would be enough, and it was Sony who insisted on 16? Please correct me if that's one of the myths I've picked up!

The best reel-to-reel with perfect setup is reckoned to be equivalent to 13 bits, so 14 would be "one better" (Compact Cassette can only manage about 9 bits at best).

 

[Nathan Raymond]

(Compact Cassette can only manage about 9 bits at best).

Depends on the deck, the tapes, and whether you are using Dolby noise reduction. The Nakamichi Dragon does better than 12 bits:

Using the 3 percent distortion reference, S/N 's for EX-II, SX, and ZX measured 50.7, 52, and 55.6 dB, respectively, with no noise reduction and no weighting. With Dolby-B and CCIR-ARM weighting, the S/N figures were 64.3, 66.2, and 68 dB, and Dolby-C increased them all the way to 73.9, 75.5, and 77.5 dB. As with many of our other measurements on the Dragon, these noise figures simply define the current state of the art in cassette-deck performance.

 

[AudioSceptic]

Well, not exactly. The high frequency content is just not there, so it’s not even “null”.



This is a bit more complex, but in essence, yes. An ADC can use noise shaping to move noise up in frequency so the audible band is clean. But so does a DAC if you play the file back again. The trouble starts if the ADC cannot push the noise far enough up and it is still present in the samples audio, like one can see in those 384 kHz examples.



Because then you can’t sell it as high-res audio anymore

Aren't "normal" files (44 or 48k), actually made this way, i.e., downsampled from masters comprising 24- or 32-bit samples at 96, 192, or 384 kHz?

 

[voodooless]

Aren't "normal" files (44 or 48k), actually made this way, i.e., downsampled from masters comprising 24- or 32-bit samples at 96, 192, or 384 kHz?

Of course they are.

 

[Herbert]

Didn't Philips originally think that 14 bits would be enough, and it was Sony who insisted on 16? Please correct me if that's one of the myths I've picked up!

No myth. Sony writes this on their website a tad biased :
One topic that caused considerable debate was the issue of the number of quantization bits, which determines the accuracy of quantization. From the beginning, Philips argued for 14 bits, whereas Doi who represented Sony favored 16 bits. Achieving a higher number of quantization bits became more difficult and expensive. Doi believed that it was worth trying to produce a 16-bit system that would last well into the 21st century. When Philips researchers asked, "Will a 11.5 cm disc with sixty minutes of recording time be okay?" Sony researchers said, "No, we want a 12 cm disc with seventy-five minutes of recording time."

Each argument was valid. Philips argued for a 11.5 cm disc because this was the same length as the diagonal length of an audiocassette. Also, this size satisfied the DIN standard and thus would be the right size for a car audio system in the European market. But it was Ohga, a trained musician, who decisively presented Sony's argument for a 12 cm, seventy-five minute disc. He argued that, "Just as a curtain is never lowered halfway through an opera, a disc should be large enough to hold all of Beethoven's Ninth Symphony." Ohga believed that the disc needed to be of a practical size for music aficionados and that 95% of all classical music pieces would fit onto a seventy-five minute disc. Therefore, a 12 cm disc was necessary to guarantee seventy-five minutes of playing time.

After hearing Ohga's argument, researchers at Philips said, "A 12 cm disc won't fit into a suit jacket pocket." "Well, let's see if it does or not," replied the Sony researchers. They measured the top pockets of a Japanese, an American and European suit jackets. The results showed that "There's no suit jacket with a top pocket size less than 14 cm wide. A 12 cm disc will be fine." It was decided that the maximum playing time would be seventy-five minutes (seventy-four minutes and forty-two seconds to be exact) and the diameter of the disc would be 12 cm. Philips also agreed to Sony's proposals for a 44.1kHz sampling frequency and 16 bits.

https://www.sony.com/en/SonyInfo/CorporateInfo/History/SonyHistory/2-08.html#block2

 

[Materhoe]

Just had a look at Pneuma by Tool in 24bit-96khz

Quite dirty i would say....

 

[RichB]

The fact that it has a wider bandwidth is no surprise, it has been known for decades that a lot of metal percussive instruments emit inaudibly high frequencies.
That doesn't mean many people can hear it.

Above say 22 kHz, it means no people can hear it.

The potential effect of a reconstruction filter implementation on the audible range could be an argument for higher sampling rates.

It is also important, the carefully constructed demos, such as the "keys" where folks have demonstrated the ability to select Hi-Res from CD res, do not identify the source of the difference.
Clearly, the playback system produced a difference in the audible range.
That difference is either, the audible effect is the result of the reconstruction filter or artifacts of the playback system modulating into the audible range.

The only thing that is not possible is to hear ultrasonics as dictated by biology and the very definition of the term.

- Rich

 

[Herbert]

It's from 1984. My first rock CD I bought together with my first CD player in 1985. Despite its age one of the best sounding recordings ever made.

It shows that if the engineers know there job and are not hindered by producers perfect digital recordings were possible since the beginning of the age of digital audio.

Edit: I also own the vinyl and its sound is very close to the CD, the vinyl used is excellent and the groove noise is one of the lowest I've heared so far.

Joe Jacksons The Verdict, from the 1984 CD-issue, recorded on a 3M Multitrack with 16bit/50kHz(!) probably transferred to 44.1 in the analog domain. Unfortunately Sonic Visualiser is not very intuitive and seems to lack simple data like RMS or headroom.

EDIT: Headroom is 0dB, RMS-20.2dB

 

[amirm]

Someone asked why the high frequency spectrum in DSD is not filtered out. Reason is that it can't! If you digitize analog using a DSD encoder, that is what it is going to do. You cannot edit that stream as is. To take out the ultrasonics, you would have to convert it to PCM, then filter it. Problem is, if you are going to deliver it as DSD, then you have to add that noise back in! It simply is the nature of the beast.

 

[tmtomh]

I'm glad these videos exist and people are learning from them, but the finding that hi rez offerings typically have either a ton of ultrasonic hash (DSD) or only low level uncorrelated garbage beyond 24 kHz is not news. Even hi rez cheerleader Stereophile was publishing analyses on this back in the 2000's. I've been looking at purchased hi rez stuff with Audition for over a decade and rarely is there much correlated content (which would be ultrasonic harmonics anyway) in the wasteland beyond the audible range. Likely because so much 'hi rez' product is sourced from ....old analog tapes.

Back in the reign of CD (circa 1992) there was a guy name James Boyk, a music teacher /pianist then at Caltech who was pushing hard the idea that we need to capture (and play back) ultrasonic harmonics that instruments generate (in an infamous articke 'There's Life Above 20 KiloHerz') , and , gee whiz here's PS Audio huckster Paul McGowan citing that article just a few years ago. Boyk himself cited the crackpottery of Tsutomu Oohashi (uncorroborated except by himself -- regardless, his papers were very popular in audiophile circles back then ) of course to back up his claim.

Though his doctoral degree was in Agriculture, Oohashi also published on such topics as 'possession trances' and the uses of artificial life, as well as composing (most famously, the music for Akira)

Talk about a circle of confusion!

Agreed - the finding is not news, and yet it needs to be repeated often, even constantly, because too many folks in the hobby are unaware of it and don't understand, and too many vested interests in the industry want it to be ignored. I'm particularly glad you mentioned Oohashi and Boyk, since these are the "sources" - and the only sources - cited by Bob Stuart, Hans Beekhuyzen, and everyone else who tries to claim there's scientific evidence for the nonsense they peddle. And as you note, these aren't even two sources, as Boyk's only contribution is to point out that musical instruments can produce harmonics above 20kHz (duh) - for the crucial next-step claim that these frequencies are audible to humans, Boyk simply relies on Oohashi's original, never-able-to-be-replicated study.

My understanding is that 24-bit (or 32-bit float) can be quite useful for recording as noted previously, and also useful in real-world conditions of mixing and production, when for example a modern multi-track rock/pop/electronic/hip-hop album might consist of a huge number of individual stems and sources, recorded/sampled in different locations with different ADCs, sample rates and bit depths, arriving on the mixing engineer's desk already having been subjected to various forms of processing, including in some cases more than one round of dither. In such situations I would think 24-bit or 32-bit resolution would make a lot of sense for mixing, processing, making potentially radical level adjustments of some tracks relative to other tracks, and so on.

For playback, however, I am perfectly content with redbook 16/44.1. I got into high-res for a while, but in the last few years I realized it was a complete waste of time, bandwidth, and money. Sure, in an ideal world where CDs had not set the format that way, I'd prefer the default to be something like 20-bit/48kHz. The higher sample rate would be just to provide more of an ultrasonic buffer in case stupid/slow/leaky filters were employed during the production process and/or on whatever equipment I might be listening to (like maybe a portable device or something). And the higher bit depth wouldn't be necessary but would make me feel better, kind of like having a DAC whose SINAD is 10dB beyond what I could ever possibly detect makes me feel a little better than a DAC whose SINAD is right at the border of what I could ever possibly detect, but still way beyond what I could actually detect in real-world situations 99.99% of the time.

Regardless, I have listened to countless high-res and redbook versions of the identical recordings/masterings, for about 12-15 years now, on successively more revealing/hi-fi iterations of my system, and there's no way in the world I can reliably distinguish them from each other. I believe there simply is no human-detectable sonic difference (as long as the downsampling and dithering of the original high-res to redbook is done competently). But even being maximally open-minded and saying it is indeed possible that I or someone else could detect a difference, my philosophy is that if I have to work so hard and listen with such concentration in order to even sometimes possibly hear a slight difference that I can't even say is necessarily better or worse, then who cares? What is the point?

 

[Herbert]

I'm particularly glad you mentioned Oohashi and Boyk, since these are the "sources" - and the only sources - cited by Bob Stuart, Hans Beekhuyzen, and everyone else who tries to claim there's scientific evidence for the nonsense they peddle. And as you note, these aren't even two sources, as Boyk's only contribution is to point out that musical instruments can produce harmonics above 20kHz (duh) - for the crucial next-step claim that these frequencies are audible to humans, Boyk simply relies on Oohashi's original, never-able-to-be-replicated study.



Well, under his alias Yamashiro Shoji, Ohashi gave himself an adequate answer, in the first three seconds .
This record is just great and one of my alltime favourites and extremely well recorded in 1976.

 

[xaviescacs]

They are sadly rare too and unless you are into big stuff like Bruckner and Mahler symphonies or, similar huge ensemble music like Verdi's Requiem you won't be listening to it!

That grabbed my attention. From my experience, two of the classical pieces for which is most difficult to set an appropriate volume for all the piece due to the high dynamic range they have are The Rite of Spring and the Firebird, both from Stravinsky. For instance, in the Action rituelle des ancêtres of the Rite, there is valuable music content consistently 25db per channel below the peak in the same track. This is from the Boulez - Deutsche Grammophon recording:

Does that measure makes sense? Or I'm lost there?

I've been exploring some other music from Mahler and other composers and although there are some points reaching the same difference, they correspond to silences or end of notes. In Stravinsky's pieces however, all music content is essential regardless its volume, and that makes those pieces quite gear demanding, in the sense that without a minimum quality system they are not understandable at all, you simply lose track of the music. The first time you listen to the Firebird, you turn up the volume at the beginning just to capture whats going on. Later on, as the level goes up, you discover that those first phrases should be just braking the silence.