Dolby's TrueHD 96K Upsampling To Improve Sound On Blu-Rays

Stowie101 writes in with a story about your Blu-ray audio getting better. "The audio on most Blu-ray discs is sampled at 48kHz. Even the original movie tracks are usually only recorded at 48kHz, so once a movie migrates to disc, there isn't much that can be done. Dolby's new system upsamples that audio signal to 96kHz at the master stage prior to the Dolby TrueHD encoding, so you get lossless audio with fewer digital artifacts. The 'fewer digital artifacts' part comes from a feature of Dolby's upsampling process called de-apodizing, which corrects a prevalent digital artifact known as pre-ringing. Pre-ringing is often introduced in the capture and creation process and adds a digital harshness to the audio. The apodizing filter masks the effect of pre-ringing by placing it behind the source tone — the listener can't hear the pre-ringing because it's behind the more prevalent original signal."

Comments

You cant hear it anyway.

[Hatta]

44.1khz audio is already transparent to the human ear. Blind studies have been done where a 16 bit 44.1khz ADC-DAC pair was inserted into a high resolution analog audio source. No significant difference was observed.

Don't waste money on the placebo effect.

Re:You cant hear it anyway.

[Anonymous Coward]

I guess that experiment failed to use monster cables then

Re:You cant hear it anyway.

[Anonymous Coward]

The hearing limit is actually about 20kHz. You need more than 40kHz sampling if you want to capture a sine wave at 20kHz.

The purpose of capturing at higher than 48kHz is to prevent sounds at frequencies above 20kHz being captured at a too-low sampling frequency, and appearing as audible frequencies. These can be removed by analog filtering, but only about one octave above the cutoff frequency. Analog filters are not ideal brick-wall filters, so 96kHz sampling is useful.

However, once the audio is acquired and digitized, software can provide a true brick-wall digital filter. This is impossible to do in analog hardware. After applying the brick wall filter, it can be sampled down to 48kHz or 44kHz with no loss. So, there is absolutely no reason to put 96kHz on disc.

The article isn't clear whether it's 96kHz on just the master, or the disc also.

Re:You cant hear it anyway.

[aardvarkjoe]

That's just because your hardware sucks. If you use the correct equipment, anyone with a discerning ear will be able to hear the difference.

Re:You cant hear it anyway.

[MightyYar]

This might be my favorite review ever on Amazon:

These cables deliver crisp clear sound and are worth every penny. The sound, in all ranges, is amazing. My panoramic eq has never sounded better. I just have one gripe. My Television sometimes won't turn off ever since I've started using these cables with my stereo surround system. In fact it's on right now despite the fact that it's not even plugged in to the electrical outlet. I'm not sure how but these cables are supplying independent power to my television and stereo receiver. It's really cut down on my electricity bill even though, at times, I've lost the ability to control my TV.

Another downside is that, occasionally, there will be high pitched shrill sounds through the speakers. Almost as if a young woman is screaming. It doesn't happen all the time though. Usually it's around 3am when the TV turns itself on. I'm not sure why. It always turns on this show called "Hell Beast". Tivo is not set to record it but, without fail, it turns on every night at 3:33 am. I'm not sure what it's about. There's some sort of gargoyle or mutant goat or something. I think it's a monster movie show. Although they never show a movie and the goat monster guy just says "I want you" over and over. I think it's British or something. I don't really understand the humor. I'm usually tending to my newborn daughter who's routinely wakes up crying because of the screaming coming out of the television. It's funny too because that goat character on the show sometimes yells the name Shannon and that's the name of my daughter. LOL...

Other than those few issues I'm really enjoying the free electricity. It's helped with $$. Especially after all the money I had to drop re-soding my lawn after some teenagers burnt a star into my front lawn. Some stupid neighborhood gang. They're calling themselves 9-9-9.

Re:You cant hear it anyway.

[SimonTheSoundMan]

I'm a sound engineer and you are totally right.

Going back in history. 44.1kHz was chosen because it syncs with PAL video frames, 48kHz syncs with NTSC. If you were doing linear editing, you can dub and cut the audio perfectly to the half frame.

44.1kHz stuck because Umatic, an analogue videotape that you could buy a PCM head as an optional extra, was chosen to create the master copies for CDs to be sent to duplication in to pressed CDs.

Re:You cant hear it anyway.

[Anonymous Coward]

Question: Was 44.1 kHz chosen in part because the integer 44100 is highly composite? It's divisible by the following factors up to its square root: 1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 14, 15, 18, 20, 21, 25, 28, 30, 35, 36, 42, 45, 49, 50, 60, 63, 70, 75, 84, 90, 98, 100, 105, 126, 140, 147, 150, 175, 180, 196, 210.

Especially interesting is that it's divisible by 7.

Prime factorization of 44100 is 2^2 x 3^2 x 5^2 x 7^2, or (2x3x5x7)^2, or just 210^2. Pretty cool, huh? Coincidence or by design?

Re:You cant hear it anyway.

[the eric conspiracy]

Here is a link to the original paper by Dr. Peter Craven where he mathematically proves that an apodizing filter can make audible improvements in sound reproduction.

http://www.aes.org/e-lib/browse.cfm?elib=12992

Unfortunately it's behind a paywall. but take my word for it's a pretty impressive piece of work.

Most people, whose sensibilities are not trained to the point where they are discriminating enough won't likely notice the difference. However the fact of the matter is the differences are measurable, the principle is based on sound math, and the results are in the perceptible audio spectrum.

Peter Craven made several important contributions to digital recording. He and Michael Gerzon did a lot to push forward the early development of surround sound technology, and made other significant contributions to the process of digital recording. In particular their work on dithering has had a big impact in improving the quality of CD recordings.

http://en.wikipedia.org/wiki/Michael_Gerzon

http://www.aes.org/e-lib/browse.cfm?elib=5872

http://www.aes.org/e-lib/browse.cfm?elib=6777

http://www.aes.org/e-lib/browse.cfm?elib=6647

Re:You cant hear it anyway.

[TheRealMindChild]

My mother always told me that I can't taste the Tuna in her chicken cassorole. I don't care WHO couldn't taste it, I could.

Re:You cant hear it anyway.

[dgatwood]

Actually, polystyrene caps can make a huge difference over electrolytic or tantalum caps in certain parts of some circuits. For example, in condenser microphones, the coupling capacitor between a microphone element and the first FET stage is a critical part of the circuit in which the signal level is very, very weak. Thus even tiny amounts of noise from cheap capacitors can have a significant effect on the final result. A fair number of cheap Chinese microphones sound dramatically better if you replace the cheap dipped tantalum caps they use with a film cap or poly cap.

We're not talking about a small difference here, either. We're talking night and day. A deaf person could just about hear the difference. :-) Replacing just the handful of tantalum capacitors in those microphones can make the difference between a muddy sound with a harsh, brittle top end and a fairly clean, accurate representation of what is being recorded... all for about five bucks and a few minutes of soldering. (Even better, the most important one—the FET coupling cap—is usually direct-wired between the capsule mount and the FET's lead, so you don't have to worry about lifting traces....)

Capacitors within the feedback path of an amplifier circuit can also degrade the sound noticeably. Admittedly, this isn't as much of an issue these days with the rise of modern, chip-based amplifier circuits, but it is still worth keeping in mind, particularly given that most condenser microphones still use transistor-based amplifier circuits.

Just to be clear, though, it doesn't have to be polystyrene film. The difference between a polystyrene cap and a traditional metal (polyester) film cap is negligible compared with the difference between film caps and electrolytic or (*shiver*) tantalum caps. Tantalum caps simply should not be within a city block of any trace that carries an audio signal.... Okay, slight exaggeration, but you get my point.

And, of course, it doesn't make sense to replace every capacitor. If it isn't in the signal path, it usually won't make much difference (though the absence of capacitors in the right places on power supply rails can cause some fun problems), and even if it is, it may or may not make much of a difference, depending on where the capacitor is in the signal path.

Re:You cant hear it anyway.

[old and new again]

he refered to karajan performance that is 73 minutes long, the story is real, but its was not BECAUSE of this that they choose 44.1

Re:You cant hear it anyway.

[guidryp]

Here is a link to the original paper by Dr. Peter Craven where he mathematically proves that an apodizing filter can make audible improvements in sound reproduction.

You can't mathematically prove something sounds better. Most adults can't even hear 16KHz, let alone 20 KHz and beyond, or detect subtle variations in those ranges.

You have to do double blind testing. Double blind testing has shown even real 24/96KHz can't be discerned from 16/44.1KHz by audiophiles and recording pros.

Anything they are trying to sell beyond this is placebo snake oil.
http://mixonline.com/recording/mixing/audio_emperors_new_sampling/

Re:You cant hear it anyway.

[sr180]

How does a sound engineer get to call themselves an engineer? Im not having a go, Im just asking...

However, for those of you quoting Nyquist, you only have half the answer. One of the side benefits of a higher frequency is lower quantisation noise - and hence a better signal to noise ratio. When you take a sample of sound, you then fit it to 16 bits. Obviously an analogue sound pressure level wont fit perfectly into a 16 bit value - so you have to fit it to the nearest one. The difference then becomes noise - which can generally be approximated as white noise (I know mathematically this is possibly incorrect, but practically its true) with its energy spread over the available frequency. Filter this noise out (which your ears will do for anything above 20-25khz) and you reduce the effective quantisation noise being heard (you have filtered out half of the noise's power) - improving the signal to noise ratio.
This obviously will not work in the case of material already sampled - as the quantisation noise is already there in its sampled form, however, it will have a similar effect for the encoding - if the encoding poduces white noise as part of its process - which (not having researched their encoding thoroughly) is likely.

Will it truely make a difference? I doubt it. TrueHD is already damn good - and the limitations are really going to be in the amplifiers and the speakers, particularly the cheap power supplies modern home amps seems to carry. I'm sure this is really just more about planned obsolescence.

Re:You cant hear it anyway.

[drkstr1]

Question: Was 44.1 kHz chosen in part because the integer 44100 is highly composite? It's divisible by the following factors up to its square root: 1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 14, 15, 18, 20, 21, 25, 28, 30, 35, 36, 42, 45, 49, 50, 60, 63, 70, 75, 84, 90, 98, 100, 105, 126, 140, 147, 150, 175, 180, 196, 210.

Especially interesting is that it's divisible by 7.

Prime factorization of 44100 is 2^2 x 3^2 x 5^2 x 7^2, or (2x3x5x7)^2, or just 210^2. Pretty cool, huh? Coincidence or by design?

./ need to let you keep some mod points in a reserve so you can use them when you come across some fine gems like these! :D

Re:You cant hear it anyway.

[Prune]

The article summary is misleading. This isn't about removing sampling artifacts, but about removing reconstruction artifacts (by the DAC's digital filter)--specifically, preringing. The 96 kHz thing here is a red herring. And as usual, Dolby is way late to the game. A few major semiconductor manufacturers added apodizing filters to their DAC chips years ago after studies showed preringing was audible.

Re:You cant hear it anyway.

[nothings]

I posted about this on twitter a month ago.

The frequency chosen had to be a multiple of 900 and had to be somewhere in a limited range of frequencies (above 40Khz, below some number I forget). The 900 comes from a factor of 300 (to guarantee it was divisible by 50 and 60 for PAL/NTSC), and a factor of 3 (the preferred number of samples per scanline; 2 was too few, 4 would have been wasteful).

There is no evidence that the specific multiple of 900 from the required range (40Khz to 47Khz) was chosen because of what the factors of the multiplier would be, but rather because the frequency wanted to be as high as possible (giving a wider region between limits of human hearing and the nyquist freq, thus making filtering it cheaper), but higher frequencies would have required encoding samples in the vertical blank part of the signal.

Certainly the fact that 900 itself is already the product of the squares of the three smallest primes is coincidence, since the factors of 300 and 3 were essentially independently motivated--the 3 wasn't chosen because it "completed the set" with the 300. Likewise, I don't believe the additional factor of 49 was chosen because of that factor. (Having more divisibility is useful in some circumstances, but 7 is such an uncommon divisor; 900*48 would be far more useful on the general divisibility front, introducing more factors of 2 and 3. But, in fact, nobody NEEDS this number to be more divisible, as it's not needed to be divisible beyond the factor of 900 that was required.)

There's a wikipedia page about it (do an "I feel lucky" search on google for "44.1").

Re:You cant hear it anyway.

[Anonymous Coward]

> NTSC is 59.97Hz, not 60Hz

*Color* NTSC is 59.97, but black and white NTSC is an even 60Hz.

Re:You cant hear it anyway.

[fluffy99]

I wish I had mod points.

I wish I had a big stick to whack all these so-called experts that are spouting total BS. What makes it worse is that some of these guys claim to be in the sound recording business and really don't understand the electronics or mathematics underlying the equipment they use.

The system I work on cost somewhere around 4 million, has 1500 channels, recording at various rates from 32k up to 192k. We calibrate it end-to-end and certify our measured data to +/- 0.2 db. So some twat on here claiming his system is good to 0.01 dB with no artifacts is really smoking something.

This Dolby system is really just trying to estimate the sampling artifacts and back them out of the digital data. The whole higher sampling rate thing is a smokescreen. As someone else pointed out, Dolby makes their money by licensing and they really want to convince consumers that they need this in their equipment.

Re:This is...

[Anonymous Coward]

No kidding. A/B/X or GTFO.

Worthless gimmick with no audible benefits

[sahonen]

Dumb, dumb, dumb. An ideal sample rate upconversion results in something that *is* identical to the source. Mathematically. It's like re-encoding a 64kbps MP3 to 192kbps. If anything you are going to *lose* quality due to inherent errors in the process.

Re:Worthless gimmick with no audible benefits

[PhrostyMcByte]

Mod parent up!

A lot of people will see a graph of PCM and think up-sampling will help make the stair-stepping be finer, less noticeable, and thus improve quality. Unscrupulous audio companies love to take advantage of this belief with up-sampling tech.

That belief is, of course, complete bullshit—the stair-stepping of PCM is merely a digital encoding which DACs use this to reproduce a full, fluid signal. There's literally nothing for up-sampling to do that could add any quality! The only thing it will do is introduce even more errors.

In some cases DACs have even behaved worse at higher sample rates—meaning in that case you'd not only have more errors from upsampling, but also more errors from the DAC.

Re:Worthless gimmick with no audible benefits

[Anonymous Coward]

Mathematically you've got 5 choices when you want to sample and have sound almost all the way up to nyquest:

1 a brick wall, phase linear filter. Mathematically the best, and yes it has pre and post ringing - the less roll off you allow the more ringing

2 you can do less filtering and allow aliasing instead. In that case you'll get a mirror image of the spectrum above the nyquest rate that wont matter much because only dogs and small children can hear that high. And less ringing

3. You can let the treble roll off a bit. In fact 48k sampling rate is more than cd just so that the roll off from 20 to 24 is longer than that from 20 to 22 and you'll get less ringing. A little roll off never killed anyone

4 you can use an old style filter with some phase shift. It just trades off preringing for postringing and delays some frequencies more than
others and is overall less efficient. Frankly the frequencies being discussed are so high no one will notice the delays. In theory you can mess up the imaging and sound a little that way. There's a reason that the industry has preferred linear phase digital filters to older style analog filters, but no doubt in the digital domain you can optimize a filter with phase delays just like you can optimize one without.

5. You can have an adaptive filter that decides between options 1 2 3 and 4 depending on some unimportant critera like masking. It's unimportant because only children can hear high enough to detect even a hint of ringing or aliasing above 20khz and as far as I know they're not the market.

Re:Worthless gimmick with no audible benefits

[DigiShaman]

Question. In theory, you could turn those stair stepped signals into vectored splines. Does anyone already do this?

Re:Worthless gimmick with no audible benefits

[Anonymous Coward]

The signal is only "stair stepped" because they chose to graph it that way. The audio signal coming out of the DAC does not look like that. Those stair-steps are happening at frequencies more than half the sampling rate -- they are eliminated from the analog output by a low-pass filter. This is essentially performing this "splining" you are talking about.

Lossless + Cinavia == Lossy

[Anonymous Coward]

May I be the first to say this- fuck Bluray, and fuck Cinavia.

I used to buy Bluray disks. Hell, I own a whole shelf full of them (about 80 titles in total). Every single one eventually got ripped to my NAS in two formats- a relatively lossless MKV file containing the original video and audio streams (up to DTS-HD MA), and a lossy x264 version for playing on crappy devices like the PS3 or 360.

Then Cinavia rolled around, which did two things:
1) It purposefully corrupts the audio stream in an attempt to encode digital information into it (go read their patents- the harder you try to pry Cinavia into an audio stream, the more damage is done to the original quality)
2) It prevented me from playing my legally purchased and legally ripped (it's legal in my country to rip disks and things you BUY) disks off my NAS on my PS3

What pisses me off the most though is that Sony is pushing Cinavia on everyone as hard as they can. AFAIK all new BR players need to be equipped with it, and most of the new BR disks are supposed to have it as well. And they're still advertising the disks as "Lossless", when in fact the audio is NOT lossless- it's lossy, the degradation of which is brought about solely by Cinavia's presence.

Before anyone yells [citation needed] at me, here's your proof straight from the Wikipedia page (http://en.wikipedia.org/wiki/Cinavia):

"Cinavia's in-band signaling introduces intentional spread spectrum phase distortion in the frequency domain of each individual audio channel separately, giving a per-channel digital signal that can yield up to 20 kilobits per second—depending on the quantization level available, and the desired trade-off between the required robustness and acceptable levels of psychoacoustic visibility. It is intended to survive analogue distortions such as the wow and flutter and amplitude modulation from magnetic tape sound recording. On playback no additional audio filters are used to cover up the distortions and discontinuities introduced."

So there you have it. Lossless is no longer lossless, because Sony insists on using this stupid fucking DRM on their stupid fucking format (as usual). Dolby's new gimmicky technology might claim to give you better lossless audio, but none of that matters the moment they drive Cinavia into the stream.

-AC

Re:Lossless + Cinavia == Lossy

[serviscope_minor]

This is a discussion site, not a peer reviewed journal or research paper.

Apodizing Filter

[Josuah]

The title is misleading if the actual goal of this is to apply an apodizing filter. I suspect the reason it's called "Advanced 96K Upsampling" is because that's much easier to get people to buy into that than a "Apodizing Filter" sticker.

The article explains how the audible benefit comes from the application of the Meridian apodizing filter, which changes the analog signal reproduced from digital data by reducing the pre-ringing. IIRC the trade-off is that post-ringing increases. The claimed benefit is that since the ringing now occurs after the "real" music of larger amplitude and as a result the ringing is masked or could be considered like an acoustic echo that naturally occurs.

The 96K upsampling is just a side-effect of wanting the extra samples when you are applying the filter.

Here's a decent summary of what is supposed to happen to the analog audio signal as a result of the filter application: Technical analysis of the Meridian Apodizing filter.

That being said, from what I've read over the past few years I think people are kind of mixed on whether or not the filter makes things better, worse, or just different but not better.

Re:Apodizing Filter

[slew]

A fancy name, but it seems to me that this mostly just a DSP textbook minimum phase filter. A minimum phase filter is just a causal filter which has minimal group delay means it can be made to sound more "analog" (since analog filters are usually mostly causal meaning there is no filter contribution from the "future"). This is of course basically tipping the hat to the sound those vinyl/tube-amp purists claim is the "best" sound (not that vinyl/tube-amp purists would actually like this as it is still an soul-less digital approximation, although perhaps listening through oxygen-free copper speaker wire might help ease them to this "approximation")...

I guess having "minimum" in the name isn't a good marketing technique thus "apodizing"...

Re:If you want to impress me

[Ogi_UnixNut]

I'd prefer if they kept movies with the "100db difference". It is far easier to apply a dynamic compressor plugin than it is to undo studio-mastered dynamic compression. In fact, I hope they do the same with music as well, so that eventually we can apply as much compression as we want for a given environment/situation.

who records 'expensive movies' at 48k?

[TheGratefulNet]

do you have a cite for that? I don't believe it.

even home recording is laughed at (technically) if you are not using 24/96. recording at 48k is just absurd. playback at 48k is fine, though; but I'm not at all convinced that million dollar (at least) movies capture audio at 48k.

if that really is true, then people have been ripped off on their blue ray purchases. one of the supposed benefits is 'better sound' and if you still get 48k (and likely 16bit audio too; as its not common to use 48/24 mode) at record time, nothing the BD can do will ever make it better than dvd. yes, dvd uses compression on dolby 5.1 or dts but its compression is actually nearly lossless *compared* to most consumer playback (not a huge S/N dac+preamp+amp) systems.

Re:who records 'expensive movies' at 48k?

[wiredlogic]

Mixing at 24/96 has some merit in the name of reducing cumulative errors. You'll be hard pressed, however, to find an ADC that produces more than 16-bits of useful, noise-free data at 96KHz for recording.

Re:If you want to impress me

[TigerPlish]

make a system that amplifies dialog to the same level as everyfucking thing else in the movie so I dont have to constantly fiddle with my remote. Why is it in music we have the loudness wars where all sound is mashed into mindless noise at the peak of volume, but in movies there HAS to be a 100db difference between scenes

So let me get this straight - you're cutting down music because of the loudness war, but you want THE SAME THING in movies? Shoot, you already have it! Just pick the mix with the most letters and acronyms in the name!

I'll give you one example, and I hope you have this dvd and a shit-hot hi-fi to go with it so you can duplicate it.

2007's Titanic release, the 3-disk set in the blue case. This one has a "5.1 dolby mix" that I wager most people use -- this is what I call the "muggle mix." For people who don't know any better. THe dialog and music are fairly close -- in fact, the dialog is too loud. This mix is compressed, just like pop music. I play this one with the volume at -52db. (95db 1w 1/m speakers.) It sounds "meh". Sure, you hear everything, and everything's fairly close, but it's "meh". Just like compressed pop.

THen there's the 2.0 Dolby Stereo mix. This is the one you want, if you want it to sound like it did in a theater. This one's uncompressed. To get a natural dialog level, I set the volume at -36 or -34, depends on my mood. At this level the sound is completely natural. WHich means when people whisper, they whisper. When people talk, they talk. When they yell, its getting loud. When Rose makes her trek down E-Deck to bust Jack free, the whole house shakes along with the boat -- and is one of the best demo bits I've ever heard for movie sound.

Same with classical music. I play most of it on the same rig as above at -36 or -34. It's soft when the orchestra's soft, and it's fucking LOUD when the conductor sticks the baton up the orchestra's collective ass.

But when I play compressed pop, it's down to -52 for moderately compressed stuff (squirrel nut zippers) and -62 for MECO's Star Wars disco thingy, which is probably the most compressed music I have.

Movies have huge dynamic range. You can either accept this, or play the muggle track.

Or, get into your receiver's or source's setup, pick DRC = ON compress the snot out of it yourself. Every DD / DTS receiver or prepro has it. It may be called different things, but it's Dynamic Range Compression.

And it's the Devils Work. It should be banned from all recordings.

Unsampling ... then re-sampling in 96KHz?

[Taco Cowboy]

Oh, c'mon !!

This is one thing that simple does NOT make any sense

If the thing was recorded in 48KHz, it's at 48KHz, and no matter how one can "un-sampling" that shit and then re-recording it in 96KHz (even at 96MHz or 96GHz), it does not boost _anything_ !!

Re:Unsampling ... then re-sampling in 96KHz?

[hairyfeet]

Exactly, this is like colorization. You can't recreate what wasn't recorded in the first place, all you can do is add shit on top. The funniest part? Ask teens and early 20s and they will tell you they LIKE the "sizzle" of MP3 because that is what they have grown up with. so not only are you adding shit that isn't there but the "artifacts' they are complaining about are ENJOYED by the younger generations which is the big target demographic everyone shoots for!

Re:Unsampling ... then re-sampling in 96KHz?

[Austerity Empowers]

Obviously you can't unsample and re-record an audio stream to reduce pre-ringing or de-apodizing all but the smallest apods. All you are doing is essentially taking harsh audio and putting it on qualudes. This simply produces depressed, harsh audio, like Norm Macdonald.

Time and money would be better spent using gold plated, neon injected, forward biased monster cables, with gallium arsenide softening strips. The justification for using gold plated, neon injected, forward biased monster cables is well known by audiophiles. The gallium arsenide softening strips work by absorbing the harsh pre-ringing frequencies by actually siphoning out the high frequencies. Silicon engineers have long known that gallium arsenide is ideal for conducting the highest frequency signalling. Used in this application it acts as sort of a apod rejection filter, allowing the ringing to be thrown free of the cable, before it can manifest as ringing, or even chirping.

However it is important to stress that the gallium arsenide softening strips must be calibrated to your eardrums carefully before use. You will need a vector analyzer and a sound meter. It's crucial that you place the softening strips in your mouth, while providing the four port network the VNA requires using both your feet and hands. You must suck on the strips until the sound meter absorbs all the s-parameters in your body, which are being slowly drained by the gallium arsenide strips. You must maintain this until the sound meter gets down to at least 3dB (or 1dB if you have especially sensitive ears). Without this step you may as well be using a walmart SPDIF cable, it will be that bad.

Re:Unsampling ... then re-sampling in 96KHz?

[msobkow]

If you treat the incoming 44.1 or 48 KHz stream of incoming signals as points on a curve, and apply Curve Fitting calculations to interpolate the intervening data points, you can mathematically recreate some of the detail.

However, this isn't necessarily accurate data -- it's just recreated, the same as when you expand a picture. But like a picture, there are different algorithms and techniques for doing the upsampling, and they "colour" the sound much as an upscaled photo may have jaggies or appear a little blurry.

What I find more interesting is the idea of combining curve approximation with a point mass. You treat the current sample as a point in time, and use acceleration curves to make the "mass" travel a path that intersects all the sample points. If your calculated mass correlates to the actual mass of the drivers in your speakers and the air they move, it should result in a more accurate recreation of the original sound curve.

In fact, I believe Mobile Fidelity got in some hot water with the USG for using just such an approach to encoding 44.1 audio disks, and had to sign a non-disclosure promising they wouldn't use the algorithms for anything other than audio processing. Apparently the USG developed similar algorithms for cruise missile guidance (missiles have mass), so even though it's an obvious and purely physical phenomena being modelled, it's a "military secret." :D

Re:Unsampling ... then re-sampling in 96KHz?

[johnwbyrd]

This is incorrect. Interpolating an audio signal using using lerp or Bezier or whatever will introduce auditory artifacts in the upper frequencies of the sound. The only mathematically correct way to upsample a signal is to perform the transformation into frequency space and then resynthesize the signal at the desired frequency with a lowpass filter.

See https://ccrma.stanford.edu/~jos/resample/ for more information on why curve fitting is incorrect.

Re:Unsampling ... then re-sampling in 96KHz?

[julesh]

Tubes naturally have electric "mass" that has to be "moved" by the changing signal strength, smoothing out the raw digitial samples into a proper analogue curve.

You can say exactly the same thing for a capacitor, and they're much cheaper, less fragile, and get to their operating point much more quickly when the circuit is switched on.

Re:You can prefectly represent anything up to Fs/2

[Man On Pink Corner]

There's no such thing as a square wave at a given frequency. A square wave is the sum of the fundamental and all odd harmonics, and a triangle wave is represented by another, similar series.

You might have sine, triangle, and square waves whose fundamentals are all at 20 kHz, but both the square and triangle waves will sound exactly the same as the sine wave if they are sampled and reproduced properly at 44.1 kHz. The antialiasing filter will remove the harmonics before the signals are digitized, resulting in three recordings of a sine wave.

Higher sampling rates allow you to use cheaper antialiasing filters, but that's hardly a constraint worth worrying about in a modern digital signal chain.

lol wut

[Alex Belits]

pre-ringing

Really? In an uncompressed audio? And the solution not only involves upsampling as a part of the process but requires the signal to stay upsampled?

My eyes are rolling at 15krpm.

Re:lol wut

[Prune]

Preringing is what the linear-phase oversampling filter in the DAC chip in the player creates. Which is also the place to fix it, by putting an apodizing filter there, and some semiconductor manufacturers do exactly that (Wolfson Micro, etc.). Dolby's approach makes no sense--they oversample 2x during mastering (needed or the apodizing filter doesn't work) and then you have to store twice the data. Why? If the DAC is doing it, then you can just feed it the usual 44.1 or 48 k. Moreover, since the DAC's filter usually oversamples by 8x to allow simpler analog filters post-DAC, it can do the apodizing much better anyway. Once again Dolby takes legit technology and implements it poorly into a lousy gimmick to sell. Instead of reading dumb marketing material and even dumber article summary on slashdot, read some peer reviewed papers discussing preringing and apodizing filters, say http://www.aes.org/e-lib/browse.cfm?elib=12992

Re:You can prefectly represent anything up to Fs/2

[Man On Pink Corner]

Try a high but more audible frequency.

It may be less confusing if I put it this way: If you can't hear a sine wave beyond, say, 20 kHz, then you are not going to be able to tell the difference between a sine wave at 7 kHz and a square wave whose fundamental frequency is 7 kHz. That's because the lowest harmonic in the square-wave signal will be at 21 kHz. Your ears will filter it out, just as the antialiasing filter in the recording system would need to do.

Now, that being said, the argument has been made that intermodulation effects in the human ear can allow us to perceive sounds beyond the usual 20 kHz limit when they mix with each other. To the extent these effects occur when listening to the source material at a given level, you could argue that the ultrasonic parts of a performance should be captured and reproduced along with everything else, and that would require a higher sampling rate.

The showstopper for this argument is that any desirable sonic content resulting from IMD at ultrasonic frequencies could only be reproduced "properly" at a specific volume level, because distortion products by definition are generated by nonlinear processes.

Re:You can prefectly represent anything up to Fs/2

[Man On Pink Corner]

The point is that you cannot distinguish a square wave from a sine wave at the same fundamental frequency, if you can't hear the odd harmonics. You cannot have a square wave at a given frequency without the odd-order harmonics. If you don't have the odd harmonics, you don't have a square wave -- you have a sine wave.

Nitpicking arguments about the frequency of a signal in the time domain are not relevant. Human hearing operates in the Fourier domain -- almost literally, if you understand how the cochlea works -- not the time domain.

And Harry Nyquist is rolling around in his grave

[Gordo_1]

Not that this whole thing isn't absurd for the reasons already discussed above, but what no one bloody well seems to understand it that an audio stream is not a godamn bitmap picture. You can't improve audio quality of *audible frequencies* by increasing resolution of the horizontal axis (sampling frequency) beyond a rate which surpasses the Nyquist frequency for human hearing. Assuming a high quality anti-aliasing filter is used and excellent quality recording and playback equipment, audio sampled at 48kHz can be unambiguously represented up to about 24kHz. 96kHz is a waste of bits.

Vertical resolution (# of bits) is the only theoretical way to improve actual audio quality further... and beyond about 16-18 bits, it's also beyond the ability of even the most diehard audiophiles to discern (in properly conducted experiments.)

Re:And Harry Nyquist is rolling around in his grav

[fluffy99]

You can't improve audio quality of *audible frequencies* by increasing resolution of the horizontal axis (sampling frequency) beyond a rate which surpasses the Nyquist frequency for human hearing.

Nyquist-Shannon notwithstanding, the range of human hearing is wider than 20kHz.

http://www.cco.caltech.edu/~boyk/spectra/spectra.htm (a properly conducted experiment)

That article says nothing about the human hearing range other than making a reference to some other unproven hypothesis. The article does show that instruments produce frequencies will above 20kHz, which was never really in question.

Re:You can prefectly represent anything up to Fs/2

[Prune]

The ear operates both in the time and frequency domains, in a manner analogous to using a very short fourier transform window when calculating a waterfall plot. As for sound above 20 kHz not being audible, studies show 120 kHz is perceptible through bone conduction: http://en.wikipedia.org/wiki/Ultrasonic_hearing and also see http://ieeexplore.ieee.org/iel5/5286202/5291232/05291285.pdf?arnumber=5291285 and other related studies showing ultrasound that is not necessarily consciously perceptible does affect perception of music.

Here's the thing

[Sycraft-fu]

It doesn't matter if there's a mathematical difference, it matters if there's a perceptible one. There's a lot out there that you can prove mathematically is more like the actual original sound wave. None of that shit matters to reproduction for human enjoyment. What matters is if the difference is perceptible to humans. The sound wave could be totally different and if humans can't hear the difference it doesn't matter.

That is the whole thing behind lossy compression. You can do an imperfect deconstruction/reconstruction of a sound wave and humans will have trouble telling the difference, or find it impossible at higher bitrates. Telling the difference as an objective matter isn't hard, you can do it on a scope, FFT, with a diff, whatever. Telling the difference listening to it is impossible (with sufficiently high bitrate, like 256k MP3).

Also don't think that just because the AES is the society for audio engineers they are immune to audio voodoo. The world of pro audio is one I play in as a hobby, and there's voodoo that goes around in it to. A friend of mine is a professional audio engineer, and a good one, but he gets in to the voodoo. He has special cables, paints the edge of his CDs green with a marker, and things like that. He's convinced he can hear a difference.

Re:And Harry Nyquist is rolling around in his grav

[dmbasso]

That said, doubling the sampling rate isn't going to do anything for a digital signal. At best, the new signal will simply play each of the old signal's samples twice.

Actually upsampling can be useful when you apply digital filters. There is no such thing as an ideal filter, so if you modify one frequency band (e.g. in a equalizer) you end up modifying all others. The higher is the sample rate the lower is this sideband interference.

Re:An exercise for the reader

[Prune]

I have one more thought on this: think of how much low level distortion is masked by how crappy speakers are in general. I think many blind tests will have to be revisited when we have truly low distortion speakers. Even electrostatic headphones are not that great. Then there's the French ionic headphones http://membres.multimania.fr/plasmapropulsion/Industrial_issues/Plasmasonic.htm but the bastards didn't measure distortion so one can only guess.
The best performing I've seen is glow discharge plasma. On Google patents you can check US 4,219,705. He used helium to create stable glow discharge (in air, glow discharge is very unstable and inevitably transitions to an arc) and then shaped it in such a way as to get a flat frequency response. Photo: http://3.bp.blogspot.com/_tb6Dp4NFH5w/SxWU_QWONKI/AAAAAAAAAnk/IgXp5VMkZlk/s1600/Plasmacell1.jpg Unfortunately, around 330 Watts in the discharge alone and only above 500 Hz (regular cone speaker for below 500) and you have to refill helium tank at welding shop periodically... But look at waterfall and impulse response: http://tinyurl.com/7d6pdnv and http://tinyurl.com/7xfppsz and THD
I decided to build this without helium. I realized I could do it once I came across microhollow cathode discharges: sandwitch a CRCLC->regulator). 135 uF total at around 3000 V is about 600 Joules or about twice a defibrillator.
One of the huge film-in-oil caps leaked (so much for "designed for pulse discharge") and the oil caught on fire and I was so startled by the mini-explosion that I broke the complex electrode structure. Haven't returned to this project yet, but I think there's something to this approach. There was also virtually no ozone (unlike the usual corona discharge speakers people drive with RF), but UV is definitely an issue to overcome.

Re:And Harry Nyquist is rolling around in his grav

[Lluc]

You can't improve audio quality of *audible frequencies* by increasing resolution of the horizontal axis (sampling frequency) beyond a rate which surpasses the Nyquist frequency for human hearing.

Nyquist-Shannon notwithstanding, the range of human hearing is wider than 20kHz.

http://www.cco.caltech.edu/~boyk/spectra/spectra.htm (a properly conducted experiment)

That said, doubling the sampling rate isn't going to do anything for a digital signal. At best, the new signal will simply play each of the old signal's samples twice.

Just because musical instruments produce frequencies above 20kHz (as shown in your link), it doesn't mean that the average human can hear them. Younger people can hear frequencies up to ~20kHz, and maybe a bit above, but most middle age adults probably cut off around 15kHz or lower. Here's one study showing 18-24 yr olds who can mostly hear 24kHz, but they're generating the sound at 117 dB -- a very dangerous level for more than just a few seconds. (http://informahealthcare.com/doi/abs/10.3109/00206098409070087?journalCode=ija)

Listening to loud sounds (>85dB) for extended periods of time will decrease the high frequency response of the human ear, so I wonder if high frequency hearing in children and teens of the last decade or two will have even worse hearing that their parents due to the ubiquitous white ear buds.

Re:And Harry Nyquist is rolling around in his grav

[poopdeville]

"X. Significance of the results
Given the existence of musical-instrument energy above 20 kilohertz, it is natural to ask whether the energy matters to human perception or music recording. The common view is that energy above 20 kHz does not matter, but AES preprint 3207 by Oohashi et al. claims that reproduced sound above 26 kHz "induces activation of alpha-EEG (electroencephalogram) rhythms that persist in the absence of high frequency stimulation, and can affect perception of sound quality." [4]
            Oohashi and his colleagues recorded gamelan to a bandwidth of 60 kHz, and played back the recording to listeners through a speaker system with an extra tweeter for the range above 26 kHz. This tweeter was driven by its own amplifier, and the 26 kHz electronic crossover before the amplifier used steep filters. The experimenters found that the listeners' EEGs and their subjective ratings of the sound quality were affected by whether this "ultra-tweeter" was on or off, even though the listeners explicitly denied that the reproduced sound was affected by the ultra-tweeter, and also denied, when presented with the ultrasonics alone, that any sound at all was being played.
            From the fact that changes in subjects' EEGs "persist in the absence of high frequency stimulation," Oohashi and his colleagues infer that in audio comparisons, a substantial silent period is required between successive samples to avoid the second evaluation's being corrupted by "hangover" of reaction to the first.
            The preprint gives photos of EEG results for only three of sixteen subjects. I hope that more will be published.

In a paper published in Science, Lenhardt et al. report that "bone-conducted ultrasonic hearing has been found capable of supporting frequency discrimination and speech detection in normal, older hearing-impaired, and profoundly deaf human subjects." [5] They speculate that the saccule may be involved, this being "an otolithic organ that responds to acceleration and gravity and may be responsible for transduction of sound after destruction of the cochlea," and they further point out that the saccule has neural cross-connections with the cochlea. [6]

Even if we assume that air-conducted ultrasound does not affect direct perception of live sound, it might still affect us indirectly through interfering with the recording process. Every recording engineer knows that speech sibilants (Figure 10), jangling key rings (Figure 15), and muted trumpets (Figures 1 to 3) can expose problems in recording equipment. If the problems come from energy below 20 kHz, then the recording engineer simply needs better equipment. But if the problems prove to come from the energy beyond 20 kHz, then what's needed is either filtering, which is difficult to carry out without sonically harmful side effects; or wider bandwidth in the entire recording chain, including the storage medium; or a combination of the two.
            On the other hand, if the assumption of the previous paragraph be wrong â" if it is determined that sound components beyond 20 kHz do matter to human musical perception and pleasure â" then for highest fidelity, the option of filtering would have to be rejected, and recording chains and storage media of wider bandwidth would be needed."

Remove 98KHz passfilter induced harmonic overtones

[TiggertheMad]

When dealing with digital inputs, too much accuracy in the audio stream produces harshness and digital fatigue.

Citations, Please. This sounds very much like analog fan boy bullshit. "Too much accuracy", really? I have seen quite a bit idiotic pseudoscience used to explain why analog is better than digital. If you just like vinyl over CDs that is fine, just say so, but don't try to snow job me with some jargon filled statement in an effort to back up your personal tastes.