Slashdot Mirror

Well, some of that could really be about volume, but it's more about averages. In a real TV show, you have dynamics... it's mid-range, people are talking, etc. Then it gets really quiet, as Col. Jack O'Neill, Samantha Carter, and T'ealk are sneaking down the hallway, trying not to be seen. Almost too quiet. Then it gets really loud, as the Replicators spring into action and everyone's shooting them to pieces... or was it a bunch of Goa'uld up to their old tricks.

Anyway... dynamics. There's a range, ideally something like 90dB or so, between quiet and loud.

Then an ad comes on. That ad has been processed through a compressor, which puts the quiet parts at 85dB, the midrange at 90dB, and the very loud parts at full range 96dB (this is assuming a 16-bit audio sample, which is what you have in most digital TV systems using AC-3 or MPEG Layer 2 audio). Now of course, they've taken that nice 16-bit potential and dropped it down to about two or three effective bits of resolution, but they don't care, long as it's as loud as your explosions ever get.

You can tell if the originator really dials down the show volume. That's really unlikely -- for one, you'd be getting huge variations in volume from channel to channel, which I've never seen. But in normal audio, the average is probably around -20dB from peak, and your loud ads with their compression probably average out at -10 to -5dB from peak... in other words, they can actually be 2x-3x louder in practice than the television show, even without playing any evil games with your show's normal volume. And if they did that, they'd really be cheating you, because the show's audio would have to be compressed, or you'd simply miss part of it.

If there are ads that really show up a VU meter louder than the loudest explosion peak, that's proof they're really messing with the overall show's volume. But I don't believe that's necessary.

In fact, popular music has already been though a loudness war. Years back, with the advent of cheap audio gear and more bands getting more involved with the technical side of their own work, music started getting louder. This was also new for the digital era -- you simply couldn't make an LP that loud, the mechanics of it were against you. But enter CD, and the issue of "too compressed" goes away, at least if you don't care about dynamics. Again, not really louder in peak, but louder over all -- more compressed, so the effect is louder. Here's the ever-present Wikipedia link: http://en.wikipedia.org/wiki/Loudness_war. And here's a short article about it as well: http://georgegraham.com/compress.html. Same principles as the loudness escalations on TV.

George Graham's album reviews include grades for dynamic compression. He reviews eclectic AAA music.

ask anyone who supported these record stores for years

and they'll say "WHAT? SPEAK UP, WHIPPERSNAPPER. DOWN-LORDIE EMM PEE WHAT? Y'ALL FROM THE FUTURE?"

Buying hard copies at retail is a geezer's activity. Once you can store your entire collection on a fingernail-sized iPod clone, and get new tracks within seconds using weekend-daddy's credit card, why on earth would you want to go out and buy a huge bit of plastic to store a copy of the two tracks you want plus eight that you don't in a medium that you'll never listen to?

Physical distribution of CDs is dead in the water. It's an inefficient, unnecessary and expensive holdover from the ancient past. You might as well give away a free buggy whip with each 'album' (another dying concept) to try and boost sales.

Lest you retort with the stale old "There will always be a market for uncompressed music", fie on that. CDs are effectively compressed. Audiophiles already need to get their fix elsewhere, and their sad devotion to their ancient religion demonstrably isn't enough to keep disks-and-mortar stores open.

CDs are dead as a retail proposition. It's time to put down the buggy whip, and move on.

because cd's are always perfect: http://georgegraham.com/compress.html

The problem isn't just compression. It's that the full dynamic range of CDs is no longer being used for most popular music. The wide dynamic range of a CD is a major reason why it is an advance over the LP--it's what gives CDs their great signal/noise ratio. There's more on the issue here:

http://georgegraham.com/compress.html
http://www.cdmasteringservices.com/dynamicrange.ht m

and pretty graphs here:
http://www.mindspring.com/~mrichter/dynamics/dynam ics.htm

I mean, I spent about $1,000 on a decent sound system, why would I do that just to listen to sound quality matched by a $50 ghetto blaster?

Good point. For most modern music 128K bitrate is just fine as even the CD's are comptessed to the point of distortion and clipping just to sound loud. Those CD's have no headroom decent S/N ratio, or dynamic range. They are not hurt by a low bitrate.

I have a pair of Yamaha NS-1000 speakers and a pair of Acoustic Research AR3's. I know I'm dating myself here, but I have more invested in quality high fidelity speakers than most people have in their entire sound system. I agree that a buck a track for a low bitrate file just doesn't cut it.

What's a crime is the way they have distorted the sound on CD's to increase the average volume at the expense of dynamic range and signal to noise ratio. Distortion is high on these recordings.
http://www.youtube.com/watch?v=3Gmex_4hreQ
http://georgegraham.com/compress.html
http://www.cdmasteringservices.com/dynamicrange.ht m

The part that sucks is the part where your great CD's source has been compressed to make it loud instead of maintaining headroom and dynamic range.

http://www.mindspring.com/~mrichter/dynamics/dynam ics.htm

http://www.stylusmagazine.com/articles/weekly_arti cle/imperfect-sound-forever.htm

http://georgegraham.com/compress.html

http://www.cdmasteringservices.com/dynamicrange.ht m

The first link is the most telling. It shows the clipped and maxxed out audio on some modern CD's compressed and squashed to just be loud constant noise. It's one of the reasons I quit buying CD's. The quality is gone to the point they sound just like a tape that was recorded at +20 Db on the VU meters. The clipping, loss of headroom and dynamic range is just the same.

You're right, but again we have to consider what people are able to hear. The number of quantization levels that are used insures that the human ear can't tell the difference between two intermediate levels.

It's funny, I have an audiophile acquaintance who swears that records are superior in every way to "digital," and for the same reasons described above. The funny thing is, because of the large number of quantization levels used in a CD, the CD's dynamic range far surpasses that of any record player. More info here

Theoretically, yes, analog would always be superior. But in reality, physical limitations of the stylus on a record player limit that medium far more than quantization limits the CD. Those same physical limits exist in the human ear, too.

So, while digital might not be "perfect" theoretically, it's "perfect enough" allowing for the limitations of the human ear.

But they're compressed... what a bummer, the reason I don't use iTunes is because I don't want compressed music... now they're not even offering CD Quality music with their new release?

Even CD's are compressed down to their sample rate. Of course, they're also compressed to maximize volume thus destroying their dynamic range.

I'd rather have a cassette that was mastered well than a modern CD that has been smashed up against the noise ceiling. A 256 kbps VBR MP3 in the hands of someone who cares will sound a lot better than a studio producer in a Porsche mashing 100 years of audio engineering in a misguided attempt to be louder than Howard Stern.

And the "superior sound quality" of both DVD-A and SACD is well outside the range of human hearing.

Okay, before we talk frequency response let's talk sample precision. The biggest limitation of 16-bit samples is an effective 98dB of dynamic range. Today's audio engineers aren't making effective use of the CD's dynamic range, and the reality is that compressors are still being utilized despite the CD being the supposed answer to this problem, which was of course much more pronounced in the days of vinyl. I'd say the fact that compressors are still being utilized is indicative that 98dB is simply not enough for the way all music is currently being engineered, not just for orchestral recordings but for rock and other genres as well. The two solutions to this problem are a better audio engineering process (i.e. better training for audio engineers), or updating the technology to resolve the issue. I'll tell you now... the former is simply not going to happen, and I see the latter as the only practical solution.

A 24-bit system offers 146 dB of dynamic range. While this seems like something which would only appeal to audiophiles who insist on absolute perfection in their orchestral recordings, the truth is that audiophiles tend to prefer vinyl even though any vinyl, even a 78 RPM record (which has a dynamic range of ~75dB) will actually have worse dynamic range than a CD due to compromises made in the vinyl engineering/cutting process. So who then benefits from greater dynamic range? The answer is everyone... 146 dB of dynamic range is more than enough to eliminate an audio engineer's need to use compressors except in the case of the extraordinarly inept. While yes, 98 dB should've been enough as well, today's audio engineers are simply failing to make use of it properly. It's sad that a technological problem is needed to address the ineptitude of today's audio engineers, but the ultimate argument is that it's ridiculous to impose unnecessary constraints on audio engineering if better technology is capable of removing them.

This is all covered quite well in this article. And here is another article which provides support for a simple claim: uncompressed recordings sound better.

In terms of frequency response, yes, human hearing extends only to 20kHz, and the Red Book stipulates that the glass masters of all CDs should be produced by passing the final cut through a 20kHz lowpass filter (the theoretical maximum frequency response of a CD is ~22kHz). What this process ignores, however, is that higher frequencies, while inaudible, are still tangible. While this area hasn't been extensively studied and is much harder to quantify, the tangibility of a live performance versus a recorded one is one of the key distinguishing characteristics, and while most of this tangibility typically comes from the bass side of things and not the treble, simply approaching the audio engineering process from an entirely psychoacoustic perspective will leave you with sound drastically different from the live performance regardless...

The fallacy that seems to have become pervasive among many people in the pop music recording field, especially among record companies, is that if a CD is pushing the absolute digital max it will somehow be louder or better on the air and presumably win more airplay, and thus sell more copies to the public. This is not true at all. Compressing a CD will contribute to on-air loudness almost unnoticeably. Radio people have the brains to turn up a CD that's recorded at a normal level, and broadcast stations' existing compressors will even everything out anyway. The only thing that is accomplished is messing up the dynamic range for those who pay their good money for CDs, "squashing" the life out of any acoustic instruments in the mix, and increasing listener fatigue.

http://georgegraham.com/compress.html

The article provides a nice perspective on the subject.

And here we have yet another Slashdotter doesn't know what they're talking about. They seem to come crawling out of the woodwork every time the "vinyl is better" boobs start slinging their BS around.

> FACT: most people can hear up to at least 30 kHz.

FACT: Nobody can hear up to 30kHz. People *might* be able to hear harmonics of sounds above 20kHz, provided those harmonics fall within the range of human hearing, but they won't be able to hear the actual pure tones themselves (as you yourself indicated). Since any medium - such as CD - that records sounds up to 20kHz will also record the harmonics of tones above 20kHz, provided they fall within the range of human hearing, what exactly would we be missing? And apart from percussion or certain electronic instruments, what instruments are out there generating gobs of ultrasonic information, anyhow? And what microphones are capable of picking up such information? And what analog tape decks are capable of recording such information? And - here's the kicker - how many speakers are capable of reproducing such information? The answer to each of these questions is, vanishing few. Many tape decks filter out or fail to record tones much beyond 20kHz. Few microphones can pick them up to begin with. And most speakers are lucky to maintain a flat frequency response even out to 20kHz, let alone to 25 or 30kHz. You'd practically have to live in a laboratory to record and then accurately reproduce ultrasonic information. A 50-year-old format like the vinyl LP certainly isn't ideal for such a thing, given its noise, distortion, dynamic range, separation and phase issues. Only the high quality analog tape decks found in professional studios or digital recording formats utilizing higher sampling rates than 44.1kHz could hope to accurately record and reproduce such audio.

>Yes, such transients are reproduced on vinyl.

Maybe on audiophile grade, quarter-speed mastered vinyl played back on a $5,000 turntable equipped with a $1,500 cartridge run through a $2,000 preamp they are. Poorly. With oceans of harmonic distortion and waves of crashing high-frequency noise. Assuming, of course, the original performance was picked up using microphones and mic preamps capable of dealing with much of anything beyond 20kHz (such mics cost in excess of $2,000, and the preamps aren't much cheaper) onto tape decks capable of recording much of anything beyond 20-25kHz. None of which is likely, outside of studiously recorded audiophile sessions.

>Your final star'ed points are just dumb. You don't give any references,
>because of course you don't have any.

You must really enjoy looking like a boob. Hey, if you want to play the (in your case, irrelevant and apparently unavailable on the web) references game, I'd love to! (Actually, one "reference" you posted is available on the web - marketing material from a stereo company plugging their overpriced audiophile gear. You should have provided us with a link to the guy selling $10,000 tinfoil hats to protect us from government mind control rays, too.) Here are my bullet points, plus any references I could dig up (though much of this should be obvious to anyone with a brain in their skull):

* Loud tics and pops caused by stray dust and wear, resulting in a *negative* signal to noise ratio - i.e. the noise can become louder than the music! (with N'Stynk, I suppose this would be a blessing in disguise . . . or simply redundant.)

Well, this one is obvious. Whenever a tick or pop is louder than the music (happens a lot with vinyl, and even with tape during quiet passages), the signal to noise ratio goes negative.

* Rumbling caused by the turntable's motor and the friction of the stylus as it passes through the groove

Another obvious point. Many turntables even include rumble measurements in their specifications, though that's for the platter only and doesn't take into account additional noise caused by the friction of the stylus dragging through the groove.

* Wow and flutter, caused by speed irregularities in the turntable's drive system and by any imperfections in the geometry of the disc.

Another spec that's included for most turntables and even analog tape decks. Hard to see how this one is, "just dumb", unless you're so ignorant you've never looked at the specs for a turntable or tape deck.

* Phase irregularities caused by the RIAA equalization and the subsequent need for the preamp to de-equalize the signal.

Another obvious point. Anytime you process the signal to emphasize or de-emphasize certain frequencies, you're going to introduce phase discrepancies. Here's a $2,000 preamp from Daniels Audio that attempts to compensate for the phase issues. Notice I say "attempts". Even a manufacturer of $2,000 stereo components won't claim to be able to eliminate such issues. And who knows what issues all that additional processing is going to introduce.

* Frequency response irregularities caused by the RIAA equalization / de-equalization process

Again, a no-brainer. If the frequency response curve used to produce the wax master doesn't precisely match the frequency response curve in your preamp (and it never will), certain frequencies are going to be emphasized upon playback while others will be de-emphasized. Here's a big page detailing the design issues faced by folks trying to build the RIAA de-equalization circuits for a preamp. Notice the difficulties he's having making the response curve come close to the RIAA ideal. Even by the end, he's off by more than a quarter dB at many frequencies, including some smack dab in the middle of the most sensitive range of human hearing.

* The inability to reproduce loud bass accurately (the cutter making the wax master would pop out of its groove if it tried to reproduce the kind of bass CDs can handle effortlessly)

For references, please see this, this, this, or this.

* The tendency for the turntable, platter and even the disc to function as microphones, picking up room reverberations and - particularly - the sound being produced by the speakers, smearing and distorting the audio in numerous ways

I should think this one would be obvious. Lots of turntable manufacturers sell heavy weights to sit on top of a record while it's playing. If you don't believe this is true, jump up and down next to your turntable while it's playing, or set it on top of a speaker pumping out a lot of bass. You'll get an "extreme" demonstration of the effect, but the truth is it's happening all the time.

* Cartridge / tonearm misalignments, causing inaccurate stylus pickup, accelerated record wear, or both.

Again, an obvious issue. Good luck getting it right!

* 30dB of stereo separation, vs. CD's 70+dB of separation

See this, or the specs for the cartridges themselves here. You'll be lucky to find a preamp that can come close to the 70-90dB of separation even a cheap CD player can provide, let alone a pickup.

* A theoretical maximum of 60dB of dynamic range for virgin vinyl of the highest quality (and only at certain frequencies - obviously, not in the low bass) vs. around 90dB of dynamic range from even the cheapest CD players, across the entire spectrum.

References to this abound. If you don't believe me, take it from an expert.

* In practice, roughly 40dB of usable dynamic range across the majority of the spectrum

See the reference above.

* A relatively flat frequency response from only around 60 Hz to 15 kHz, with severe rolloffs beyond those limits.

This one has been covered already.

* The need for mastering engineers to severely compress and re-equalize the signal in order to steer clear of the format's limitations relative to CD, which requires no such distortion-educing compensation.

Again, see the references above.

* Pitch and frequency errors caused by the speed difference between the cutter used to produce the wax master and your turntable.

That's another obvious fact to anyone but a blithering idiot.

* The tendency of the media itself to wear out as its played, and to be damaged during routine handling with audible results

Well, duh. On to dissect the remainder of your post:

>The reverse of most of what you say is true. E.g. your claim
>of 60dB dynamic range is nuts: the range is over 100 dB.
>You are confusing the noise floor of a high-hiss record with
>dynamic range--but you can hear 20 dB into that noise, and a
>good record need not have high hiss. Vinyl has poor bass???
>It's much better than CD. And so on.

Oh my. There doesn't seem to be anything left to dissect. I've already covered these points up above. Vinyl is *lucky* to hit 60dB of dynamic range with audiophile pressings played back on incredibly expensive equipment. No "confusion" with vinyl's truly outrageous noise floor is necessary. And the dynamic range decreases drastically as the length of the record increases - a problem digital formats don't suffer from. And as for vinyl's bass performance, I think half the links I posted up above note how crappy vinyl is at capturing loud, low bass.

Next time, you might want to learn something about a subject before you proceed to open your mouth and cram your foot down your throat.

It's rather disingenuous of you to say that the audio on vinyl is compressed. It is, but not in the same sense as an MP3 is a compressed digital audio signal, so your statement is misleading. In this case it means that the dynamic range has been flattened a bit, and on pop music vinyl this was generally done to increase the average level of the music, based on the idea that louder music sold better. But they do this with CDs too for no apparent reason, and to the detriment of the quality of the sound -- in fact, the practise erases the increased dynamic range that's the primary advantage of CDs over vinyl as far as I'm concerned.