Digital Audio 101

[rcarlberg]

Over on the "Your Old Vinyl Collection" thread I posted a link to an article, entitled "Vinyl vs. CD Myths refuse to die." It's worth reading. In it the author links to ANOTHER article, "Is the sound on vinyl records better than on CDs or DVDs?", which is ALSO worth reading. The author of the first article says, "The content on that [second] page is, frankly, tripe. On a quick scan I counted six outright errors - mostly uninformed opinion being expressed as fact." Then he goes on to say, "Perhaps at some point I'll create a post here specifically refuting the technical errors on that page". A quick search of Rich Pell's articles on EE Times since January 2010 however does not show that he did.

No matter. The refutation is really easy. The article nicely summarizes the surprisingly prevalent misconceptions about digital audio, so I think it might be helpful to an informed discussion to point out exactly what the unnamed author of that undated and unreferenced article got exactly wrong.

It can be expressed in three letters: DAC.

To digitize a signal, you run it through an ADC, an analog-to-digital converter. This slices up the incoming analog waves into 44,000 slices per second based on 65,536 possible values. This part the article got exactly right.

The results are shown on the graph (pink line) overlaying the 10,000 Hz source sine wave (black line) and the graph correctly shows 4.4 samples.

Pretty rough approximation, you have to agree! It looks like a big ugly square wave -- not the nice smooth 10,000 Hz sine wave it is supposed to be reproducing.

If you've ever accidentally hooked up a speaker to a raw digital output (as I have), that's EXACTLY what you hear: ugly screeching square waves!

What the graph fails to point out, and what the author failed to include, and what a lot of people misunderstand, is that we don't listen to this square wave output. The raw digital bitstream must be first run through a DAC, digital-to-analog converter, to turn it back into sine waves so we can listen to it. Once this is done, something very similar to the original waveform comes out.

How similar? Well you might say "There's a lot of missing information in that square wave approximation" (if fact the article's author does say this) but you have to remember two things: 1) the DAC's job is to smooth out the square wave jumps in level (all 65,536 of them) and it uses the same process, in reverse, that the ADC used to create that square wave. Therefore with a matched ADC/DAC pair the end result is very, very close. 2) The example shown in the graph is a 10,000 Hz sine wave. We don't listen to 10,000 Hz sine waves. In fact the frequency range of even the highest percussion instruments, like a triangle, is only 3-5000 Hz. There are overtones and resonances in the 10,000 Hz range, but no fundamentals.

And remember, below 10,000 Hz you're getting a lot more than 4.4 samples per wavecycle. 8.8 for 5,000 Hz, 17.6 for 2,500 Hz, 35.2 for 1,250 HZ and so forth.

So do you lose "overtones and resonances" with a 44 Khz sampling frequency? Not with a good DAC, and not according to Nyquist theorum. If you aren't familiar with Nyquist it's worth Googling.

Going back to the graph above, would sampling at 192,000 times per second (red line) produce a more accurate result? Well yes -- if you were listening to the square wave output. But if you're using a DAC, the answer is, "Maybe yes, maybe no. It all depends on the DAC." Certainly (and proven in numerous double-blind tests) the human ear CANNOT HEAR THE DIFFERENCE under normal circumstances. That's the cold hard truth.

And that is why products that tout "super fidelity" are mostly blowing smoke up your ass (...to get closer to your wallet without being seen).

And why people who say "analog LPs produce a more accurate representation of the original analog signal" are full of shit.

[Magic Mountain]

Thanks for posting. Nice to get some science for a change rather than anecdotal evidence.

[Jubal]

Thanks R.
The funny part about the "analog kid" was he criticized digital for doing the very thing that vinyl does. It records using a curve very different from the original waves, because vinyl requires RIAA equalization. Both formats record something substantially different and the format calls for re-conversion (RIAA EQ for vinyl, Digital-to-analog conversion for digital).
The bottom line is what sounds best to your ears. The big problem occurs when people who do not really know what live music sounds like attribute accuracy to a format they like simply because it sounds "better" to them. Many analog recordings use fifth order distortion to give music a "warmer" feel.
To my ears, there are circumstances where a album can actually sound "better" (to me) on vinyl. If I am honest though, it is not because the vinyl is more accurate. It is because the vinyl (RIAA conversion) may smooth out overtones I hear in the digital recording that may be construed as harsh, but are nonetheless accurate representations of the original recording.
That said, I purchased a new turntable two years ago and have about 50 vinyl albums. Those albums I feel sound "better" in the stereo format.

[rcarlberg]

A - B Comparison

I was recently on a Bill Frisell kick, and bought a CD of his 1986 ECM album with bassist Marc Johnson (and drummer Peter Erskine) called "Bass Desires." Gives me a perfect example to compare, since I have a near-mint LP of the same album. The CD doesn't say it was remixed, so I think this is a pretty fair comparison.

Starting both of them at the same time allows me to switch back and forth between them.

1) You've never mistake one for the other. The CD doesn't have any surface noise.

2) The mid-bass (100-250 Hz) is fuller on the LP. Gives it a heavier bottom. But the low bass -- under 100 Hz -- is deficient on the LP compared to the CD, as evidenced by the kick drum. Both sound fine and balanced in isolation, but A - B comparisons show a clear difference.

3) Similarly the CD sizzles in the high end. The LP is... "warmer." It's rolled off above about 8,000 Hz. Again the effect isn't objectionable, just very clear in direct comparison. Actually the pops and clicks are remarkably sharp and clear, so I don't think it's the cartridge's fault.

4) Soundstage & imaging -- because Erskine's snare is rolled off on the LP, you miss all of the studio "air" around his drumkit. It's subtle, but once I noticed it it was very apparent. The LP has a lot less soundstage, a lot less sense of the size of the studio. I credit this entirely to the reduced high end on the LP. The mids and lows are very comparable, the transients and FR sound equally good. But the LP is playing behind a wool carpet compared to the CD.

[Wounded Land]

It's funny, but I did a very similar comparison the other night, but with Stevie Ray Vaughan's Texas Flood. This was an album that I bought first on cassette when I was probably 14 years old. Later on, I bought it on CD, and I just recently got my hands on a vintage pressing of the LP.

On the LP, the dynamics of the guitar and vocals were clearly superior. When Stevie dug in, his guitar just jumped out front, although there was a trace of harshness to the vocal at parts. The bass was huge, so much so that I had to kick in the NAD's tone controls and really dial the bass back significantly. After a good cleaning, there was no audible surface noise or any significant clicks or pops.

On the CD, there was more clarity overall, and more of a sense of space (real or electronic). It was generally a "smoother" overall presentation. It was more polite. On "Dirty Pool," when Stevie takes that solo in which he's essentially tremolo picking chords black metal style, the individual strings of the guitar were more present on the LP. On the CD I was more aware of the chords themselves, but on the LP I was more aware of the fact that a guitar was playing, if that makes sense.

I can definitely understand why people, hearing both, would prefer the CD. But for this recording, on this system, in this room, the LP sounded more like an actual band playing actual instruments.

Conclusion: Hard and fast rules are hard to come by in audio.

NP: Beethoven Symphony No. 9 [on CD]

[trurl]

http://productionadvice.co.uk/high-s...c-sound-worse/

Are high sampling rates making music sound worse? Good article.

[rcarlberg]

I read that. Seems like a bit of a stretch.

[rcarlberg]

It's funny, but I did a very similar comparison the other night, but with Stevie Ray Vaughan's Texas Flood.

Was it remixed?

Comparisons are meaningless if so.

[Birdy]

What it really comes down to in the long run is that everybody likes what they like, "science" aside and arguing or trying to convince everybody ad nauseum one way or the other gets really stupid. Both have their merits/disadvantages.

Enjoy the music in whatever format you enjoy.

[rcarlberg]

What it really comes down to in the long run is that everybody likes what they like, "science" aside and arguing or trying to convince everybody ad nauseum one way or the other gets really stupid. Both have their merits/disadvantages.

Enjoy the music in whatever format you enjoy.

I agree with this. People can prefer whatever they like.

However the science is not ambiguous nor unsettled. If your preferences ignore science you cannot redefine the science to fit your preferences.

[trurl]

I agree with this. People can prefer whatever they like.

However the science is not ambiguous nor unsettled. If your preferences ignore science you cannot redefine the science to fit your preferences.

Exactly... people are free to prefer the quaint sound of music on wax cylinders if they prefer and that's fine but there is an objective reality as to the physics of the medium and why it sounds like it does.

[Jimmy Giant]

Wow, this is probably the most even tempered thread I've seen on this subject. I'm torn because I've never had a high end turntable, which I feel is required to make the stereo comparisons (along with a good CD player).
I always hated the clicks and pops and it diminished my listening pleasure. Higher sample rate DVDs do sound amazingly better than CDs. Remember, the more accurate information (higher sample rate) on the input leads to more accuracy on the output. Yes, having great matched AD/DA converters are necessary, but the easiest way to say this is "garbage in, garbage out." You want as little "approximation" as possible.
The things I've heard in (properly done) DVD-A and SACD just flat out astound me. Vinyl can never match that mostly because separating the instruments out over more channels leads to more accurate sound and less frequency interference. And space! When a nearly 50 year old recording (Moody Blues - Days of Future Passed) can come through with virtually no noise and clear, dynamic orchestral sound in DTS (even 96k, not 192k), you would picture frame the vinyl version and just hang it on the wall.
But yes, the bottom line is what sounds best to YOU. We don't all have the same freq response to what's left of our hearing!

[Magic Mountain]

Higher sample rate DVDs do sound amazingly better than CDs. Remember, the more accurate information (higher sample rate) on the input leads to more accuracy on the output. Yes, having great matched AD/DA converters are necessary, but the easiest way to say this is "garbage in, garbage out." You want as little "approximation" as possible.
The things I've heard in (properly done) DVD-A and SACD just flat out astound me.

From all that I've read, this is patently not true, all else being equal. By all else being equal, means the same source material, mixing, mastering, etc. It has been shown in double blind tests that subjects, both trained and untrained listeners could not tell the difference between a DVD-A/Bluray than a downsampled 16/44.1 cd of the same material. Many times 24/192 sounds better because there is better source material for these.

[Supersonic Scientist]

Agree that this is a very mature and grown-up discussion on the topic and I have to say, I've learned a lot too. Keep the scientific info coming. I'm digging it!

[wideopenears]

Good thread.

[rcarlberg]

From all that I've read, this is patently not true, all else being equal.

Usually though (as you noted) all else is NOT equal -- the 24/96, 32/192, SACD, DVD-A stuff is remastered, improved, recorded specifically for the format or otherwise worthy of the format. Otherwise, why bother?

I'm in favor of the highest quality format that one can afford. Of course it has to be top-to-bottom. It doesn't do any good to upsample something from a low fidelity master.

And that includes "super-bit masters" of classic Rudy Van Gelder recordings from the 1960s. As good as they are, I'm not convinced there's enough there to tax a standard 16/44 encoding.

[rcarlberg]

people are free to prefer the quaint sound of music on wax cylinders if they prefer

Q: Is there a music genre that you avoid?
A: Rock-n- roll. Period. Any of it. Hate it. Worse thing that happened to music. Hurt all types of music. They took blues and ruined it. It’s the cancer of music….ate into everything. Killed Country music, that’s for sure.

Everyone has an opinion.

[simon moon]

A - B Comparison

I was recently on a Bill Frisell kick, and bought a CD of his 1986 ECM album with bassist Marc Johnson (and drummer Peter Erskine) called "Bass Desires." Gives me a perfect example to compare, since I have a near-mint LP of the same album. The CD doesn't say it was remixed, so I think this is a pretty fair comparison.

Starting both of them at the same time allows me to switch back and forth between them.

1) You've never mistake one for the other. The CD doesn't have any surface noise.

2) The mid-bass (100-250 Hz) is fuller on the LP. Gives it a heavier bottom. But the low bass -- under 100 Hz -- is deficient on the LP compared to the CD, as evidenced by the kick drum. Both sound fine and balanced in isolation, but A - B comparisons show a clear difference.

3) Similarly the CD sizzles in the high end. The LP is... "warmer." It's rolled off above about 8,000 Hz. Again the effect isn't objectionable, just very clear in direct comparison. Actually the pops and clicks are remarkably sharp and clear, so I don't think it's the cartridge's fault.

4) Soundstage & imaging -- because Erskine's snare is rolled off on the LP, you miss all of the studio "air" around his drumkit. It's subtle, but once I noticed it it was very apparent. The LP has a lot less soundstage, a lot less sense of the size of the studio. I credit this entirely to the reduced high end on the LP. The mids and lows are very comparable, the transients and FR sound equally good. But the LP is playing behind a wool carpet compared to the CD.

What do you attribute the roll-off at 8K to?

It is not inherent to vinyl. I have test records by several different manufacturers (from the 70's and recent manufacture) that have frequency sweeps from 20 - 20K with no roll-off (at either end).

Cutting lathes are capable of encoding out to 100K. Most 70's commercial vinyl was capable 35K.

There could be a couple of things going on.

The loss of low frequencies can be caused by a poor match between the effective mass of the tonearm and the compliance of the cartridge.

The loss of high frequencies could be caused high inductance coils in the cartridge. Most low to mid price moving magnet cartridges are guilty of this.

It also could be due to a less than good phono stage.

[Jimmy Giant]

It has been shown in double blind tests that subjects, both trained and untrained listeners could not tell the difference between a DVD-A/Bluray than a downsampled 16/44.1 cd of the same material. Many times 24/192 sounds better because there is better source material for these.

Well, in the past, I've been able to tell the difference on occasion, it does depend on the listener. No, can't always tell, but yes, there's a difference when something is remix/remastered and many times the newer output sounds better because of it. That's typically where the digital version, when done properly, can sound better than the LP. Again, not always. We've gone through a period of "engineers" crushing the signal with max compression to give it a perceived improvement in sound. Louder is better, right? I guarantee these people would never get a job remastering classical music, or would be fired day 1. Many have since realized the errors of their ways, creating another wave of remasters. And then there are botched remasters like the first Japanese Yes remasters. Lost a lot of money there. Also, I don't know what happened with the Genesis box SACDs, but the DTS sounded much better to me, and I would usually pick SACD over DTS.
Bottom line is, there are a lot of variables in play here, not just theory. Nobody's ears are the same, sound systems are vastly different, room space is critical. All that does make a difference when you're splitting hairs.

[rcarlberg]

Simon:
Yup, all true. I do not know exactly where my roll-offs are occurring (it's an old table and I rarely listen to vinyl anymore). You're absolutely spot-on, they're not necessarily on the LP.

Although LF rolloff was common in order to get more playing time per side. And HF rolloff was common to limit record wear. But as you say, there was no physical limitation that made it so.

Without the bass cut, you’d only have about 5 minutes on your LP side.

Although 25kHz response is possible, excessive transients are a problem. There are several reasons for this. [...] RIAA curve [...] tiny coils (one for each channel) in the cutting head [....] The coils are helium cooled but still can reach 200 degrees Centigrade. A circuit breaker is used to prevent catastrophic destruction. This doesn’t all add up to the limitation it seems, because it is still possible to cut levels higher than can be played back.

Source

[Wounded Land]

Was it remixed?

Comparisons are meaningless if so.

Nope, it was an original issue CD and an original issue LP.

[rcarlberg]

Nope, it was an original issue CD and an original issue LP.

But did they come from the same master?

[Wounded Land]

I have no reason to suspect otherwise. This was released in 1983, so I wouldn't expect separate masters to be made.

What about your Bill Frisell albums?

[rcarlberg]

What about your Bill Frisell albums?

As far as I can tell, yes. They were both 1986 -- well into the CD era. Nineteen 83 would be just as the cusp of the beginning of CDs.

[AZProgger]

In the past couple of nights, I did an experiment similar to Wounded Land's. First I down converted the finale movement of Prokofiev's 7th Symphony from 24/96 to 16/44.1. I also down converted my own digitization of Les Cathedrales by Fireballet from 24/96 to 16/44.1. Lastly I compared the Emerson, Lake & Powell CD to Vinyl. The only alteration I made in the down conversion process was the addition of dithering. This is a must for such conversion, and if anything, made it a more fair comparison.

In all three cases, the 16/44.1 and CD sounded a little more crowded, like the instruments were bumping into each other. The hi-res and vinyl sounded a little more like the instruments had room to breath. All instruments stood out more as their own entities. Listening over Sennheiser headphones, the Symphony seemed to lose 2 to 4 percent of stereo separation with the down conversion.

Most dramatic was the E, L & Powell comparison. As I'm sure you're aware, this album is heavy on the reverb, like most albums from that time frame. On the CD, the reverb sounded thick while on the vinyl, it sounded more light and airy. On the CD, Greg's bass disappears in the mix so it's barely detectable. On the vinyl, it's more prominent. There was also more stereo separation on the vinyl, but that may have been a function of the turntable, not necessarily the recording itself.