Faith-Based Electronics (IMO Addendum)

[Jag]

David, that was not a literal explanation it was an analogy (apparently not a very good one).

I would say, "no one records an album that they think doesn't sound good". Still, there are plenty of stinkers out there!

The late 80's/early '90's was when I was first working as a recording engineer, although I'd already been working professionally as a guitarist since the early '80's. In many of the situations I worked in, the "senior people" were still working they way they always had. While many of the older guys would talk about the improvements the technology offered, very few actually seemed to appreciate the practical application. Many of the fine engineers and producers I knew were still recording, mixing and mastering as if nothing had changed (I've since spoken to some who have said as much). Granted, I'm not in a major market, but we are not hicks up here either.

I have played on, engineered and produced many studio projects. I have listened to recordings played back in the studio and know how that sounds. I'm sure there are others here who do as well.

I take your point on the record companies cutting corners. I even said, "Many classic vinyl albums were initially digitized without any regard for the change in technology and they sound like crap". However I don't think cutting corners accounts for the, at the time, "new product" sounding crappy. I'm pretty sure early vinyl didn't sound great before people had a time to figure out the best techniques either.

[David Schwab]

David, that was not a literal explanation it was an analogy (apparently not a very good one).

I would say, "no one records an album that they think doesn't sound good". Still, there are plenty of stinkers out there!

Absolutely, but my point was that they don't record it thinking it will sound better on vinyl. I expect my recordings to sound just the way I hear them, because that's how I want them to sound. Then you do the usual of auditioning them on various monitors to make sure they are even sounding and have the vibe you want.

The late 80's/early '90's was when I was first working as a recording engineer, although I'd already been working professionally as a guitarist since the early '80's. In many of the situations I worked in, the "senior people" were still working they way they always had. While many of the older guys would talk about the improvements the technology offered, very few actually seemed to appreciate the practical application. Many of the fine engineers and producers I knew were still recording, mixing and mastering as if nothing had changed (I've since spoken to some who have said as much). Granted, I'm not in a major market, but we are not hicks up here either.

Yeah, I've been there. I remember telling an engineer/studio owner that I wanted to take my distorted guitar signal, and run it though a pitch shifter with the left -12 cents and the right +12 cents. He looked at me and said "that will sound like crap". I knew it wouldn't, because I used that trick on many of my home recordings. He was surprised on how it sounded. People that have been doing things long enough get set in their ways.

I take your point on the record companies cutting corners. I even said, "Many classic vinyl albums were initially digitized without any regard for the change in technology and they sound like crap". However I don't think cutting corners accounts for the, at the time, "new product" sounding crappy. I'm pretty sure early vinyl didn't sound great before people had a time to figure out the best techniques either.

I think we have unfortunate trends in music production, like auto pitch correction. Listen to some 80s recordings to hear all the over use of gated reverbs on the drums. It was cool sounding at the time, but hasn't aged well! Even some old rock albums sound like crap that were on vinyl, like the early Aerosmith stuff. Rocks is a lousy sounding record, but the music is great. Some of the old Chick Corea and Return to Forever albums are awful sounding. And these were all analog and mastered to vinyl. So lousy recordings of good (and not so good) music have been around forever. Digital just highlights crappy technique and tracking.

Very old vinyl records are pretty bad sounding, but look at the technology used. You might as well be listening to them with a tin can and a string. Thomas Edison's lab is very close to me in West Orange. I was there recently and they play back some old recordings. You can just about make out the instruments! But the vocals are pretty clear.

[Mike Sulzer]

Oboes sound more like oboes...

They do not. Perhaps you do not know the very fine set of early digital classical recordings made in the 70s by Telarc. They were released on vinyl at the time, and then released on cd in the 80s when possible. The cds are far superior. Instruments sound more natural, the imaging is better, there is more air around the sounds, and the dynamic range is oh so much better.

The lesson here is that you want to use as a test case a sound that is intended to be reproduced naturally. That is how you best show up the limitations and strengths of a system.

[FunkyKikuchiyo]

Mike - by no means am I going to deny the successes of digital recording and reproduction. I'm just saying that when vinyl hits a home run, it really can't be ignored. The examples are there. I also think it is a tad arrogant to assume that current digital technology has hit a point of absolute perfection that can never be improved upon. I think it was on this board I read that early writings about Edison's wax cylinders claimed that the recordings couldn't be distinguished from a real human voice. This kinda reminds me of that.

"It is not an assumption. The signal must be processed for vinyl. That's a fact."

"The bottom line is you are used to the softening and dynamic range compression you get with vinyl. "

The assumption you made is that you told me what I like and don't like, and why that is. You can say that, but it is still an assumption. Guys, I'm not arguing vinyl's vast superiority, I'm saying that it has had some great successes and while certain sacrifices are often made, there is some seriously good stuff there and I think critical listening is in order to figure out what we're missing.

"The guitar amp analogy does not work because guitar amps are sound modifiers. They are not hi-fi. Plug your guitar DI into the board. That's what your guitar sounds like. People use amps because they distort in s pleasing way. But you don't use a Marshall stack to listen to music, right? That would sound like crap. You want your playback device to reproduce that Marshal stack as it was recorded, not add more distortion to it. "

Early guitar amps used the same technology as pretty much any other amplifying device, it just so happened that while other types of amplification evolved, guitar amps stayed mostly the same because they liked them that way. Yes a guitar sounds crappy through a mixer... what's your point? MY point is that because mixers are more transparent than guitar amps and designed for a different purpose, it doesn't mean that every mixer sounds the same, because each one is designed and built differently, and those designs impact the sound. Mixers don't enter a "transparent" pantheon where tonal changes don't happen. I think we can all agree that different mixers sound differently. The difference is that guitar amp designers exploit these design elements in a different manner than people who design mixers, or hi-fis or MP3 players. It isn't that hard of a concept, people... you're all smarter than this, I think you're just enjoying arguing too much.

[RedHouse]

...I think a true remaster, of anything originally recorded for vinyl, should include a REMIX as well. This would not sound like the vinyl original but, I don't think it would suck either....

I totally disagree with that, totally. Re-eq them but leave the tracks alone.

The Zeppelin "remasters" were re-mixed when re-mastered and don't sound any better IMHO, I'd even say worse. Another that comes to mind is the (Deep Purple) Machine Head re-mix/master jeez-o if it wasnt for Pacey some of those songs don't even sound like DP doing them.

[salvarsan]

Re: early digital conversions - compressed analog filtered through a steep lumpy filter to remove the Nyquist frequency, into a 16 bit flash ADC, given a long period pseudo-random dither to raise the apparent noise floor and sneakily boost low level signals, and from there to a streaming tape drive -- there were audible process artifacts in Soundstream's early digital audio. I remember how awful it made flute stops sound.

These days, the analog signal runs into 24bit sigma-delta converters running ~4x as fast, so fast that tape noise can be treated as a periodic complex signal and be partially suppressed in post-processing. The music is extremely well represented. All that's left to chance is the mixdown engineer's taste, and that may be what folks are complaining about.

Oh, and calling a science a faith-based endeavor... don't do that.
It might get you unwanted attention.

[Mike Sulzer]

The early Telarc recordings were not made with the original 37.5 KHz sampling system, but rather with a modified system using 50 KHz sampling, allowing the filters to be relaxed. The purpose of the filters is not to remove the Nyquist frequency, but rather to remove content above the Nyquist frequency which otherwise folds into the wanted pass band causing a form of unwanted distortion. Dithering the A/D converter removes imperfections, systematic errors in the least significant bits, at the expensive of increasing the noise somewhat. I do not know what you mean by "sneakily boost low level signals". I do not remember how awful it made flute stops sound; I do remember how high frequencies in low level percussion could be heard distinctly for the first time, and how the delayed sounds from the double basses appeared to roll around the hall, something very difficult to achieve.

Of course it is better to sample fast and use digital filters to remove the high frequency content in order to sample at 44.1 KHz for the CD, but when has an early attempt ever been perfect?

I am confused: who was calling science a faith-based endeavor?

Re: early digital conversions - compressed analog filtered through a steep lumpy filter to remove the Nyquist frequency, into a 16 bit flash ADC, given a long period pseudo-random dither to raise the apparent noise floor and sneakily boost low level signals, and from there to a streaming tape drive -- there were audible process artifacts in Soundstream's early digital audio. I remember how awful it made flute stops sound.

These days, the analog signal runs into 24bit sigma-delta converters running ~4x as fast, so fast that tape noise can be treated as a periodic complex signal and be partially suppressed in post-processing. The music is extremely well represented. All that's left to chance is the mixdown engineer's taste, and that may be what folks are complaining about.

Oh, and calling a science a faith-based endeavor... don't do that.
It might get you unwanted attention.

[David Schwab]

I am confused: who was calling science a faith-based endeavor?

The title of the thread, which I think was intended to be tongue in cheek?

[David Schwab]

I totally disagree with that, totally. Re-eq them but leave the tracks alone.

The Zeppelin "remasters" were re-mixed when re-mastered and don't sound any better IMHO, I'd even say worse. Another that comes to mind is the (Deep Purple) Machine Head re-mix/master jeez-o if it wasnt for Pacey some of those songs don't even sound like DP doing them.

Same thing with some of the King Crimson re-mix/remasters.

[Mike Sulzer]

The title of the thread, which I think was intended to be tongue in cheek?

Oh, thanks, that is the way I took it, too, and then forgot about it!

[David Schwab]

The assumption you made is that you told me what I like and don't like, and why that is.

No, you weren't saying you like the sound of vinyl, you were saying it reproduces the sound of the original recording better, and that you can hear what you interpret as what the guitar really sounded like, etc. I said you are hearing a processed version that was intended for vinyl. It sounds different from the CD version because you are likely hearing two different master tapes (for newer recordings). The vinyl does not offer a more accurate reproduction of the original. It has its own sound, which it imparts on the original. You might like that sound, but it's like running all your music though a signal processor. And you might like that too. So just get a dual 30 band EQ and compressor. Then you can make your CD sound like vinyl.

So I wasn't talking about what you like, I was talking about which medium offers better fidelity.

[rjb]

Humor: a comic, absurd, or incongruous quality causing amusement

The title of the thread, which I think was intended to be tongue in cheek?

Yes, the title was intended to be tongue in cheek!

In my original post, I postulated that the CD version of a Howling Wolf LP sucked because the digital medium couldn't capture the chi of the original performance.
You countered with a boatload of boring facts, including the industry standard sampling rate of 44.1 kHz. :tongue in cheek:
I then proclaimed that "chi pulsates at frequencies above the visible light spectrum".

If my claim is true, then, obviously, chi cannot be captured at such a measly sampling frequency!
To prevent aliasing, the chi would have to be filtered out before even reaching the A/D converter!
See, isn't that hilarious? Get it? Get it?

Man, I hate having to explain my jokes.

Have fun. Seriously.
-rb

[Steve Conner]

But science is faith-based. See Hume's discussion of the problem of induction, for instance.

The main difference between science and religion is that we don't expect the water to turn to wine unless we saw it happen before.

[rjb]

But science is faith-based....

Oh, noooooooo!
What have I done, what have I done?

-rb

[Mike Sulzer]

But science is faith-based. See Hume's discussion of the problem of induction, for instance.

The main difference between science and religion is that we don't expect the water to turn to wine unless we saw it happen before.

The degree of faith-basedness in science is very small, much smaller than implied by the usual examples in discussions of induction. For example, suppose you wanted to argue that Maxwell's equations need a small correction term. You might attempt to verify this by making some very complete and difficult observations. But more likely, you would be able to rule out such a possibility by the use of logic and already existing observations. Perhaps it would require a change in special relativity which is known to be impossible.

Sure, you can always argue that the laws of physics could be different tomorrow. But I think it is unlikely that you would still be here to talk about it.