Thoughts on Dylan on PBS and Analog

[PHILANDDON]

Like lots of people, I'm a big fan. I'm amazed at his fearlessness as a performer.

Here's a thought, only analog can handle the type of wild singing Dylan does. (He's got some mic technique when he's in the mood but often he just puts his mouth right up to the mic and whispers and screams his whole way through). With digital he'd be clipping all over the place or they'd have to use a whole busload of compressors and limiters to tame the signal.

Here's another thought, I'm not sure Dylan would've become so famous if he had recorded in a direct to digital age. His performances are so dynamic and analog captures the musicality of that sort of thing so much better than digital.

The special really made me think YET AGAIN about tracking to analog. Digital would just be so cool if it worked as well..........

 

[apl]

With digital he'd be clipping all over the place or they'd have to use a whole busload of compressors and limiters to tame the signal.

Here's another thought, I'm not sure Dylan would've become so famous if he had recorded in a direct to digital age. His performances are so dynamic and analog captures the musicality of that sort of thing so much better than digital.

Pshaw. The best analog equipment has only got about 60dB of dynamic range. 24-bit digital gives you more dynamic range than any analog equipment you can connect to. Yeah, you'd need some dynamics processing but you'd have to do that anyway.

 

[Dr ZEE]

24-bit digital gives you more dynamic range than any analog equipment you can connect to.

that's why you need analog recorder to begin with!,
....especially! if you're recording the guy who has problem controlling his own static (hair that is ) - so forget about trying to control his dynamic :

 

[apl]

24-bit digital gives you more dynamic range than any analog equipment you can connect to.

That sentence was ambiguous, sorry.

Any mic/pre, keyboard, or other instrument that you'd record would have less dynamic range than a 24-bit ADC.

 

[FALKEN]

Pshaw. The best analog equipment has only got about 60dB of dynamic range.

This is simply bullshit. Try looking up some specs on some r2r's. you are mistaken.

24-bit digital gives you more dynamic range than any analog equipment you can connect to. Yeah, you'd need some dynamics processing but you'd have to do that anyway.

in the manner of range; yes. but analog captures more detail between the bits that digital cannot.

 

[Dr ZEE]

Any mic/pre, keyboard, or other instrument that you'd record would have less dynamic range than a 24-bit ADC.

that's why you need analog recorder.
or did I say this already?
Of course, that is if you record to produce music-record and not to do some lab-work for a company, manufacturing 24-bit ADC(s).

 

[apl]

I stepped in it, didn't I?

This argument is worse than conservative v. liberal in the Cave!

Or the evolution v. ID arguments!

So what's the noise floor on a 2" R2R running at 15 ips?

 

[Dr ZEE]

I stepped in it, didn't I?

This argument is worse than conservative v. liberal in the Cave!

Or the evolution v. ID arguments!

So what's the noise floor on a 2" R2R running at 15 ips?

imho: cons vs libs, darvin vs creator and ESPECIALLY! - noise floor on a 2" analog tape recorder - all these things are irrelevant to Thoughts on Dylan on PBS and Analog.

Dylan is DYLAN! This guy would be who he IS regardless of anything...imho, again.

Analog recording is not a 'past'... and it is not a 'conservative way' - it is present and it's simply a great way to record and produce great music.

Digital recording is irrelevant to analog recording. It can be used for what it does and for what it is useful in respect to music production.

Analog tape recording noise floor "issue" is pretty much irrelevant to anything ... practically speaking.

 

[PHILANDDON]

There's this other thing. I don't have any thing scientific to back me up, but it seems to my ears that analog is kinder to someone with less than perfect singing pitch like dylan; an off note in digital just sounds off whereas an off note in analog sounds human or charming. of course, if the note is wildly off this doesn't apply. whatdayathink?

 

[FALKEN]

I think f- singers who can't sing. 'cept for bobby of course. ha!

 

[snow lizard]

Pshaw. The best analog equipment has only got about 60dB of dynamic range. 24-bit digital gives you more dynamic range than any analog equipment you can connect to. Yeah, you'd need some dynamics processing but you'd have to do that anyway.

My understanding of how bit depths work for digital recording is that each bit gives you something like 6 dB of dynamic range. So yes, if you increase the bit depth, you're increasing dynamic range.

On 16 bit systems, you get a range of 96 decibels. In the range of -90 to -96, you're working off of one bit. The amount of variation within each "bit block" will exponentially double as you work your way up the chain.

So in that last 6 dB block, all sound wave data that exists will play back at the same amplitude. That isn't realistic behaviour of the waveform, and analog simply does not behave like this.

So theoretically, a 24 bit system offers 144 dB of dynamic range. Find a preamp or a board or a condenser mic that can give you that much. Damn near most of them don't. Many will get you to -96 dB easily enough, but even in 24 bit recording, the math isn't too dificult to understand.

-90 to -96 in 24 bit will use 9 of those bits. The word length works like this:

2^9 = 512

That gives you 512 choices, or variations in amplitude to choose from, between the range of -90 and -144. It's the total range to that point. 8 bits represents the range that falls below that.

2^8 = 256

Now 512 - 256 is 256. The number becomes more accurate if you say that 0 doesn't count (no signal), so you'd subtract 1.

So in 24 bit, the range in which the amplitude can vary between -90 and -96 is now divided by 255 steps. The condenser mic and preamp can still represent the waveform more accurately than that. Once you start performing more math on the wave file with plugins and such, these areas of the waveform will suffer tremendously, making the advantage of all this extra range come into question.

Considering that the old, old digital recorders of the '80's that used tape typically ran on something like 28 bits, I think we're getting short changed.

To say that 24 bit digital has more dynamic range than analog is sort of true, but it's also necessary to achieve better dynamic accuracy, so it negates the point.


sl

 

[cjacek]

Didn't we do this a short while ago ?

 

[apl]

Hey, sl, I’m not necessarily disagreeing with you, but I’d like to clarify some things.

On 16 bit systems, you get a range of 96 decibels. In the range of -90 to -96, you're working off of one bit. The amount of variation within each "bit block" will exponentially double as you work your way up the chain.

The ADC divides the working range of voltage into 65,536 equal increments in a 16 bit converter. Each sample taken at the time dictated by the sample clock is measured for voltage and assigned a value between zero and 65,536. This value and its time is recorded in the data file. For a 24 bit ADC, the range of voltage is divided up into 16,777,216 increments.

So in that last 6 dB block, all sound wave data that exists will play back at the same amplitude. That isn't realistic behaviour of the waveform, and analog simply does not behave like this.

The last 6dB block is the center around zero volts. If the ADC’s max range was set to ±1.000V, the last 6dB is between 0 Volts and 30.5µV. I would expect that if you were recording with a very nice LDC and pre, the background acoustic noise in the studio and the electrical noise would easily be smearing that area of the signal.

Maybe one of the things that’s confusing about your explanation is the use of dB. The difference between -90 and -96 dB is one bit, but it’s the least significant bit, worth 30.5µV, but the difference between 0 and -6 dB is also one bit, but it’s worth 1 Volt because that’s the most significant bit.

The noise created by not being able to represent less than 30.5µV is called quantization noise.

Once you start performing more math on the wave file with plugins and such, these areas of the waveform will suffer tremendously, making the advantage of all this extra range come into question.

Most of that signal processing is done at 32 bits, so those nuances are preserved.

Considering that the old, old digital recorders of the '80's that used tape typically ran on something like 28 bits, I think we're getting short changed.

Can you provide a link? I am unaware of 28 bit ADCs being available in the 80s.

Here is a useful article on digital signal theory.

IMHO, the preference for analog is due to the aurally pleasant artifacts the analog process introduces, and not because analog is capturing a signal more accurately.

 

[My Name]

WOW!! It's amazing how a relatively interesting thread can decend into one that is sooooooooooooooooo boring.

"How many protest singers are there"

 

[Dr ZEE]

...the preference for analog is due to the aurally pleasant artifacts the analog process introduces...

The whole process of production of aurally pleasant music-record IS ABOUT an array of introductions of aurally pleasant "artifacts" (if you wish to call it that way).

that's why you need analog recorder, if you record to produce aurally pleasant music-record (not to do some lab-work)
or did I say this already?

 

[Dr ZEE]

"How many protest singers are there"

I'd say, Much more than anyone can stomach.

 

[PHILANDDON]

would anyone care to comment on what I said about analog being gentler to pitch (i.e. an off singing note) than digital? has anyone else noticed this? is there any science to back up what i'm hearing?

 

[guitarfreak12]

I just wanna say that Bob, (hey Bob, you mind if I call you Bob) is the most inspirational musician I have ever seen. Like my wife always tells me "it's not what you did, it's what you didn't do" He didn't do so many things. He didn't sing well, he really didn't even play that well, he definately didn't tame his dynamics, and the best one of all he HATED popularity. That special that PBS put on was so funny. The press were asking him "deep" questions and he was just saying "I don't know what do you think? I'm just singing, I wasn't speaking about anything". The best one, was this beetnik guy who seemed to be the standard music lover back then. He asked about his album picture with his motorcycle shirt on, he wanted so bad for Bob to have some deep all powerful, larger than life meaning for every artistic thing he did, and Bob said, "I never really looked at the picture, I don't remember it at all." the guy was crushed. Here's my theory, back then, (and I'll add, I am a youngling, but my Dad was around then) I gather that everyone was so desperately looking for any sign of rebelion and so excited about rebeliion just for the sake of rebelion, that it even came down to a guy who completely rebelled against all of what pop musicians set as a standard for what you're supposed to be. He just did what he liked and that's it. So people loved that because of what it was....rebellion. And I really do dig his songs, they do sound good. I might be way off, but whatever.

 

[mickee]

Anyway...I honestly think that Dylan is "the most popular but worst" singer ever and he most be the first one to be amazed about that...

 

[apl]

WOW!! It's amazing how a relatively interesting thread can decend into one that is sooooooooooooooooo boring.

I'm sorry, I was under the impression that the thread was about the feasability of Dylan's success in the digital domain because the thread starts with the following:

Here's a thought, only analog can handle the type of wild singing Dylan does.

Here's another thought, I'm not sure Dylan would've become so famous if he had recorded in a direct to digital age.

That's what I've been discussing.