Why analogue and not digital?

[apl]

Fine but why persist in burdening everyone else with it.

Why not? If it's a serious discussion then, yeah, it'll probably be a burden to get to the bottom of it. If we just wanna throw our subjective preferences around, that's fine, too, but it really doesn't enhance our understanding of what we're talking about.

And who's coerced into being burdened?

No, it was exaggerated, probably in an attempt to make the point unmistakeably clear................obviously it wasn't clear enough for you.

It was another bad thought experiment.

You really could digitize a face in three dimension and calculate the limits of accuracy of the reconstuction based on digital signal theory. In this case, instead of time you would have a displacement such as left to right or up and down. And in terms of digital signal theory, measuring data on a face that's 20 cm wide every cm is the same as sampling every nine seconds in a three minute song, and using only four bits.

 

[wado1942]

explain, and find countermeasures for those differences.

That's exactly what I'm saying. There's something we're missing, that we're unable to explain with science what our comparatively primitive ears tell us. I'd like to discover more about it and while I have theories, there is no instrumentation on earth yet that can test the things I want to test.

No, it was exaggerated, probably in an attempt to make the point unmistakeably clear

Also right on the money. It's not a pathetic thought experiment at all. It's exactly the same. You take measurements, write down those measurements, use those measurements to build a new model. The scale in thought experiments are not important. I could have used MMs or uMs as my base and it still wouldn't overlook the fact that information is missing and that an imperfect creature is trying to fill in those details based on an estimated guess. Would you have felt better if I used a 2 dimensional image? It wouldn't have changed anything. In my "industrial technology" class ages ago, we did that exact experiment using photographs and graph paper. We layed graph paper over a photo and put dots where lines in the photos intersected the lines of the graph. Then we removed the photo and drew in lines to connect the dots. None of the completed images were exact replicas of the photos. At least we had a memory of the original photo to tell us how much we should curve a line and which way to curve it. We also weren't stuck to puting dots just on the corners of the graph paper. A computer has very ridged sampling intervals and no memory of the original so it just guesses based on adjacent samples.

Might I mention that much of Newton's and Einstein's theorems are based on thought experiments that they were unable to test scientifically in many cases. Yet they have created the most accurate theorems to predict the behavior of our natural world than anybody else before or after them, particularly when it comes to the motion of stellar bodies.

 

[Dr ZEE]

I can't quite figure out what his POV is.

My Practical POV is: choose what's best for the task based on experience.

My somewhat "philosophical" POV is not so easy to describe as the practical one. But I'll try my best in a shortest possible way.
I'm an independence freak. And I believe in separation of powers .
I don't like the idea of delegating the faith of my waves solely to the ingenuity, ability and honesty of engineers, manufacturers and entrepreneurs.
Softly speaking, I don't exactly trust neither of them .
"Converting" my waves into a "list of discrete values" does exactly that - it puts the plot of my work in "jeopardy of trust" and leaves nothing behind.
I don't mind, on the other hand, delegating to them (all three of the above) all the powers that they grab and eat when providing me with tools that treat my waves with respect and as an Objective Phenomena, the integrity of which is indispensable.
And I'd gladly do my part and pay my fees. And I would gladly deal with dust and dirt that comes with and my lack of control over the Indispensable Fenomina, becuase that's life.

My POV in one short line is: A true objectivist ultimately is an analog freak.

 

[apl]

That's exactly what I'm saying. There's something we're missing, that we're unable to explain with science what our comparatively primitive ears tell us. I'd like to discover more about it and while I have theories, there is no instrumentation on earth yet that can test the things I want to test.

Our ears are very sophisticated. so much so that creationists ask, "How could this just happen?" Did you know that our ears are discrete in the frequency domain? But also very sophisticated is the signal analysis provided by the brain. It's so good that it can hear things that can't be measured.

You take measurements, write down those measurements, use those measurements to build a new model. The scale in thought experiments are not important. I could have used MMs or uMs as my base...

That would change your analogous sampling rate from 0.11Hz (cm) to 1.1Hz (mm) to 1111Hz (µm). Yes, there would still be missing information, but you can calculate exactly how much error there would be in your reconstruction. FWIW, I think you know that the µm resolution would do a very good likeness.

Would you have felt better if I used a 2 dimensional image? It wouldn't have changed anything. In my "industrial technology" class ages ago, we did that exact experiment using photographs and graph paper. We layed graph paper over a photo and put dots where lines in the photos intersected the lines of the graph. Then we removed the photo and drew in lines to connect the dots. None of the completed images were exact replicas of the photos. At least we had a memory of the original photo to tell us how much we should curve a line and which way to curve it. We also weren't stuck to puting dots just on the corners of the graph paper. A computer has very ridged sampling intervals and no memory of the original so it just guesses based on adjacent samples.

That's still three dimensional, ie, x, y, and a grayscale or color value and you can calculate exactly how much error is possible in the algorithm.

Might I mention that much of Newton's and Einstein's theorems are based on thought experiments that they were unable to test scientifically in many cases. Yet they have created the most accurate theorems to predict the behavior of our natural world than anybody else before or after them, particularly when it comes to the motion of stellar bodies.

Some are good, some are bad. Most of the ones I've seen about digital are bad. Connecting dots is always bad.

Here's my fave thought experiment.

 

[apl]

My Practical POV is: choose what's best for the task based on experience.

Well said.

My somewhat "philosophical" POV is not so easy to describe as the practical one. But I'll try my best in a shortest possible way.
I'm an independence freak. And I believe in separation of powers .
I don't like the idea of delegating the faith of my waves solely to the ingenuity, ability and honesty of engineers, manufacturers and entrepreneurs.
Softly speaking, I don't exactly trust neither of them .
"Converting" my waves into a "list of discrete values" does exactly that - it puts the plot of my work in "jeopardy of trust" and leaves nothing behind.
I don't mind, on the other hand, delegating to them (all three of the above) all the powers that they grab and eat when providing me with tools that treat my waves with respect and as an Objective Phenomena, the integrity of which is indispensable.
And I'd gladly do my part and pay my fees. And I would gladly deal with dust and dirt that comes with and my lack of control over the Indispensable Fenomina, becuase that's life.

My POV in one short line is: A true objectivist ultimately is an analog freak.

I really, really wish I had a TSR-8 with dbx. And I wish it had needles instead of LEDs. I love needles.

 

[evm1024]

Exactly. Some of us would like to measure, explain, and find countermeasures for those differences.



And some of us would like to characterize those, develop the transfer functions, and write plugins.

Your face analogy is pretty pathetic. It would be like making a recording and sampling once every five seconds or so.

Now here's a blind/blind subjective experiment that somebody with, say, a MIDI controlled pipe organ could do. You could have three things to listen to.

1. Live source through the signal chain, ie, organ-mic-mixer-amp-monitors. Might be tough to keep the flanking paths down.

2. A digital reproduction.

3. A nice analog reproduction.

An organ might not be the thing to use.... Or at least if you were to use it you would need to have it played only once thus MIDI is not needed.

The pipes of an organ are sensitive to temp and air pressure as well as air density (think humidity and barometric pressure).

Imagine a large organ with 10,000 pipes all slightly out of tune to each other.

-E

 

[apl]

An organ might not be the thing to use....

A really good guitar player maybe. The idea is to be able to blindly expose the subject to three versions, the live through the signal chain, the digital reproduction, and the analog reproduction. The musician would have to play pert near exactly the same several times.

 

[apl]

Do the R2Rs make clunks and whirrs when you're trying to record in the same room as the deck? Like when you punch in and out and stuff?

 

[Dr ZEE]

I really, really wish I had a TSR-8 with dbx. And I wish it had needles instead of LEDs. I love needles.

I thought you are an engineer. Why wasting time wishing around?
Heat your gun, plug your scope, grab your screwdriver, screw something - knock your socks off.

/later

p.s.. apl, sorry, but your "humor" does not stick well, buy, hey, it's all good, after all , you are an engineer.

 

[apl]

I thought you are an engineer. Why wasting time wishing around?
Heat your gun, plug your scope, grab your screwdriver, screw something - knock your socks off.

/later

p.s.. apl, sorry, but your "humor" does not stick well, buy, hey, it's all good, after all , you are an engineer.

Just because I could doesn't mean it would be cost or time effective.

 

[Dr ZEE]

...doesn't mean it would be cost or time effective.

That cancels your "really, really wish".
So, then, I guess, what can you do? hmmmm, - just stick to and stare at your current blinkers, what ever blinks there, if anything

 

[Tim Gillett]

Frequency response is NOT the only thing that changes with sample rate changes. Otherwise, we'd still be using 44.1KHz for everything. But 96KHz is now the standard because it sounds cleaner. Most of the distortion is pushed beyond the human hearing range. This is not just an issue of converter quality but the inherant quality of the sample rate itself. Many classical recordings are done either DSD or 192KHz BECAUSE they can hear the difference.

Seems like I have to repeat myself yet again. I wasnt discussing the pro's and cons of 44.1 vs 96. In a very early post, the guy who started this thread posted your snapshot of the 10khz waveform, or its 44.1 representation, and on that basis made the assertion, or assumption, that with digital recording, audio information - whole chunks of audio information - is lost AT ALL AUDIO FREQUENCIES, right down to 20hz.

That of course is nonsense. You dont believe that, I dont believe that, but he does. Would you mind telling him he's mistaken, that yes, people record at 96khz but not to improve the resolution at 20hz or even 1khz but much further up the audio spectrum. Since he apparently respects your views, maybe he'll believe it coming from you.

I was just trying to help the guy out. I made that very clear in the original post.

Once again, please read the discussion in context. Is that too much to ask?

Cheers Tim

 

[A Reel Person]

Damn!...

I luv u guyz!!!.............

 

[apl]

That cancels your "really, really wish".
So, then, I guess, what can you do? hmmmm, - just stick to and stare at your current blinkers, what ever blinks there, if anything

I gots needles on my receiver. Sometimes I move the knob to make them wiggle.

 

[apl]

Seems like I have to repeat myself yet again. I wasnt discussing the pro's and cons of 44.1 vs 96. In a very early post, the guy who started this thread posted your snapshot of the 10khz waveform, or its 44.1 representation, and on that basis made the assertion, or assumption, that with digital recording, audio information - whole chunks of audio information - is lost AT ALL AUDIO FREQUENCIES, right down to 20hz.

That of course is nonsense. You dont believe that, I dont believe that, but he does. Would you mind telling him he's mistaken, that yes, people record at 96khz but not to improve the resolution at 20hz or even 1khz but much further up the audio spectrum. Since he apparently respects your views, maybe he'll believe it coming from you.

I was just trying to help the guy out. I made that very clear in the original post.

Once again, please read the discussion in context. Is that too much to ask?

Cheers Tim

Exactly. Tryin' to keep it honest.

 

[pianodano]

I gots needles on my receiver. Sometimes I move the knob to make them wiggle.

My tape recorder has needles. http://www.danskittlethorp.com/Studio_setup_011_op_800x600.jpg

 

[cjacek]

My tape recorder has needles. http://www.danskittlethorp.com/Studio_setup_011_op_800x600.jpg

Man, that is a beautiful deck indeed...

-----

 

[cjacek]

the guy who started this thread posted your snapshot of the 10khz waveform, or its 44.1 representation, and on that basis made the assertion, or assumption, that with digital recording, audio information - whole chunks of audio information - is lost AT ALL AUDIO FREQUENCIES, right down to 20hz.

That of course is nonsense.

No, what is 'nonsense', rather, is the assertion [or assumption] that with digital recording audio information is NOT lost [yes, at all frequencies]. You don't have to possess a degree in physics or mathematics to know that when you sample a sound per second [whatever the fixed number sample], you're omitting information, whether that is in the form of sound staging, depth, ambiance, coherence, air or what have you..

-----

 

[A Reel Person]

What's...

What's a needle? Some kind of virtual device for a bar meter?

 

[cjacek]

But this was a discussion about which sounds better, right?

Wrong! The thread title (and I suspect it's intent) asks why one and not the other.

Yup, I believe that is correct, ausrock.

The analog vs digital, which sounds better, debate, is an old one so I thought I'd mix it up a little, with a more open theme and I see I've done that alright..

No doubt it's the sound too but also a host of other incentives which drive people to stay or go to analog. I think thats been outlined well, herein.

---