The Truth about D/A and A/D conversion and Digital Media Primer

Yes, I recall watching that video which makes perfect sense and thinking, how many articles have I seen that offered the stair sine wave as an explanation why digital sounds less analog compared to the other (inferior types of media).

Of course it always was and always will be, the quality of the recording, mixing and mastering that affects the outcome of the final sound. Not so much the type of media.

Higher Resolution version of the files have also been mostly debunked in blind studies in which sound and recording engineers themselves participated with no better than a 50/50 guess, meaning they could not tell a difference.

Having said all that, a native high resolution recording with attention paid to the whole chain, sounds just as analog but with all the benefits of an incredibly low noise floor and perceived detail and resolution that has all of the analog smoothness and naturalness to the sound.

13 bits equivalent of open reel to reel

6 bits of cassette.

Vinyl can have a dynamic range of 70 dB (but often limited to 50 dB)*so 70/6=12 bits

Nothing compares to the potential of digital done right. Until something even better is invented.

Thanks for sharing, it was a good refresher.
 
Kind of says everything there is to know in one brief paragraph...

"Digital can give us even more signal-to-noise and dynamic range, but at a certain point, it’s our analog circuits that just can’t keep up. In theory, 16-bit digital gives us 96 dB of dynamic range. But in practice, the dynamic range of a 16-bit audio file can reach well over 100 dB – Even as high as 120 dB or more.

This is more than enough range to differentiate between a fly on the wall halfway across your home and a jackhammer right in front of your face. It is a higher “resolution” than any other consumer format that came before it, ever. And, unless human physiology changes over some stretch of evolution, it will be enough “resolution” for any media playback, forever."
 
TheOctopus said:
"Digital can give us even more signal-to-noise and dynamic range, but at a certain point, it’s our analog circuits that just can’t keep up. In theory, 16-bit digital gives us 96 dB of dynamic range. But in practice, the dynamic range of a 16-bit audio file can reach well over 100 dB – Even as high as 120 dB or more.
Click to expand...

So please tell me who is recording music that actually contains 96dB of dynamic range. Also explain how “in practice” you can get 100dB-120dB of dynamic range from 16 bits and where are those recordings.
 
Vinyl (D2D) can have a dynamic range of 110-120 dB.
R2R tape (15 or 30 ips 1/2-track) records audible signal below the noise floor and easily reaches 90 dB of dynamic range.

xiph.org and Monty Montgomery have been disseminating their propaganda for years (I didn't look at the dates on the sources above but it's been at least 10 years since I first ran across it, maybe 15)

I am very pro-digital recording and storage, but no one is helped by biased information sharing such as that in the OP, either pro or anti digital, vinyl, tape, CD, hi-res or whatever. Unfortunately as has been frequently noted, the digital equipment used in most professional recording today is far from the state-of-the-art, and the music recorded with more SOTA digital gear is not of the appeal or quality of that which is not.

There is so much we just don't know about audio perception, and relatively little valid research compared to that for video; vision is felt to be far more important than hearing to most people. Very little (if any) perceptual research involving hearing would be of acceptable quality if it were in the biomedical field, and even in that field only about 5-10% of what is published stands up to critical analysis of methods and conclusions. So take any "conclusions" about differences (or lack of) between different digital formats with a healthy amount of skepticism.
 
mep said:
So please tell me who is recording music that actually contains 96dB of dynamic range. Also explain how “in practice” you can get 100dB-120dB of dynamic range from 16 bits and where are those recordings.
Click to expand...

A symphony orchestra playing full out can easily reach 96 to 98 decibels, and certain brass and percussion instruments have registered 130 to 140 at close range. The decibel debate: Sound and the symphony - Chicago Tribune

So there is that...

Then there is the fact that while it may be barely possible to squeeze that kind of dynamic range on digital, 24 bits is theoretically 144dB, from softest to loudest crescendo/passage, I’m not aware of many systems that can reproduce it. The typical noise floor of a quiet room is approximately 30-35dB so one needs to be above that for the softest and add 105 on top. My system can’t do that nor would I ever attempt it.
 
rbbert said:
Vinyl (D2D) can have a dynamic range of 110-120 dB.
R2R tape (15 or 30 ips 1/2-track) records audible signal below the noise floor and easily reaches 90 dB of dynamic range.

xiph.org and Monty Montgomery have been disseminating their propaganda for years (I didn't look at the dates on the sources above but it's been at least 10 years since I first ran across it, maybe 15)

I am very pro-digital recording and storage, but no one is helped by biased information sharing such as that in the OP, either pro or anti digital, vinyl, tape, CD, hi-res or whatever. Unfortunately as has been frequently noted, the digital equipment used in most professional recording today is far from the state-of-the-art, and the music recorded with more SOTA digital gear is not of the appeal or quality of that which is not.

There is so much we just don't know about audio perception, and relatively little valid research compared to that for video; vision is felt to be far more important than hearing to most people. Very little (if any) perceptual research involving hearing would be of acceptable quality if it were in the biomedical field, and even in that field only about 5-10% of what is published stands up to critical analysis of methods and conclusions. So take any "conclusions" about differences (or lack of) between different digital formats with a healthy amount of skepticism.
Click to expand...
I have never seen those numbers before. Can you share the source of this information?
 
TheOctopus said:
I’m not finding the numbers mentioned.
Click to expand...

Small wonder as the dynamic range of plain 16 bit PCM is 96 dB.
This doesn’t mean that it is limited to 96.
Somebody mentioned Pacific Microsonics. They invented the HDCD
Indeed there are tricks to increase the dynamic range but observe it is no longer Redbook.
High Definition Compatible Digital - Wikipedia.

Likewise you can get 96 db out of 14 bits.
This happened to Philips. They sought the CD would be 14 bits but Sony won and it became 16.
All Philips DACs where 14 bit at that time.
Using noise shaping, they managed to squeeze the desired 96 dB dynamic range out of them.
Maybe the most radical example is DSD, it is single bit but using noise shaping the dynamic range is close to 150 dB.

Anyway, using noise shaping or any other technique, you can get any dynamic range but is no longer plain PCM, no longer CD (Redbook), hence you need dedicated hardware.

BTW: if you want a bigger dynamic range, plain 24 bit PCM will do the job.
It goes down to -144 dBFS ; way below the noise floor of any recording or playback chain.
All modern DAC’s support this.
No need for special hardware.😊
 
How do you avoid the fact that music which goes through the proverbial cocktail blender of analog converted to digital and then converted back again doesn’t lose something? And we won’t even begin to discuss the overuse of compression with digital. And yes, I’ve heard all the arguments about RIAA being a form of compression too. Yeah, yeah, just listen I say! I’ll take RIAA over digital compression every day of the week.

An analog recording to tape pressed to vinyl sounds superior in so many ways. Dimensionality being one of them. Naturalness in tonality being another. I’m surrounded by the best DAC’s in the world. You won’t convince me otherwise. My employees and I shake our heads all day long doing shootouts. Yesterday was a pure analog recording: Classic Records Led Zep II Vs every single digital version we could find. No contest. The CR version was better in every way. The cymbals sounded so natural. The entire presentation is more believable in so many ways.

If it’s recorded digitally from the get go, you may as well stick to that, for better or worse. Digital recordings pressed to vinyl sound, well, like digital.

And we don’t listen to scopes from the 1970’s either.

The digital Vs analog has been beaten to death. Even Mike L who, like my store, has some of the best digital, vinyl and tape equipment will no doubt agree. Even my long running thread “why no turntable?” converted some hard core naysayers once they tried it.

It’s not really worth getting into it again is it? If you’re impressed by the specs of digital, fantastic, stick to digital. Today, digital is very good. But without a turntable, how do you really know?


Sent from my iPhone using Tapatalk Pro
 
TheOctopus said:
A symphony orchestra playing full out can easily reach 96 to 98 decibels, and certain brass and percussion instruments have registered 130 to 140 at close range. The decibel debate: Sound and the symphony - Chicago Tribune

So there is that...
Click to expand...

Yeah, and jet engines put out over 120dB, but no one is capturing 120dB recordings of that either. Just because a symphony orchestra can hit peaks of 96dB-98dB doesn't mean any recording of said performances are capturing that amount of dynamic range.

TheOctopus said:
Then there is the fact that while it may be barely possible to squeeze that kind of dynamic range on digital, 24 bits is theoretically 144dB, from softest to loudest crescendo/passage, I’m not aware of many systems that can reproduce it. The typical noise floor of a quiet room is approximately 30-35dB so one needs to be above that for the softest and add 105 on top. My system can’t do that nor would I ever attempt it.
Click to expand...

Never mind the theoretical ability to record 144dB of dynamic range, we aren't coming close to capturing 96dB of dynamic range and making commercial copies of the digital files available for purchase or streaming. We still have some people that think that because 16/44.1 is capable of 96dB of dynamic range, that means that every 16/44.1 recording they play has 96dB of dynamic range.

The real truth is that most people who have good stereo systems and source material would do backflips if the recordings they played averaged 70dB of dynamic range.
 
mep said:
The real truth is that most people who have good stereo systems and source material would do backflips if the recordings they played averaged 70dB of dynamic range.
Click to expand...

I wonder if this is correct.
Let’s assume we have a proper 24 bit recording with sufficient headroom.
The loudest a DAC can play is 0 dBFS
Assume 20 dB headroom so the loudest passage is -20 dBFS.
Lets also assume a very quiet recording chain so indeed there is musical information at -120 dBFS
Effectively this recording will have a dynamic range of 100 dB.
Obvious if we turn down the volume to 0, we won’t hear a thing.
If we turn it up a little we will hear something.
To hear the full dynamic range we probably need to amplify in such a way that the moment this -20 dBFS is played, we have a SPL of 100 dB at the listening place.

As far as I could judge, there is no relation between the dynamic range and how loud we decided to play.
 
rbbert said:
...

xiph.org and Monty Montgomery have been disseminating their propaganda for years (I didn't look at the dates on the sources above but it's been at least 10 years since I first ran across it, maybe 15)

I am very pro-digital recording and storage, but no one is helped by biased information sharing such as that in the OP, either pro or anti digital, vinyl, tape, CD, hi-res or whatever. Unfortunately as has been frequently noted, the digital equipment used in most professional recording today is far from the state-of-the-art, and the music recorded with more SOTA digital gear is not of the appeal or quality of that which is not.

...
Click to expand...

The term “propaganda” is typically used to describe something misleading. Both videos, however, show reproducible, scientific experiments based on mathematically proven (i.e., demonstrated) theorem such as the Nyquist-Shannon Theorem (Nyquist–Shannon sampling theorem - Wikipedia) which in turn uses Fourier transform (Fourier transform - Wikipedia). The Fourier transform have been around since the 19th century; the Nyquist-Shannon Theorem was mathematically demonstrated/proven (Mathematical proof - Wikipedia) in 1930s. Actually, to label or suggest that such scientific evidence is “propaganda” without providing any scientific backup does strike me of propaganda.
 
Vincent Kars said:
I wonder if this is correct.
Let’s assume we have a proper 24 bit recording with sufficient headroom.
The loudest a DAC can play is 0 dBFS
Assume 20 dB headroom so the loudest passage is -20 dBFS.
Lets also assume a very quiet recording chain so indeed there is musical information at -120 dBFS
Effectively this recording will have a dynamic range of 100 dB.
Obvious if we turn down the volume to 0, we won’t hear a thing.
If we turn it up a little we will hear something.
To hear the full dynamic range we probably need to amplify in such a way that the moment this -20 dBFS is played, we have a SPL of 100 dB at the listening place.

As far as I could judge, there is no relation between the dynamic range and how loud we decided to play.
Click to expand...

These two statements are in conflict with each other. However, your statement that there is no relation between dynamic range and how loud you set your playback levels are true. The digital loudness wars proved that. The race to embrace compression in order to achieve maximum recording levels showed that you could have very loud recordings with zero dynamic range with some people not understanding the difference.
 
Fourier transform is also that funky stuff that makes it possible for the MRI, CAT scan, Ultrasound and other medical devices to image the patient giving the doctor a diagnosis. I don't think anyone can diminish that point. :)
 
My philosophy, in this madness of directional fuses and all kinds of illogical things that are high end audio, has been unwavering for the past decade. Enjoy whatever connects with the inner music lover more. What we hear on our end, is not what the music sounded like in that moment of time-space... That moment is gone and has been replaced by a copy that has been altered and flavored by many processes necessary to deliver the music to us. So if one focuses more on pleasing the inner music lover and less on the audio perfectionist, I believe there is more satisfaction to be derived from the hobby of listening to recorded music. Media choice is a highly personal preference and they are all flavored one way or another. Listening to music does not have to be about science of perfect sound while wearing a white lab coat, unless of course that is what turns you on with this hobby. :)
 
You must log in or register to reply here.
Back
Top