Post photos of your media room!

[Artisan Fan]

Yeah, the music I listen to was all recorded when lps were the primary format, thus hearing them on LP gives them a warmer, fuller sound like they were ment to be heard, opposed to digitally enhanced music of CDs/DVDs.

Much of this music was likely recorded in analog so that gives the analog format an advantage in addition to the higher resolution of LP over CD.

It's amazing what great finds there are in LPs. On Saturday I bought for a whopping $4 a 50s era Everest recording that was mint. The sound and performance are to die for.

 

[A Y]

Originally Posted by Artisan Fan

Much of this music was likely recorded in analog so that gives the analog format an advantage in addition to the higher resolution of LP over CD.

Umm, no. What was that about half-truths you mentioned earlier?

--Andre

 

[RJman]

Originally Posted by Andre Yew

Umm, no. What was that about half-truths you mentioned earlier?

--Andre

 

[Artisan Fan]

Originally Posted by RJman

Agreed.

 

[dusty]

Originally Posted by Andre Yew

Umm, no. What was that about half-truths you mentioned earlier? --Andre
Technically speaking, music recorded in analog has an infinite resolution, or at least a noise-limited resolution. This argument doesn't really mean much given the bitrates for modern digital recording, but an analog recording will always be more faithful to the original source than a digitized one.

 

[A Y]

Originally Posted by dusty

Technically speaking, music recorded in analog has an infinite resolution, or at least a noise-limited resolution. This argument doesn't really mean much given the bitrates for modern digital recording, but an analog recording will always be more faithful to the original source than a digitized one.

Some points to consider:

1. Infinite resolution implies infinite energy, so nothing has infinite resolution.

2. Noise is the limit of resolution. "Noise-limited resolution" is like saying "darkness-limited light". Analog and digital resolution are always limited by noise.

3. Analog recording is not necessarily more faithful. It depends on the bandwidth and noise of the recording process relative to the source. For example, a cassette tape recording will always be inferior to a 96 kHz/24-bit digital recording.

Analog also has the severe disadvantage in that every copy is lossy.

--Andre

 

[dusty]

Originally Posted by Andre Yew

Some points to consider: 1. Infinite resolution implies infinite energy, so nothing has infinite resolution. 2. Noise is the limit of resolution. "Noise-limited resolution" is like saying "darkness-limited light". Analog and digital resolution are always limited by noise. 3. Analog recording is not necessarily more faithful. It depends on the bandwidth and noise of the recording process relative to the source. For example, a cassette tape recording will always be inferior to a 96 kHz/24-bit digital recording. Analog also has the severe disadvantage in that every copy is lossy. --Andre
Preface: I'm not a signal processing guy. 1. I'm not very well-versed in this sort of thing, so you'll have to refresh my memory as to what you mean here. When I say analog has an infinite resolution, I essentially mean that it has no resolution (and therefore no resolution limit on quality). Doesn't the infinite resolution = infinite energy only apply to digital sampling? 2. How is noise the limit of digital resolution and not the sample rate? Given a sample rate you can calculate the probability that information is lost. If this rate is sufficiently low, isn't that the limit rather than the noise? You can always push the noise lower in a system (until you get to quantum noise), but if you have a crappy sample rate there's nothing you can do about it. Analog recording does not have the sampling problem and is limited by noise only. 3. The fundamental issue at hand is that both digital and analog sources are recorded with analog detectors, and a digital reproduction is by definition a degradation of the analog signal from the detector. What I'm trying to say here, and correct me if I'm wrong, is that a perfect analog recorder is just noise-free, while a perfect digital recorder must be both noise-free and have an infinite sample rate.

 

[A Y]

Originally Posted by dusty

When I say analog has an infinite resolution, I essentially mean that it has no resolution (and therefore no resolution limit on quality). Doesn't the infinite resolution = infinite energy only apply to digital sampling?

Resolution is the ability to tell the difference between two points in a signal. The smaller the difference you can detect, the higher the resolution. The smallest difference you can detect is limited by the noise in the recording. Noise levels have to be lower than the smallest difference you're trying to detect otherwise you couldn't tell the difference between an actual level difference and that caused by random noise.

You also have to be able to tell the difference between very large signals. Infinite resolution implies that you can detect infinitely large differences, hence infinite energy.

2. How is noise the limit of digital resolution and not the sample rate?

Sampling rate limits bandwidth not noise. Bandwidth is the range of low to high frequencies. Noise in digital is limited by the number of bits for each sample. There are tricks you can play to trade off one for the other, but sampling rate in general does not limit resolution.

Analog recording does not have the sampling problem and is limited by noise only.

Analog is limited by other things. For example, the bandwidth of magnetic tape limits analog tape recording to less than 10 octaves. Analog's limitations generally reflect the physical limitation of the physical media, and all that implies. That makes its limitations somewhat unpredictable and dependent on how it was made. For example, an LP has different limitations than a tape recording.

Digital is also limited by physical media, but digital specs make certain guarantees about the noise and bandwidth, so it can be made independent of media limitations.

3. The fundamental issue at hand is that both digital and analog sources are recorded with analog detectors, and a digital reproduction is by definition a degradation of the analog signal from the detector.

Yes, both use analog detectors, but there is a dangerous assumption behind your reasoning that "analog" represents one way of encoding. There are myriad ways of analog encoding and all of them cause degradation. Magnetic tape turns an electrical signal into arrangements of magnetic domains on a piece of plastic. LP turns it into bumps on V-shaped walls on vinyl. Older technologies turn things into little holes or cuts in paper.

Almost all recording is about encoding some signal from one form into another form --- there's always some kind of transformation happening.

Each of these methods is about as unnatural as digital recording, but digital recording is the only method that gives you certain guarantees about what it can and cannot record. I think this is partly the reason why it's so easy to criticize digital recording, because its specs are easily available and comprehensible. In order to make the same criticism (and recognition) of analog's limitations requires a broad range of knowledge related to the physics of a particular analog medium.

--Andre

 

[Artisan Fan]

Analog is limited by other things. For example, the bandwidth of magnetic tape limits analog tape recording to less than 10 octaves. Analog's limitations generally reflect the physical limitation of the physical media, and all that implies. That makes its limitations somewhat unpredictable and dependent on how it was made. For example, an LP has different limitations than a tape recording.

Technical considerations aside, what I have found is that analog tape is capable of storing is extremely high resolution sound. Done well a good analog tape transfer to LP can be extremely high quality.

As for digital recordings, we have been recording at 24/176 for our recent classical and gospel sessions. We often have a backup hard drive running at 24/88.2 or 16/44.1. The difference of 24/176 over 16/44.1 is material and substantial.

So that's my experience on formats at a 30K level after working on dozens of sessions since 1990.

 

[RJman]

Yeah Andre, I think you're going to have to post pictures of your media room before we can take any more posts from you seriously.

 

[A Y]

Originally Posted by RJman

Yeah Andre, I think you're going to have to post pictures of your media room before we can take any more posts from you seriously.

The tinfoil on all of my components is preventing my digital camera from acquiring any images.

--Andre

 

[Artisan Fan]

Originally Posted by Andre Yew

The tinfoil on all of my components is preventing my digital camera from acquiring any images.

--Andre

Maybe you need an analog film camera then....

 

[Artisan Fan]

Maybe this old MTV clip can shed some light:

 

[dusty]

Originally Posted by Andre Yew

Resolution is the ability to tell the difference between two points in a signal. The smaller the difference you can detect, the higher the resolution. The smallest difference you can detect is limited by the noise in the recording. Noise levels have to be lower than the smallest difference you're trying to detect otherwise you couldn't tell the difference between an actual level difference and that caused by random noise.

You also have to be able to tell the difference between very large signals. Infinite resolution implies that you can detect infinitely large differences, hence infinite energy.



Sampling rate limits bandwidth not noise. Bandwidth is the range of low to high frequencies. Noise in digital is limited by the number of bits for each sample. There are tricks you can play to trade off one for the other, but sampling rate in general does not limit resolution.



Analog is limited by other things. For example, the bandwidth of magnetic tape limits analog tape recording to less than 10 octaves. Analog's limitations generally reflect the physical limitation of the physical media, and all that implies. That makes its limitations somewhat unpredictable and dependent on how it was made. For example, an LP has different limitations than a tape recording.

Digital is also limited by physical media, but digital specs make certain guarantees about the noise and bandwidth, so it can be made independent of media limitations.



Yes, both use analog detectors, but there is a dangerous assumption behind your reasoning that "analog" represents one way of encoding. There are myriad ways of analog encoding and all of them cause degradation. Magnetic tape turns an electrical signal into arrangements of magnetic domains on a piece of plastic. LP turns it into bumps on V-shaped walls on vinyl. Older technologies turn things into little holes or cuts in paper.

Almost all recording is about encoding some signal from one form into another form --- there's always some kind of transformation happening.

Each of these methods is about as unnatural as digital recording, but digital recording is the only method that gives you certain guarantees about what it can and cannot record. I think this is partly the reason why it's so easy to criticize digital recording, because its specs are easily available and comprehensible. In order to make the same criticism (and recognition) of analog's limitations requires a broad range of knowledge related to the physics of a particular analog medium.

--Andre

Thanks for clarifying. I think we're on the same page, though I completely forgot about bandwidth considerations.

 

[kronik]

Another audiophile pot-stirring:

DACs aside, what's the difference between a Denon AVR4308 and a McIntosh MC207 in regards to digital SQ? Same signal path (let's say FLAC via HTPC/AuzenTech sound card), same cabling, assumed room/phase-correction, speakers driven by the same amp (bypassing the Denon's internal), etc. I, unfortunately, have neither the resources nor the ability in some cases to audition the differences - I'm very curious as to the assessments of others on this board. We're not talking analog capabilities at all. I realize certain speakers will emphasize flawed component design in some cases.. we're not talking 15k/pair speakers.