 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
90.208.167.156
In Reply to: RE: "Bits of resolution" posted by b.l.zeebub on May 22, 2009 at 00:19:11
First off, let me apologise cos I mispoke earlier, you can indeed resolve signal below the noise floor but not below the noise density floor in both analogue and digital, so my apologies for mixing up the two terms in my original post.>> The point is that while in digital resolution and possible noise floor are both determined by the bit rate. This is obviously not the case when it comes to a cutting head in analog, noise floor and resolution are not linked.
That's not strictly correct, the word length does not necessarily determine the resolution, an extreme example being DSD with a word length of 1 bit but a dynamic range (up to 20Hz) of >120dB. Furthermore, as stated above you can resolve signal below the quantisation noise floor up until you hit the noise density floor at the cost of much higher distortion, there are ways round this but that's a separate discussion.
>> good cutting head has an upper frequency limit of 50kHz and thus can reproduce detail above its noise floor that 16/44.1 digital can not match. This can heard by how accurately cymbals are reproducedYou are confusing upper frequency cutoff with the noise floor.
Music making the painting, recording it the photograph
Edits: 05/22/09Follow Ups:
"You are confusing upper frequency cutoff with the noise floor."
a) I am not.
b) You appear to be confusing noise floor/dynamic range with resolution.
There is no point speaking across each other, so let's start over with clear cut definitions.What is your definition of resolution?
What is your definition of dynamic range?
How is the noise floor related to the upper frequency limit?Music making the painting, recording it the photograph
Edits: 05/23/09 05/23/09
Resolution: A systems ability to reproduce fine detail.
Dynamic range: Difference between noise floor and possible max output before overload.
The noise floor is not usually related to upper frequency limit.
However in digital (disregarding DSD) noise floor is related to bit depth and bit depth determines largely the resolution as well (so does sampling rate to some extent). A cutting head is a mechanical devise as such its resolution is directly related to how fast it can react. The speed of reaction also determines upper frequency limit. Thus in case of a cutting head the upper frequency limit is linked to its resolution.
In different domains things relate differently to each other, it's really not that hard to comprehend when you actually think about it.
You seem to have commingled two types of definitions, colloquial and scientific.> > Resolution: A systems ability to reproduce fine detail.
Sounds like a colloquial definition. I cannot comment on this as I am not certain of your definition of 'fine detail', what is it?
> > Dynamic range: Difference between noise floor and possible max output before overload.
This is reasonably strick definition of SNR but not for dynamic range as the dynamic range can exceed the SNR at the cost of higher distortion.
> > However in digital (disregarding DSD) noise floor is related to bit depth and bit depth determines largely the resolution as well (so does sampling rate to some extent).
Actually, this is incorrect as DSD is simply an extreme case of a fairly common digital phenomenon. Here the term "Effective Number of bits (ENOB)" determines the SNR rather than the actual number of bits.
> > A cutting head is a mechanical devise as such its resolution is directly related to how fast it can react. The speed of reaction also determines upper frequency limit. Thus in case of a cutting head the upper frequency limit is linked to its resolution.Looks like you are referring to frequency resolution i.e. bandwidth. In otherwords, LPs have arguably greater frequency extension than CD. It's huge stretch to you use this to explain an LP's ability to resolve signal below the noise floor though.
Music making the painting, recording it the photograph
Edits: 05/26/09
If you do not know what fine detail is you probably can't hear it.
As for dynamic range lets have a look at the figures given by the manufacturer of one of the most highly regarded microphones: Neumann and their U87:
Noise level: 12dB
Maximum SPL: 117dB
Dynamic range: 105dB
Signal to Noise Ratio: 82dB
(all a-weighted, mic set to cardioid response pattern)
I never said anything about resolving anything BELOW noise floor. Got no idea where you got that from or what your point could possibly be. I do know though that it helps reading AND comprehending any post that you try to answer to.
It is not possible to resolve anything below noise floor. The fact that one can have a conversation in a noisy environment is a psycho-acoustic phenomenon and irrelevant here.
Try doing that in a language in which you are capable to converse but not 100% fluent and you'll find it practically impossible. Here it is the brain filling in the gaps (ie the parts which fall below the noise level) with sounds it thinks should be there.
You still do not seem to able to grasp that resolution in a mechanical system is related to the highest frequency it can reproduce. This is not necessarily the case for digital or electro-magnetic systems (like analogue tape).
> > If you do not know what fine detail is you probably can't hear it.That's not a definition, is it?
> > You still do not seem to able to grasp that resolution in a mechanical system is related to the highest frequency it can reproduce. This is not necessarily the case for digital or electro-magnetic systems (like analogue tape).
The frequency resolution, dynamic range and SNR of any device, analog or digital, is related to the highest frequency it can reproduce i.e. bandwidth limited that's why spec sheet you posted states that the figures are A-weighted.
> > I never said anything about resolving anything BELOW noise floor...It is not possible to resolve anything below noise floor.
But then you go ahead and publish specs of a device that does exactly that i.e resolve signal below the noise floor, oh well!
Music making the painting, recording it the photograph
Edits: 05/26/09 05/26/09
Can you actually read?
Or may be it's the maths you struggle with:
Noise floor 12dB, max SPL 117dB, DR 105dB.
117-12=105
The signal to noise ratio is referenced to a 94dB signal level.
94-12=82
At no point do the specs imply any resolution below noise floor.
It really is not that hard…
I have to conclude that you are either purposefully obstinate or far from the sharpest tool in the box. I consider this closed because there is no point arguing with someone who can't even understand the simplest specs.
I am just glad that audio is just your hobby because if it were your honest business your family would clearly starve.
> > The signal to noise ratio is referenced to a 94dB signal level .A point you deliberately omitted in your original post. Anyhow thanks for the correction wrt Neumann U87 spec and the various ad hominems along the way.
Music making the painting, recording it the photograph
Edits: 05/26/09
"You are confusing upper frequency cutoff with the noise floor."
Not really the discussion revolves around resolution and it's a term made popular long before digital audio became common place.
You've been somewhat obstinate in this whole discussion on resolution holding firm to definition and verbiage that serves your purpose.
Truth be it told for a conversation to be worthwhile you need to make an effort at least attempt to understand the points others are making here instead of bullying through with verbiage suiting your POV only.
There's no reason for you to have to obfuscate verbiage in order to dismiss the importance of the following comment -
"> > good cutting head has an upper frequency limit of 50kHz and thus can reproduce detail above its noise floor that 16/44.1 digital can not match. This can heard by how accurately cymbals are reproduced"
So the original poster isn't really confused the question becomes are you incapable of understanding his POV or are you intentonally trying to score points simply on syntaxical grounds?
> > Truth be it told for a conversation to be worthwhile you need to make an effort at least attempt to understand the points others are making here instead of bullying through with verbiage suiting your POV only.Please elaborate.
> > good cutting head has an upper frequency limit of 50kHz and thus can reproduce detail above its noise floor that 16/44.1 digital can not match.
TAH: You are confusing upper frequency cutoff with the noise floor.> > So the original poster isn't really confused the question becomes are you incapable of understanding his POV or are you intentonally trying to score points simply on syntaxical grounds?
On which syntaxical grounds do you feel I am trying to score points?
Music making the painting, recording it the photograph
Edits: 05/23/09 05/23/09
reconstruction process (ie. interpolation and smoothing filters) as well as gain (ie. information loss due to fixed discreet sampling values).
"On which syntaxical grounds do you feel I am trying to score points?"
Most blatantly your comment that the author of the post was confused. He wasn't confused, you just ignored his perspective and the point he was attempting to make based on holding firm to your syntax without consider his.
"Please elaborate."
Bandwidth and resolution are being used differently by different posters. It shouldn't be too difficult to understand what meaning the poster is intending based on his comments.
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
