Can You Provide an Objective Defintion of "Noise" ? ...

We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
Maybe I'm just an outlier.
On 2nd thought, let's not give up so soon on that. What's the current accepted definition of noise?
Source: Oxford English Dictionary.

Noise:

Definition [2]: technical Irregular fluctuations that accompany a transmitted electrical signal but are not part of it and tend to obscure it.

Definition [2.1] Random fluctuations that obscure or do not contain meaningful data or other information.

From: http://www.oxforddictionaries.com/us/definition/american_english/noise

.

Deterministic (being the adjective form of the noun "determinism", which is defined below):

The doctrine that all events, including human action, are ultimately determined by causes external to the will.

From: http://www.oxforddictionaries.com/us/definition/american_english/determinism

.

Irregular:

Definition [1.1]: Occurring at uneven or varying rates or intervals.

From: http://www.oxforddictionaries.com/us/definition/american_english/irregular

.
Detail Man wrote:

One little "beef" (that also came up when conversing with GB on his recent thread). If one follows your above statement, then it would seem that any periodic ("aka "pattern") "noise" that appears within a recorded image is (also, accordingly) not considered by you to be "noise" ? If and when such things appear in recorded images, it seems (for me) hard to refer to them as constituting "desired signal(s)". Is it the case that such periodic phenomena are not "noise" ?

Does it seem to make reasonable sense in your thinking to fashion three separate categories:

"random noises"; and

"signals"; and

"periodic signals" generated by the imaging hardware that are not "signal" and are not "noise" ?
JimKasson wrote:

First, in my way of thinking, more than one noise source generates "noise" not "noises".

If you must construct your imaging system model this way, please substitute "deterministic"* for periodic.

Thanks,

Jim

*always producing the same output if the underlying machine passes through the same sequence of states.


From: http://www.dpreview.com/forums/post/54682753.

.

It seems to me that Jim's stated description of the term "determinism" differs from the definition existing in the Oxford English Dictionary. Wikipedia states the following information:

Determinisic System:

In mathematics and physics, a deterministic system is a system in which no randomness is involved in the development of future states of the system. A deterministic model will thus always produce the same output from a given starting condition or initial state.

From: http://en.wikipedia.org/wiki/Deterministic_system

.

Whether one is describing periodic readout components (generated within the imaging device itself) existing within recorded image-data that are unrelated to the scene recorded, or whether one is describing physical scratches existing on a vinyl disc used for reproducing recorded audio sounds producing sounds unrelated to the originally recorded sounds, ...

... both cases of such (aesthetically undesired, and clearly unrelated to the original scene or sounds intended to be represented) phenomena propagating through and to the output of any deterministic system(s) (as defined on the Wikipedia web-page quoted from and linked-to above) represent what may be reasonably considered to constitute undesirable components (arising soley out of the machinery involved in recording, processing, storage, or reproduction).

Both examples presented above are (for most viewers/listeners, anyway) considered to constitute what are undesirable components that are clearly not related to the original scene/sounds intended to be recorded, processed, and subsequently presented to our sensory faculties.

The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.

DM
 
Last edited:
Surely you'd agree that a scratch on a record produces noise, and that such noise has a deterministic cause, no?
No, I wouldn't. Not in this context. But, like you say, I'm an engineer, and probably incapable of understanding you.

[Added later: Or, more to the point, I'm comfortable in using words that are part of the common parlance that have different meaning in engineering and color science contexts than they do in normal speech. "Negative feedback", is an example, which I noticed recently that none other than Robert Rubin used to mean what I'd call "positive feedback with adverse results". The word "color" (or colour, depending on your location), is a word that, when I'm using it as a color scientist, has specific meaning which varies with context, and in no case, should I use it in those ways when I'm picking out paint with my wife, or I'll be in big trouble. "Hue" is another word like that. So is "brightness". So I'm not uncomfortable using the word noise in a particular way in a particular context.]
Just trying to disambiguate. Are you saying that what the scratch on my record produces on playback is not noise? Or are you saying that it doesn't have a deterministic cause? Imagine a thousand scratches on one record, enough to produce a random distribution of sonic disturbances on playback. Does that move us closer together in your view?

We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
I think the scratch(es) would probably correspond to PRNU in our lingo so, interpreting Jim, it would fall in a tricky gray area which could either be considered noise or not noise depending on the context. Sounds like he's ok with calling PRNU noise .
 
Surely you'd agree that a scratch on a record produces noise, and that such noise has a deterministic cause, no?
No, I wouldn't. Not in this context. But, like you say, I'm an engineer, and probably incapable of understanding you.

[Added later: Or, more to the point, I'm comfortable in using words that are part of the common parlance that have different meaning in engineering and color science contexts than they do in normal speech. "Negative feedback", is an example, which I noticed recently that none other than Robert Rubin used to mean what I'd call "positive feedback with adverse results". The word "color" (or colour, depending on your location), is a word that, when I'm using it as a color scientist, has specific meaning which varies with context, and in no case, should I use it in those ways when I'm picking out paint with my wife, or I'll be in big trouble. "Hue" is another word like that. So is "brightness". So I'm not uncomfortable using the word noise in a particular way in a particular context.]
Just trying to disambiguate. Are you saying that what the scratch on my record produces on playback is not noise? Or are you saying that it doesn't have a deterministic cause? Imagine a thousand scratches on one record, enough to produce a random distribution of sonic disturbances on playback. Does that move us closer together in your view?

We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
I think the scratch(es) would probably correspond to PRNU in our lingo so, interpreting Jim, it would fall in a tricky gray area which could either be considered noise or not noise depending on the context. Sounds like he's ok with calling PRNU noise .
The audio signal resulting from phonograph-stylus confronting a single physical scratch on a disc are surely a bit complex - in that they represent a deformation of a vinyl record "groove" existing upon the (nearly) vertical inner surfaces within which the stereo audio information is "cut" in a (physically) matrixed form (L+R signal, and L-R signal) after also being "pre-emphasized" (a frequency response with an increasing magnitude transfer-function over a range of increasing frequencies applied).

Note that the (relevantly steady) repetion-rate of the scratch's sonic impact - and whatever their (resembling) "impulsive" composite spectral signatures may be - is clearly not a random event ...

(Based upon a linear audio signal-path), the superposition in time of a number of individual physical scratches seems to "fare no better" (unless, perhaps, their numbers approach unbearable levels).

One might argue that humans do not at very well perceive audio frequencies corresponding to 33.333, 45, etc. revolutions per Minute. However, two things about such physical scratches seem clear. They can quite often be audible to the listener in what are perceived to be aesthetically objectionable extents, and the spectral signature poduced by a phonograph stylus (when confronting a "scratch" upon each individual revolution of the vinyl record disc) seems to be "anybody's guess" where it might come to the arts of "modelling and prediction". Clear as mud. I do not see the "conceptual bridge" between such mysteries and "PRNU" myself. What is your thinking ?

.

On the other hand, regarding "PRNU noise", Emil Martinec published:

Pixel response non-uniformity (PRNU) is an additional effect which can be measured via the same technique just outlined. The non-uniformity turns out to be only a fraction of a percent, and consequently it is justifiable to treat this random variation in pixel response as another noise source to be added in quadrature to read and shot noise.
 
Last edited:
The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.

OK, DM. You don't like my answer. You want some thing that cleaves desired from undesired. Got it. What is your answer to the question you originally posed?

Jim
 
The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.
OK, DM. You don't like my answer. You want some thing that cleaves desired from undesired. Got it. What is your answer to the question you originally posed?
Again my take. Not speaking for DM.

I posited that being "accepted as signal" by an information consumer and simultaneously "deleterious to its purpose" (and ultimately /our/ purposes) were two such characteristics.

DM is in a related area with his notion of "desired". Our desires are keyed to satisfying the purposes in our undertakings.

If our information consumer (a device in this case) did not somehow "accept" something in the first place, it could never be deleterious to its purpose. An AM receiver accepts impulse noise which modifies the modulated signal, thereby degrading it as a matter of degree.

A single-ended connection works similarly. However once we use a balanced line, then the spurious impulse noise is rejected, or "not accepted as signal".
 
OK, DM. You don't like my answer. You want some thing that cleaves desired from undesired. Got it. What is your answer to the question you originally posed?
Again my take. Not speaking for DM.

I posited that being "accepted as signal" by an information consumer and simultaneously "deleterious to its purpose" (and ultimately /our/ purposes) were two such characteristics.

DM is in a related area with his notion of "desired". Our desires are keyed to satisfying the purposes in our undertakings.

If our information consumer (a device in this case) did not somehow "accept" something in the first place, it could never be deleterious to its purpose. An AM receiver accepts impulse noise which modifies the modulated signal, thereby degrading it as a matter of degree.

A single-ended connection works similarly. However once we use a balanced line, then the spurious impulse noise is rejected, or "not accepted as signal".
So, Luke, is capture metameric error noise? Is barrel distortion noise? Is clipping noise? They all seem to fit your definition.

Jim
 
The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.
OK, DM. You don't like my answer. You want some thing that cleaves desired from undesired. Got it. What is your answer to the question you originally posed?
Hi Jim,

My intention in asking others if they could fashion an objective definition for what (ought, implied) be considered "noise" is intended to get to a useful "cleavage" (as you say) in the course of numerical measures of (shall we say, in the most erudite of terms) lens-camera system "gooderness" ... :P

Seeems like nobody wants to touch a definition of "signal" - which is somewhat understandable (as what is, in presentation to our senses, "desireable" may well have subjective aesthetic elements).

So, I thought it would be interesting to (instead) approach such a goal (of considering metrics of system "gooderness") by attemping to define what aspects exist of that which is considered "undesireable" (at least, by some significant proportion of viewers).

Given some particular controlled situation such as shooting a test-target of some sort where the lens-camara system is illuminated with some particular controlled spectral source (ie, CIE-D50,etc.), my hope is that it should be possible to fashion a metric that includes certain non-random elements (in particular, periodic components of image-sensor readout processes).

Your listing of various (and variable) elements that are introduced by the lens-system that ought not be considered makes sense, and I indicated agreement with all such thoughts - excepting what appear to be random elements of camera/subject-movement. However, I make an assumption that the test-procedure in mind would take place with positionally stable camera system and test-scene.

I see that you have asked (Luke):

Is capture metameric error noise? Is barrel distortion noise?

My answer would be "no".

Is clipping noise?

My answer would be that, while surely "undesirable", such is under user/tester control - and is a phenomena that it would be important that any test-procedure would ensure is not a factor.

My interest and concern is largely born out of statements made by some persons where (it seems to me, anyway) that periodic components arising out of imaging hardware - bearing no relation to the scene photographed, and probably largely agreeably "undesirable" - can (or ought) to be discounted/ignored (conceptually, and presumably also in the derivation of metrics) merely on the basis that they are spectrally random in nature. Such approaches seem to me puzzling - as they seem to exist in theoretical worlds that bear limited relation to the viewing experience.

If anything considered to be "noise" must necessarily consist only of entirely random elements, this not only ignores the usage of the term "noise" in electronic engineering as (also) including elements that are periodic (or "almost periodic") in their nature - but (more importantly) ignores elements that (it seems to me) rightly should be taken into account numerically within the elements considered in the derivation of such "metrics" (derived from measurements, not theory).

My interest is in metrics that sincerely attempt to quantify the nature of the "viewer experience" where it comes to actual measured lens-camera systems (as opposed to theoretical metaphors) - which is "part and parcel" of "reducing ideas into actual practice".

DM
 
Last edited:
Seeems like nobody wants to touch a definition of "signal" - which is somewhat understandable (as what is, in final presentation to our senses, "desireable" may well have subjective aspects).

So, I thought it would be interesting to (instead) approach such a goal (of considering metrics of system "gooderness") by attemping to define what aspects of that which is considered "undesireable" (at least by most viewers).

Given some particular controlled situation such as shooting a test-target of some sort where the lens-camara system is illuminated with some particular controlled spectral source (ie, CIE-D50,etc.), my hope is that it should be possible to fashion a metric that includes certain non-random elements (in particular periodic components of image-sensor readout processes).

Your listing of various (and variable) elements that are introduced by the lens-system to not be considered makes sense, and I indicated agreement with such thoughts - excepting what appear to be random elements of camera/subject-movement. However, I make an assumption that the test-procedure in mind would take place with positionally stable camera system and test-scene.

I see that you have asked (Luke):

Is capture metameric error noise? Is barrel distortion noise?

My answer would be "no".
In that case, I don't understand how you are deciding what to call noise.
Is clipping noise?

My answer would be that, while surely "undesirable", such is under user/tester control - and is a phenomena that it would be important that any test-procedure would ensure is not a factor.

My interest and concern is largely born out of statements made by some persons where (it seems to me, anyway) that periodic components arising out of imaging hardware - bearing no relation to the scene photographed, and probably agreeably "undesirable" - can (or ought) to be discounted/ignored (conceptually, and presumably also in the derivation of metrics merely on the basis that they are spectrally random in nature). Such approaches seem to me puzzling - as they seem to exist in theoretical worlds that bear limited relation to the viewing experience.

If anything considered to be "noise" must necessarily consist only of entirely random elements, this not only ignores the usage of the term "noise" in electronic engineering as (also) including elements that are periodic (or "almost periodic") in their nature - but (more importantly) ignores elements that (it seems to me) rightly should be taken into account numerically within the elements considered in the derivation of such "metrics" (derived from measurements, not theory).
MM, it will come as no surprise to you that, as a long-time double-E (50 years as of last June), I disagree about your characterization of how the term noise is used in electrical engineering, although I admit that there's more than a little inconsistancy. I could provide some examples to buttress my contention. However, I seem to be the only one taking my position here, and I don't see my convincing you or Luke of the rectitude of my position as a likely occurrence.

And so, I will bow out of this discussion unless there are specific questions for me. I wish you and Luke luck with your endeavors.

Jim
 
Seeems like nobody wants to touch a definition of "signal" - which is somewhat understandable (as what is, in final presentation to our senses, "desireable" may well have subjective aspects).

So, I thought it would be interesting to (instead) approach such a goal (of considering metrics of system "gooderness") by attemping to define what aspects of that which is considered "undesireable" (at least by most viewers).

Given some particular controlled situation such as shooting a test-target of some sort where the lens-camara system is illuminated with some particular controlled spectral source (ie, CIE-D50,etc.), my hope is that it should be possible to fashion a metric that includes certain non-random elements (in particular periodic components of image-sensor readout processes).

Your listing of various (and variable) elements that are introduced by the lens-system to not be considered makes sense, and I indicated agreement with such thoughts - excepting what appear to be random elements of camera/subject-movement. However, I make an assumption that the test-procedure in mind would take place with positionally stable camera system and test-scene.

I see that you have asked (Luke):

Is capture metameric error noise? Is barrel distortion noise?

My answer would be "no".
In that case, I don't understand how you are deciding what to call noise.
Is clipping noise?

My answer would be that, while surely "undesirable", such is under user/tester control - and is a phenomena that it would be important that any test-procedure would ensure is not a factor.

My interest and concern is largely born out of statements made by some persons where (it seems to me, anyway) that periodic components arising out of imaging hardware - bearing no relation to the scene photographed, and probably agreeably "undesirable" - can (or ought) to be discounted/ignored (conceptually, and presumably also in the derivation of metrics merely on the basis that they are spectrally random in nature). Such approaches seem to me puzzling - as they seem to exist in theoretical worlds that bear limited relation to the viewing experience.

If anything considered to be "noise" must necessarily consist only of entirely random elements, this not only ignores the usage of the term "noise" in electronic engineering as (also) including elements that are periodic (or "almost periodic") in their nature - but (more importantly) ignores elements that (it seems to me) rightly should be taken into account numerically within the elements considered in the derivation of such "metrics" (derived from measurements, not theory).
MM, it will come as no surprise to you that, as a long-time double-E (50 years as of last June), I disagree about your characterization of how the term noise is used in electrical engineering, although I admit that there's more than a little inconsistancy. I could provide some examples to buttress my contention. However, I seem to be the only one taking my position here, and I don't see my convincing you or Luke of the rectitude of my position as a likely occurrence.

And so, I will bow out of this discussion unless there are specific questions for me. I wish you and Luke luck with your endeavors.
Jim, no disrespect is intended. The use of the term "noise" in EE is a subset, perhaps a small subset, of its application. You can have noise in biological systems that might or might not be considered electrical. Auditory noise, for example, is mechanical in nature, and it is only a contingent matter of fact that it should ever involve electrical signals in downstream processing. I'm looking for a definition that is necessary (i.e., true in all possible worlds). However, this is not to suggest that decades of work in electrical engineering should be devalued. We both have a claim on this real estate, and for my purposes, that's just fine.
 
OK, DM. You don't like my answer. You want some thing that cleaves desired from undesired. Got it. What is your answer to the question you originally posed?
Again my take. Not speaking for DM.

I posited that being "accepted as signal" by an information consumer and simultaneously "deleterious to its purpose" (and ultimately /our/ purposes) were two such characteristics.

DM is in a related area with his notion of "desired". Our desires are keyed to satisfying the purposes in our undertakings.

If our information consumer (a device in this case) did not somehow "accept" something in the first place, it could never be deleterious to its purpose. An AM receiver accepts impulse noise which modifies the modulated signal, thereby degrading it as a matter of degree.

A single-ended connection works similarly. However once we use a balanced line, then the spurious impulse noise is rejected, or "not accepted as signal".
So, Luke, is capture metameric error noise? Is barrel distortion noise? Is clipping noise? They all seem to fit your definition.
The first case I don't know enough about to comment on.

The second case is acceptable to me under the broadened definition. In cases where we restrict ourselves to electrical or photoelectrical phenomena, then you could certainly exclude it just on the grounds that it is not such a phenomenon. But insofar as my definition is concerned when applied solely to electrical phenomena, then as near as I can tell, my definition will fit and explain all of the traditional cases that you know well.

The third case is a bit trickier, partly because it involves /signal loss/. I'd have to think about that a bit more.
 
Jim, no disrespect is intended.
Luke, I didn't sense any. I'm not going away mad, it's just that you guys are headed in a direction that I can't help you with.

Jim
 
Seeems like nobody wants to touch a definition of "signal" - which is somewhat understandable (as what is, in final presentation to our senses, "desireable" may well have subjective aspects).

So, I thought it would be interesting to (instead) approach such a goal (of considering metrics of system "gooderness") by attemping to define what aspects of that which is considered "undesireable" (at least by most viewers).

Given some particular controlled situation such as shooting a test-target of some sort where the lens-camara system is illuminated with some particular controlled spectral source (ie, CIE-D50,etc.), my hope is that it should be possible to fashion a metric that includes certain non-random elements (in particular periodic components of image-sensor readout processes).

Your listing of various (and variable) elements that are introduced by the lens-system to not be considered makes sense, and I indicated agreement with such thoughts - excepting what appear to be random elements of camera/subject-movement. However, I make an assumption that the test-procedure in mind would take place with positionally stable camera system and test-scene.

I see that you have asked (Luke):

Is capture metameric error noise? Is barrel distortion noise?

My answer would be "no".
In that case, I don't understand how you are deciding what to call noise.
The following statements assume linearity in the photosite-assemblies and ADC conversions.

Well, I agreed with you that:
  • Flare.
  • Diffraction.
  • Lens aberrations.
  • Lens falloff.
... are not considered by me to be "noise" here. ... so I don't understand your lack of understanding expressed above. For a given test-scene and physical lens-system placement (independent of camera settings), none of the above vary - whereas periodic image-sensor readout (dare I call it "noise") is not a desirable component of recorded images. It's magnitude relative to the maximum ADC ADU count (in most cases) does vary with changes in camera settings.

"Pixel-level crosstalk" is complex phenomenon involving "signal" as well as "noise" (as well as whatever you would prefer to call spectrally periodic phenomena arising out of the camera hardware). Due to that (upon further reflection) I would not entirely discount it as having "noise" components. How one might endeavor to actually measure "pixel-level crosstalk" seems unclear.
Is clipping noise?

My answer would be that, while surely "undesirable", such is under user/tester control - and is a phenomena that it would be important that any test-procedure would ensure is not a factor.

My interest and concern is largely born out of statements made by some persons where (it seems to me, anyway) that periodic components arising out of imaging hardware - bearing no relation to the scene photographed, and probably agreeably "undesirable" - can (or ought) to be discounted/ignored (conceptually, and presumably also in the derivation of metrics merely on the basis that they are spectrally random in nature). Such approaches seem to me puzzling - as they seem to exist in theoretical worlds that bear limited relation to the viewing experience.

If anything considered to be "noise" must necessarily consist only of entirely random elements, this not only ignores the usage of the term "noise" in electronic engineering as (also) including elements that are periodic (or "almost periodic") in their nature - but (more importantly) ignores elements that (it seems to me) rightly should be taken into account numerically within the elements considered in the derivation of such "metrics" (derived from measurements, not theory).
MM, it will come as no surprise to you that, as a long-time double-E (50 years as of last June), ...
Happy "silver technical anniversary" to you, sir !
... I disagree about your characterization of how the term noise is used in electrical engineering, although I admit that there's more than a little inconsistancy. I could provide some examples to buttress my contention.
No need to do so. I have read articles (I suspect numbering in the hundreds over the years) that have within their text utilized phrases such as "switching noise", "transient noise", etc. relating to what are indeed non-random periodic processes existing within electronic circuits andsystems in electronics industry periodicals such as Electronic Design, Electronic Design Notes, EE Times, etc.
However, I seem to be the only one taking my position here, and I don't see my convincing you or Luke of the rectitude of my position as a likely occurrence.

And so, I will bow out of this discussion unless there are specific questions for me. I wish you and Luke luck with your endeavors.
As stated, the inspiration of this thread arose out of what I view as a seemingly strange tendency on the part of some (Great Bustard, and it seems perhaps yourself ?) to exempt the consideration and the quantifiction of periodic components appearing in recording images that arise soley out of the camera hardware itself (as opposed to the photographed scene) when considering applicable elements to be included in "metrics" intended to describe system "gooderness" from the actual perspective of the "perceptual aesthetic experience" of viewers - and thus seems clearly relevant.

I don't mean to speak for you, though. Do you perceive any reason why such should be exempted ?

DM
 
Last edited:
We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
Maybe I'm just an outlier.
On 2nd thought, let's not give up so soon on that. What's the current accepted definition of noise?
Source: Oxford English Dictionary.

Noise:

Definition [2]: technical Irregular fluctuations that accompany a transmitted electrical signal but are not part of it and tend to obscure it.

Definition [2.1] Random fluctuations that obscure or do not contain meaningful data or other information.

From: http://www.oxforddictionaries.com/us/definition/american_english/noise
That works quite nicely for me.
Whether one is describing periodic readout components (generated within the imaging device itself) existing within recorded image-data that are unrelated to the scene recorded, or whether one is describing physical scratches existing on a vinyl disc used for reproducing recorded audio sounds producing sounds unrelated to the originally recorded sounds, ...

... both cases of such (aesthetically undesired, and clearly unrelated to the original scene or sounds intended to be represented) phenomena propagating through and to the output of any deterministic system(s) (as defined on the Wikipedia web-page quoted from and linked-to above) represent what may be reasonably considered to constitute undesirable components (arising soley out of the machinery involved in recording, processing, storage, or reproduction).

Both examples presented above are (for most viewers/listeners, anyway) considered to constitute what are undesirable components that are clearly not related to the original scene/sounds intended to be recorded, processed, and subsequently presented to our sensory faculties.

The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.

DM
Systematic "undesirable components", such as banding and flare would not be sources of noise, in my opinion (which is not to say, of course, that they are not more "evil" than noise itself).
 
We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
Maybe I'm just an outlier.
On 2nd thought, let's not give up so soon on that. What's the current accepted definition of noise?
Source: Oxford English Dictionary.

Noise:

Definition [2]: technical Irregular fluctuations that accompany a transmitted electrical signal but are not part of it and tend to obscure it.

Definition [2.1] Random fluctuations that obscure or do not contain meaningful data or other information.

From: http://www.oxforddictionaries.com/us/definition/american_english/noise
That works quite nicely for me.
Whether one is describing periodic readout components (generated within the imaging device itself) existing within recorded image-data that are unrelated to the scene recorded, or whether one is describing physical scratches existing on a vinyl disc used for reproducing recorded audio sounds producing sounds unrelated to the originally recorded sounds, ...

... both cases of such (aesthetically undesired, and clearly unrelated to the original scene or sounds intended to be represented) phenomena propagating through and to the output of any deterministic system(s) (as defined on the Wikipedia web-page quoted from and linked-to above) represent what may be reasonably considered to constitute undesirable components (arising soley out of the machinery involved in recording, processing, storage, or reproduction).

Both examples presented above are (for most viewers/listeners, anyway) considered to constitute what are undesirable components that are clearly not related to the original scene/sounds intended to be recorded, processed, and subsequently presented to our sensory faculties.

The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.
Systematic "undesirable components", such as banding and flare would not be sources of noise, in my opinion (which is not to say, of course, that they are not more "evil" than noise itself).
Aha ! Finally, a "peep" out of his eminence Lord Bustard. So ... if we are going to endeavor to fashion "metrics" surrounding image-sensor "righteousness/awesomeness/gooderness", does it not make sense to (at least attempt to) "factor in" such "evils" existing within said devices ? Please advise ...

Please note that flare results from optical sources interacting with lens-systems - whereas "banding" (which you have stated that - though "evil" - you do not consider to be "noise") arises within the very image-sensors that you were recently hoping to fashion a "metric" for ...

.

I took a wrong turn, it was a right turn
My turn to have me a ball
The boys at the shop told me just where to stop
If I, wanted to play for it all
I didn't know I'd find her on daytime TV
My whole world lies waiting behind door number three

I chose my apparel, I wore a beer barrel
And they rolled me to the very first row
I held a big sign that said,
'Kiss me I'm a baker, and Monty I sure need the dough'
Then I grabbed that sucker by the throat until he called on me
Cause my whole world lies waiting behind door number three

And I don't want what Jay's got on his table
Or the box Carol Marel points to on the floor
No I'll hold out just as long as I am able
Until I can unlock that lucky door....well
She's no big deal to most folks
But she's everything to me
My whole world lies waiting behind door number three

Oh Monty, Monty, Monty, I am walking down your hall
Got beat, lost my seat, but I'm not a man to crawl
Though I didn't get rich, you son of a
[expletive deleted],
I'll be back, just wait and see
Cause my whole world lies waiting behind door number three
Yes my whole world lies waiting behind door number three

- "Door Number Three", by Jimmy Buffet and Steve Goodman

.


DM ... :P
 
Last edited:
However, I seem to be the only one taking my position here, and I don't see my convincing you or Luke of the rectitude of my position as a likely occurrence.

And so, I will bow out of this discussion unless there are specific questions for me. I wish you and Luke luck with your endeavors.
As stated, the inspiration of this thread arose out of what I view as a seemingly strange tendency on the part of some (Great Bustard, and it seems perhaps yourself ?) to exempt the consideration and the quantifiction of periodic components appearing in recording images that arise soley out of the camera hardware itself (as opposed to the photographed scene) when considering applicable elements to be included in "metrics" intended to describe system "gooderness" from the actual perspective of the "perceptual aesthetic experience" of viewers - and thus seems clearly relevant.
It helps to include the idea that a complex system involves many subsystems in which what gets passed from one subsystem to another are independently characterized in terms of "signal" and "noise". What one subsystem produces might be considered noise to another subsystem. The power supply subsystem, for example, due to poor filtering, often passes noise into other subsystems, regardless of what the global inputs are.
 
We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
Maybe I'm just an outlier.
On 2nd thought, let's not give up so soon on that. What's the current accepted definition of noise?
Source: Oxford English Dictionary.

Noise:

Definition [2]: technical Irregular fluctuations that accompany a transmitted electrical signal but are not part of it and tend to obscure it.

Definition [2.1] Random fluctuations that obscure or do not contain meaningful data or other information.

From: http://www.oxforddictionaries.com/us/definition/american_english/noise
That works quite nicely for me.
Whether one is describing periodic readout components (generated within the imaging device itself) existing within recorded image-data that are unrelated to the scene recorded, or whether one is describing physical scratches existing on a vinyl disc used for reproducing recorded audio sounds producing sounds unrelated to the originally recorded sounds, ...

... both cases of such (aesthetically undesired, and clearly unrelated to the original scene or sounds intended to be represented) phenomena propagating through and to the output of any deterministic system(s) (as defined on the Wikipedia web-page quoted from and linked-to above) represent what may be reasonably considered to constitute undesirable components (arising soley out of the machinery involved in recording, processing, storage, or reproduction).

Both examples presented above are (for most viewers/listeners, anyway) considered to constitute what are undesirable components that are clearly not related to the original scene/sounds intended to be recorded, processed, and subsequently presented to our sensory faculties.

The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.

DM
Systematic "undesirable components", such as banding and flare would not be sources of noise, in my opinion (which is not to say, of course, that they are not more "evil" than noise itself).
This beckons for a conceptual argument. The dictionary is at its heart, a history book, and its definitions don't substitute for a conceptual argument.
 
However, I seem to be the only one taking my position here, and I don't see my convincing you or Luke of the rectitude of my position as a likely occurrence.

And so, I will bow out of this discussion unless there are specific questions for me. I wish you and Luke luck with your endeavors.
As stated, the inspiration of this thread arose out of what I view as a seemingly strange tendency on the part of some (Great Bustard, and it seems perhaps yourself ?) to exempt the consideration and the quantifiction of periodic components appearing in recording images that arise soley out of the camera hardware itself (as opposed to the photographed scene) when considering applicable elements to be included in "metrics" intended to describe system "gooderness" from the actual perspective of the "perceptual aesthetic experience" of viewers - and thus seems clearly relevant.
It helps to include the idea that a complex system involves many subsystems in which what gets passed from one subsystem to another are independently characterized in terms of "signal" and "noise".
True enough. Not sure why you have stated that, however.
What one subsystem produces might be considered noise to another subsystem.
Possibly.
The power supply subsystem, for example, due to poor filtering, often passes noise into other subsystems, regardless of what the global inputs are.
As stated, I simply would like to see the inclusion of periodic (as well as random) "aesthetically undesirable from a viewer's perspective" elements existing with recorded images included within "metrics" endeavoring to quantify (camera) system "righteousness/awesomeness/gooderness".

Understanding how others define the term "noise" might help me understand others' thinking.
 
Last edited:
However, I seem to be the only one taking my position here, and I don't see my convincing you or Luke of the rectitude of my position as a likely occurrence.

And so, I will bow out of this discussion unless there are specific questions for me. I wish you and Luke luck with your endeavors.
As stated, the inspiration of this thread arose out of what I view as a seemingly strange tendency on the part of some (Great Bustard, and it seems perhaps yourself ?) to exempt the consideration and the quantifiction of periodic components appearing in recording images that arise soley out of the camera hardware itself (as opposed to the photographed scene) when considering applicable elements to be included in "metrics" intended to describe system "gooderness" from the actual perspective of the "perceptual aesthetic experience" of viewers - and thus seems clearly relevant.
It helps to include the idea that a complex system involves many subsystems in which what gets passed from one subsystem to another are independently characterized in terms of "signal" and "noise".
True enough. Not sure why you have stated that, however.
It was an editing mistake, my fault. I was addressing another point by someone else, but accidentally edited it out. Apologies.
 
We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
Maybe I'm just an outlier.
On 2nd thought, let's not give up so soon on that. What's the current accepted definition of noise?
Source: Oxford English Dictionary.

Noise:

Definition [2]: technical Irregular fluctuations that accompany a transmitted electrical signal but are not part of it and tend to obscure it.

Definition [2.1] Random fluctuations that obscure or do not contain meaningful data or other information.

From: http://www.oxforddictionaries.com/us/definition/american_english/noise
That works quite nicely for me.
Whether one is describing periodic readout components (generated within the imaging device itself) existing within recorded image-data that are unrelated to the scene recorded, or whether one is describing physical scratches existing on a vinyl disc used for reproducing recorded audio sounds producing sounds unrelated to the originally recorded sounds, ...

... both cases of such (aesthetically undesired, and clearly unrelated to the original scene or sounds intended to be represented) phenomena propagating through and to the output of any deterministic system(s) (as defined on the Wikipedia web-page quoted from and linked-to above) represent what may be reasonably considered to constitute undesirable components (arising soley out of the machinery involved in recording, processing, storage, or reproduction).

Both examples presented above are (for most viewers/listeners, anyway) considered to constitute what are undesirable components that are clearly not related to the original scene/sounds intended to be recorded, processed, and subsequently presented to our sensory faculties.

The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.
Systematic "undesirable components", such as banding and flare would not be sources of noise, in my opinion (which is not to say, of course, that they are not more "evil" than noise itself).
Aha ! Finally, a "peep" out of his eminence Lord Bustard. So ... if we are going to endeavor to fashion "metrics" surrounding image-sensor "righteousness/awesomeness/gooderness", does it not make sense to (at least attempt to) "factor in" such "evils" existing within said devices ? Please advise ...
Absolutely. As I said, systematic "undesirable components", such as banding and flare can most certainly be more undesirable than noise.
Please note that flare results from optical sources interacting with lens-systems - whereas "banding" (which you have stated that - though "evil" - you do not consider to be "noise") arises within the very image-sensors that you were recently hoping to fashion a "metric" for ...
Indeed.
.

I took a wrong turn, it was a right turn
My turn to have me a ball
The boys at the shop told me just where to stop
If I, wanted to play for it all
I didn't know I'd find her on daytime TV
My whole world lies waiting behind door number three

I chose my apparel, I wore a beer barrel
And they rolled me to the very first row
I held a big sign that said,
'Kiss me I'm a baker, and Monty I sure need the dough'
Then I grabbed that sucker by the throat until he called on me
Cause my whole world lies waiting behind door number three

And I don't want what Jay's got on his table
Or the box Carol Marel points to on the floor
No I'll hold out just as long as I am able
Until I can unlock that lucky door....well
She's no big deal to most folks
But she's everything to me
My whole world lies waiting behind door number three

Oh Monty, Monty, Monty, I am walking down your hall
Got beat, lost my seat, but I'm not a man to crawl
Though I didn't get rich, you son of a
[expletive deleted],
I'll be back, just wait and see
Cause my whole world lies waiting behind door number three
Yes my whole world lies waiting behind door number three

- "Door Number Three", by Jimmy Buffet and Steve Goodman

.


DM ... :P
While I get the reference to my post title, I'm afraid we listen to different music:

*talking*
Oh my god
Becky, look at her butt
Its so big
She looks like one of those rap guys girlfriends
Who understands those rap guys
They only talk to her because she looks like a total prostitute
I mean her butt
It's just so big
I can't believe it's so round
It's just out there
I mean, it's gross
Look, she's just so black

*rap*
I like big butts and I can not lie
You other brothers can't deny
That when a girl walks in with an itty bitty waist
And a round thing in your face
You get sprung
Wanna pull up tough
Cuz you notice that butt was stuffed
Deep in the jeans she's wearing
I'm hooked and I can't stop staring
Oh, baby I wanna get with ya
And take your picture
My homeboys tried to warn me
But that butt you got
Make Me so horney
Ooh, rump of smooth skin
You say you wanna get in my benz
Well use me use me cuz you aint that average groupy

I've seen them dancin'
To hell with romancin'
She's Sweat,Wet, got it goin like a turbo vette

I'm tired of magazines
Saying flat butts are the thing
Take the average black man and ask him that
She gotta pack much back

So Fellas (yeah) Fellas(yeah)
Has your girlfriend got the butt (hell yeah)
Well shake it, shake it, shake it, shake it, shake that healthy butt
Baby got back

(LA face with Oakland booty)

I like'em round and big
And when I'm throwin a gig
I just can't help myself
I'm actin like an animal
Now here's my scandal

I wanna get you home
And UH, double up UH UH
I aint talkin bout playboy
Cuz silicone parts were made for toys
I wannem real thick and juicy
So find that juicy double
Mixalot's in trouble
Beggin for a piece of that bubble
So I'm lookin' at rock videos
Knockin these bimbos walkin like hoes
You can have them bimbos
I'll keep my women like Flo Jo
A word to the thick soul sistas
I wanna get with ya
I won't cus or hit ya
But I gotta be straight when I say I wanna --
Til the break of dawn
Baby Got it goin on
Alot of pimps won't like this song
Cuz them punks lie to hit it and quit it
But I'd rather stay and play
Cuz I'm long and I'm strong
And I'm down to get the friction on

So ladies (yeah), Ladies (yeah)
Do you wanna roll in my Mercedes (yeah)
Then turn around
Stick it out
Even white boys got to shout
Baby got back

(LA face with the Oakland booty)

Yeah baby
When it comes to females
Cosmo ain't got nothin to do with my selection
36-24-36
Only if she's 5'3"

So your girlfriend throws a Honda
Playin workout tapes by Fonda
But Fonda ain't got a motor in the back


;-)
 
We've already got some general acceptance on a cluster definition for "noise", so I don't see a reason /in principle/ why the scratched-record case wouldn't be admitted.
Maybe I'm just an outlier.
On 2nd thought, let's not give up so soon on that. What's the current accepted definition of noise?
Source: Oxford English Dictionary.

Noise:

Definition [2]: technical Irregular fluctuations that accompany a transmitted electrical signal but are not part of it and tend to obscure it.

Definition [2.1] Random fluctuations that obscure or do not contain meaningful data or other information.

From: http://www.oxforddictionaries.com/us/definition/american_english/noise
That works quite nicely for me.
Whether one is describing periodic readout components (generated within the imaging device itself) existing within recorded image-data that are unrelated to the scene recorded, or whether one is describing physical scratches existing on a vinyl disc used for reproducing recorded audio sounds producing sounds unrelated to the originally recorded sounds, ...

... both cases of such (aesthetically undesired, and clearly unrelated to the original scene or sounds intended to be represented) phenomena propagating through and to the output of any deterministic system(s) (as defined on the Wikipedia web-page quoted from and linked-to above) represent what may be reasonably considered to constitute undesirable components (arising soley out of the machinery involved in recording, processing, storage, or reproduction).

Both examples presented above are (for most viewers/listeners, anyway) considered to constitute what are undesirable components that are clearly not related to the original scene/sounds intended to be recorded, processed, and subsequently presented to our sensory faculties.

The premise that - rather than incorporating such phenomena into differentiations between that which is in presentation desired and that which is undesired (as in the perhaps simplisticly, but understandably, worded rhetorical categories of "signal" and "noise") - a third category must be created (in order to satisfy a particular individual definition of "noise" as having to necessarily only apply to random components), seems to me to be substantively counter-productive in what (seem to be) reasonably understandable processes of determining (including numerically quantifying) what may amount to components perceived by human beings as being "desired" and "undesired" components existing within representations of scenes/sounds which are output by systems in visual or auditory presentations provided to our sensory faculties.

DM
Systematic "undesirable components", such as banding and flare would not be sources of noise, in my opinion (which is not to say, of course, that they are not more "evil" than noise itself).
This beckons for a conceptual argument. The dictionary is at its heart, a history book, and its definitions don't substitute for a conceptual argument.
Sure. Words, and their meanings, can change over time. However, when I think of "noise", I always think of random, not systematic, components, and I most certainly distinguish between banding, flare, and noise, for example.
 

Keyboard shortcuts

Back
Top