android
Go Back   abi>>forums > Software & Media > Software

Reply
 
Thread Tools Search this Thread Display Modes
  #21  
Old 05-07-2012, 03:32 PM
saratoga saratoga is offline
Rockbox Developer / Moderator
 
Join Date: Apr 2007
Posts: 3,627
Default

Rereading these replies, I think the core problem is a misunderstanding about equalization. Some background might help put things into context.

The output of any system (like an amplifier) can be decomposed into a linear and nonlinear component. The linear component (which ideally should be the only component), contains the frequency response of the system. The nonlinear component contains things like harmonics, intermodulation, distortion, etc that give different amplifiers different sounds. For instance, the vaunted sound of a tube amplifier is due to nonlinear interactions between different gain stages that lead to a complicated nonlinear, time varying distortion that is extraordinarily difficult model or reproduce. Meanwhile, a neutral amplifier should have essentially negligible nonlinear components (and so sound neutral because it add no new sounds to the signal).

When we talk about EQ, what we mean is that we apply a linear system to a device that introduces a frequency dependent gain. The key point though is that this is a linear correction, and so it can only impact the linear part of signal. This frequency dependent gain then corrects the linear response of the entire audio chain, from DAC, to amp, to speakers to room acoustics.

The reason this is important is that acoustics of almost any listening environment or pair of headphones have a strongly frequency dependent attenuation/gain. Its virtually impossible to design a room or pair of headphones that has a perfectly flat frequency response, and even if you did the act of putting a human being in it will probably mess up that frequency response. Hence EQ is very important to accurate reproduction. Without it rooms will have uneven bass as different frequencies couple more efficiently or less efficiently. Similar effects happen with headphones.

As for nonlinear effects, they generate new frequencies that were not present in the original signal, and so we cannot process them out. Once you add nonlinearity, you are stuck with it. Contrast that to any linear terms, which you can adjust however you like. Provided the output impedance of an amplifier is low compared to its load (hence, that it is not being used to drive an oversized load), one can always make any amplifier have a perfectly flat frequency response just by putting a trivial EQ circuit into into its input stage. And indeed this is generally done, since its fairly common for amplifiers to naturally have a frequency dependent gain. Doing so however does not mean that have made a transparent amplifier, since you can still have serious nonlinear effects that would render its output poor.

Thus, when doing an amplifier test, EQ is very important, since you don't know how well equalized any given amplifier is internally. And at the same time, you don't care, since you can (and for serious applications, must) set that independently. If you fail to do this, much more serious problems in the amplifiers output (the nonlinear effects) could be masked uneven frequency response leading to spurious test results.

Now of course you absolutely can do double blind tests without applying EQ. In that case you would do so to look at how audible the combined effects of linear and nonlinear effects are. However, this is problematic for several reasons. First, in a high fidelity environment, you will use EQ, and so linear part of the test is not really applicable. Second, since the linear and nonlinear parts of the test cannot be separated, its difficult to interpret the results of the test. In that case for instance, if you have a positive result, would it be due to frequency response (and so not generalizable to other rooms or headphones) or distortion (generalizable to other equipment)? The interpretation of such results is thus very uncertain, and arguably, less interesting.

Ultimately the purpose of any double blind tests depends on what you want to know about audibility. If you care about having a perfectly flat frequency response for some application (perhaps you cannot use an EQ for some reason?) then non-equalized tests might be interesting in spite of their limitations. But generally, that is not the case.
__________________
Interested in Google's Summer of Code ? PM me.
Reply With Quote

Advertisement [Remove Advertisement]

  #22  
Old 05-07-2012, 03:43 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

Quote:
Originally Posted by The DarkSide View Post
Regardless, no matter what you may claim, technobabble is technobabble. I know quite enough to understand that your full of bs. Simple as that,... Wordplay to seem superior, overly intelligent as it may, shows a lack of the ability to actually learn from the conversation. I refuse to delve into all the technical aspects for a REASON. It's totally unnecessary. Simple as that. For information's sake I do research certain things - FLAC vs LAME, the quality of the components & devices I'm interested in, etc. But I'm confident in the knowledge that I've learned enough here, and on the sites linked in some ppl's sig's, to make an educated decision about what's good and bad in the audio world. I do need to over qualify every aspect of it. You do, and that's quite sad.
Not at all. I am an experienced and trained electronics technician. You apparently are not. I do not claim to have superior knowledge of whatever your chosen professional field is, or would I downgrade you for speaking more confidently when discussing it than I likely could.

Quote:
I guess, when it comes down to it, you consider yourself an "audiophile"? I guess I have another reason I don't further my studies in the area,...I refuse to be cursed with that hideous moniker. I'se likes being dumb!!!
Sorta. There are problems in the audio world... Things suggested (such as exotic interconnects or fancy speaker wire) that really are not relevant until you get into longer lengths of wire. On interconnects, it is very rare to need anything "fancy", and anything beyond "midgrade" is generally a waste. With speaker wire, unless you are dealing with a huge room, and large amounts of power, it probably doesn't make a huge amount of difference, either.

With amplifiers, it comes down to the room you are in, the speakers being used, how great of expectation you have, and countered by just how good your hearing really is.

The problems I have are less with the tests themselves, than the conclusions drawn by them and the subsequent universal application of them.
Reply With Quote

  #23  
Old 05-07-2012, 03:53 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

@ Saratoga

Are you arguing that equalization does not alter the level relationship between the fundamental and harmonic frequencies, or can attenuate harmonics?
Reply With Quote

  #24  
Old 05-07-2012, 04:02 PM
saratoga saratoga is offline
Rockbox Developer / Moderator
 
Join Date: Apr 2007
Posts: 3,627
Default

Quote:
Originally Posted by mutescream View Post
The resistive load of the wire itself. Are you ignoring the first law of thermodynamics? What is happening to the energy impeded by the resistive load? It turns into heat. The hotter the wire is, the greater its resistance is.
If you think its the first law of thermodynamics, then you're completely full of crap.

Quote:
Originally Posted by mutescream View Post
As previously stipulated, the relevance of inductance and capacitance is directly proportional to the length of the wire run.
Answer the damn question.

Quote:
Originally Posted by mutescream View Post
Reread the wire challenge again. The problem with it, is that it never acknowledges there are situations where it is NOT applicable.
So I just reread it, and I think it makes no mention at all of this:

Quote:
Originally Posted by mutescream View Post
I was attacking that the test in question indicated that 16/18gau wire was suitable for all applications.
As far as I can tell, nothing you've actually said relates to anything actually tested. Its not even clear to me that you understand the article at all.


Quote:
Originally Posted by mutescream View Post
Does this matter? I am simply arguing that a blanket statement is not universally true.
Yes I think it does. If you're going to tell me that something isn't vanishingly small, I expect you to show that it is not. Do so now.


Quote:
Originally Posted by mutescream View Post
So, what exactly was the point of his test? If you counter everything that makes an amplifier different from its contemporaries, and give it the same sonic signature, it's pretty self evident that it will provide indistinguishable results.
See post above. You need to understand more about differences in amplifers before I can reply to this.

Quote:
Originally Posted by mutescream View Post
Were that not the case (that equalization can compensate for colorization), it would have been much easier to distinguish between tube amps and solid state. We are also talking about human perception here, not lab test equipment (for measuring results).

Were that not the case, we wouldn't have people believing that "loud is always better", we wouldn't subsequently have the poorly mastered recordings we do today (with minimal dynamic headroom) or the need to normalize these tests for volume (for SQ measurements within relative listening levels).
I'm going to try and rewrite this without all the double negatives. If I get it wrong, please let me know:


Quote:
Originally Posted by mutescream View Post
Were that the case (that equalization can't compensate for colorization), it would have been much easier to distinguish between tube amps and solid state.
Translation: If things had been more different the test would have been easier. I think I agree.

Quote:
Originally Posted by mutescream View Post
We are also talking about human perception here, not lab test equipment (for measuring results).
Translation: meaningless

Quote:
Originally Posted by mutescream View Post
Were that the case, we would have people believing that "loud is always better",
Translation: no idea.

Quote:
Originally Posted by mutescream View Post
we wouldn't subsequently have the poorly mastered recordings we do today (with minimal dynamic headroom) or the need to normalize these tests for volume (for SQ measurements within relative listening levels).
I give up!

Quote:
Originally Posted by mutescream View Post
Yes, but his test apparently proves that you can eq either a solid state amp or a tube amp to be indistinguishable from one another (to a listener), under his testing criteria.
Not really. His test criteria exclude any amps that are specifically designed to introduce heavy distortion like tube based guitar amps. Its really only a test of amplifers that are designed to be neutral, which is only a subset of tube amplifiers.

I would think it would be obvious that anyone could ABX certain guitar amps for instance due to the massive distortion they intentionally build into them.

Quote:
Originally Posted by mutescream View Post
This alone should bring into question the validity of his chosen parameters.
Honestly, if you think you've come up with some trivial observation that would undermine everything, and that somehow no one over the year has thought of it, its probably a good sign that you should reread the text because you're misunderstanding things. People aren't so stupid as you seem to think.

Quote:
Originally Posted by mutescream View Post
Only if you ignore that audio performance is dependent on your expectation.
What is this supposed to mean with respect to a double blind test?

Quote:
Originally Posted by mutescream View Post
Were performance not related to audibility,
Its not.

Quote:
Originally Posted by mutescream View Post
then impedance mismatches would be a non-issue, and available power in relation to efficiency would be a non issue.
????????

Seriously, I can't believe waded through all that bullshit. How does one post so much while saying so little.
__________________
Interested in Google's Summer of Code ? PM me.
Reply With Quote

  #25  
Old 05-07-2012, 04:08 PM
saratoga saratoga is offline
Rockbox Developer / Moderator
 
Join Date: Apr 2007
Posts: 3,627
Default

Quote:
Originally Posted by mutescream View Post
Not at all. I am an experienced and trained electronics technician. You apparently are not. I do not claim to have superior knowledge of whatever your chosen professional field is, or would I downgrade you for speaking more confidently when discussing it than I likely could.
I don't know what a "electronics technician" does, but I don't believe you have any particular knowledge of electronics. Pretty much every electrical thing you've said has been wrong. I don't think you're fooling anyone else either, so maybe you shouldn't be appealing to your own authority.
__________________
Interested in Google's Summer of Code ? PM me.
Reply With Quote

  #26  
Old 05-07-2012, 04:32 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

Quote:
Originally Posted by saratoga View Post
If you think its the first law of thermodynamics, then you're completely full of crap.
The First Law of Thermodynamics (Conservation) states that energy is always conserved, it cannot be created or destroyed. In essence, energy can be converted from one form into another.
http://www.emc.maricopa.edu/faculty/...bookener1.html
In a series circuit (such as a wire in line between a speaker and an amp) the wire provides a resistive load. Which is a voltage drop. What do you think happens with that voltage? Oh, crap... It's converted to heat.

Quote:
So I just reread it, and I think it makes no mention at all of this:
The wire test states that no one can tell the difference between "fancy" speaker wire and 16gau ripcord. It makes this as a blanket statement, and does not allow for limitations of applicability.

Quote:
Yes I think it does. If you're going to tell me that something isn't vanishingly small, I expect you to show that it is not. Do so now.
First explain the relevance, when refuting the universal applicability of a blanket statement. I will concede that in the vast majority of listening rooms, it will not make an audible difference. I am not arguing that. I am arguing that there are situations where it is not universally applicable.

Quote:
Translation: If things had been more different the test would have been easier. I think I agree.
Correction, "without things being as they are, you would be correct".

Quote:
Translation: meaningless
No less so than your apparent claim that equalizers cannot attenuate harmonics, that they are somehow immune to an equalizer's effects.

Quote:
Translation: no idea.

I give up!
Ad hominems, to fill in for your lack of a reasonable rebuttal?

Quote:
Not really. His test criteria exclude any amps that are specifically designed to introduce heavy distortion like tube based guitar amps. Its really only a test of amplifers that are designed to be neutral, which is only a subset of tube amplifiers.
Please show that this was the case.

Quote:
Honestly, if you think you've come up with some trivial observation that would undermine everything, and that somehow no one over the year has thought of it, its probably a good sign that you should reread the text because you're misunderstanding things. People aren't so stupid as you seem to think.
Look at the mediocrity that people settle for on a daily basis, before you assert that claim in the future. We wouldn't have the horrendous state of politics that we do, and the iPod would not be as ubiquitous as it is, if people were not that gullible.

Quote:
What is this supposed to mean with respect to a double blind test?
What it means, is that not all amplifiers can perform the same audibly, because they have differing operational capabilities. They can only be equal, if the superior one is hobbled sufficiently.

Quote:
Its not.
Really? Then why were amps not allowed to reach the clipping range? Oh, because some amplifiers can outperform others.

Quote:
Seriously, I can't believe waded through all that bullshit. How does one post so much while saying so little.
The feeling is mutual, here.
Reply With Quote

  #27  
Old 05-07-2012, 04:34 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

Quote:
Originally Posted by saratoga View Post
I don't know what a "electronics technician" does, but I don't believe you have any particular knowledge of electronics. Pretty much every electrical thing you've said has been wrong. I don't think you're fooling anyone else either, so maybe you shouldn't be appealing to your own authority.
Prove that anything that I have said about electricity is factually incorrect, then.
Reply With Quote

  #28  
Old 05-07-2012, 05:05 PM
saratoga saratoga is offline
Rockbox Developer / Moderator
 
Join Date: Apr 2007
Posts: 3,627
Default

Quote:
Originally Posted by mutescream View Post
The First Law of Thermodynamics (Conservation) states that energy is always conserved, it cannot be created or destroyed. In essence, energy can be converted from one form into another.
http://www.emc.maricopa.edu/faculty/...bookener1.html
In a series circuit (such as a wire in line between a speaker and an amp) the wire provides a resistive load. Which is a voltage drop. What do you think happens with that voltage? Oh, crap... It's converted to heat.
Ok then, use the 1st law to compute the resistive losses given the numbers you've posted above. When you do so, please think about why maybe I would think that was an odd choice, verses, say Kirchoff's law

Quote:
Originally Posted by mutescream View Post
The wire test states that no one can tell the difference between "fancy" speaker wire and 16gau ripcord. It makes this as a blanket statement, and does not allow for limitations of applicability.
As far as I can tell this is just a flat out fabrication. On the outside chance that I'm wrong, please quote the portion of the article claiming this.

Quote:
Originally Posted by mutescream View Post
First explain the relevance, when refuting the universal applicability of a blanket statement. I will concede that in the vast majority of listening rooms, it will not make an audible difference. I am not arguing that. I am arguing that there are situations where it is not universally applicable.
My essential point here is that its a retarded argument. I'm encouraging you to actually crunch the numbers because:

1) It seems unlikely that you would have even brought this up if you knew what you were talking about. Therefore I'm pretty sure you're bullshitting here and have no idea how to do it.
2) I want you to see how stupid this argument is

I think this is pretty reasonable. After all, its your claim. Do the work.

Quote:
Originally Posted by mutescream View Post
No less so than your apparent claim that equalizers cannot attenuate harmonics, that they are somehow immune to an equalizer's effects.
So I don't know how to respond to this because:

1) Its not even really a coherent idea, its just a babble of words
2) Its obviously not something I would have claimed, since the things I claim are ideas
3) I can't tell if you're just trolling me, or actually this intellectually dishonest.

Quote:
Originally Posted by mutescream View Post
Ad hominems, to fill in for your lack of a reasonable rebuttal?
An ad hom is actually when you try to discredit an argument by attacking a person. If I discredit an argument by pointing out that its not a grammatically tractable construct, that is not an ad hom.


Quote:
Originally Posted by mutescream View Post
Please show that this was the case.
If you read through the test criteria they disregard any equipment with significant artifacts, distortion, etc. They would have to. Otherwise someone could just show up with a busted amp and walk away with the prize money.

Quote:
Originally Posted by mutescream View Post
Look at the mediocrity that people settle for on a daily basis, before you assert that claim in the future. We wouldn't have the horrendous state of politics that we do, and the iPod would not be as ubiquitous as it is, if people were not that gullible.
How does other's mediocrity excuse your own laziness?

Quote:
Originally Posted by mutescream View Post
What it means, is that not all amplifiers can perform the same audibly, because they have differing operational capabilities. They can only be equal, if the superior one is hobbled sufficiently.
Great! Thats a real claim, something that could be debated and discussed. Now provide test results showing that to be the case and we'll do so.


Quote:
Originally Posted by mutescream View Post
The feeling is mutual, here.


Clearly its not since you didn't even bother replying to my post explaining your misconceptions.

Quote:
Originally Posted by mutescream View Post
Prove that anything that I have said about electricity is factually incorrect, then.
Sure, this is wrong:

Quote:
Originally Posted by mutescream View Post
Quote:
Originally Posted by saratoga
It is not what you were saying. You brought up voltage, it is current that matters. You are now talking about capacitance, when it is conductance that matters.
Reread, I bought up BOTH current and voltage. Voltage matters, because it relates to how much heat will be generated on the wire (and impact the resistance of the wire overall).
You claimed that the current as well as the 103 VAC open circuit voltage "mattered" in order to calculate the resistive losses. In fact, the current alone is sufficient for a given piece of wire:

P = I^2*R_wire

(note that I can solve for the losses per foot that without knowing the open circuit voltage nor the length of the wire, and given the open circuit voltage I cannot solve that equation without knowing more information, specifically the load impedance and the total length of wire. Thus, the open circuit voltage is both irrelevant and insufficient)

Its a minor mistake, but its the kind of thing most people who do this stuff realize without even thinking. At least when you mentioned the total circuit voltage the first thing I thought was "who cares, its the current that determines how hot a wire gets". At the time I assumed you just misspoke, now I see that you don't really understand what the difference is. Which makes it kind of galling to hear you talking down to other people about electronics.
__________________
Interested in Google's Summer of Code ? PM me.
Reply With Quote

  #29  
Old 05-07-2012, 05:08 PM
The DarkSide's Avatar
The DarkSide The DarkSide is offline
Ultra Senior Member
 
Join Date: Feb 2008
Location: Qns Vlg, NYC
Posts: 16,054
Default

I have a totally better idea,...clean up this thread by moving all this BS to a separate thread to keep the topic correct. Then, put a certain member on "ignore" and be happy with the evening!

Last edited by The DarkSide; 05-08-2012 at 01:56 PM.
Reply With Quote

  #30  
Old 05-07-2012, 05:09 PM
saratoga saratoga is offline
Rockbox Developer / Moderator
 
Join Date: Apr 2007
Posts: 3,627
Default

Quote:
Originally Posted by The DarkSide View Post
I have a totally better idea,...clean up this thread by moving all this BS to a separate thread to keep the topic correct. Then, put a certain member on "ignore" and be happy with the evening!
I'll probably just delete this crap once its over. Until then, its probably easier for everyone if we keep it in the same thread.
__________________
Interested in Google's Summer of Code ? PM me.
Reply With Quote

  #31  
Old 05-07-2012, 05:12 PM
The DarkSide's Avatar
The DarkSide The DarkSide is offline
Ultra Senior Member
 
Join Date: Feb 2008
Location: Qns Vlg, NYC
Posts: 16,054
Default

Quote:
Originally Posted by saratoga View Post
I'll probably just delete this crap once its over. Until then, its probably easier for everyone if we keep it in the same thread.
Yeah, I guess your correct. I'm just getting a kick out of all the wordplay. I make it a point to be nontechnical, not confusing. I hate when a person tries to impress via techspeak,...it's funny.
Reply With Quote

  #32  
Old 05-07-2012, 05:53 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

Quote:
Originally Posted by saratoga View Post
Ok then, use the 1st law to compute the resistive losses given the numbers you've posted above. When you do so, please think about why maybe I would think that was an odd choice, verses, say Kirchoff's law
Because you seemed unfamiliar with the concept entirely, and I wanted to pick one that would be more universally accessible? FFS, the laws of thermodynamics have even been discussed on the Simpsons.



Quote:
As far as I can tell this is just a flat out fabrication. On the outside chance that I'm wrong, please quote the portion of the article claiming this.


Conversely, please show me anywhere in that article that they admitted a limitation in applicability.

Quote:
My essential point here is that its a retarded argument. I'm encouraging you to actually crunch the numbers because:

1) It seems unlikely that you would have even brought this up if you knew what you were talking about. Therefore I'm pretty sure you're bullshitting here and have no idea how to do it.
2) I want you to see how stupid this argument is

I think this is pretty reasonable. After all, its your claim. Do the work.
You claim here that I don't know what I am talking about, prove it. I've conceded that from the beginning that this would only be relevant for large signals being sent through rather long wire runs. I've also stated that this is directly proportional to the length of the wire for applicability.

Quote:
So I don't know how to respond to this because:

1) Its not even really a coherent idea, its just a babble of words
2) Its obviously not something I would have claimed, since the things I claim are ideas
3) I can't tell if you're just trolling me, or actually this intellectually dishonest.
You are the one that has claimed that an equalizer cannot compensate for harmonic differences between amplifiers, not I. I am calling you on your intellectual dishonesty here.

Quote:
An ad hom is actually when you try to discredit an argument by attacking a person. If I discredit an argument by pointing out that its not a grammatically tractable construct, that is not an ad hom.
Actually, I admit to being slightly off there on the logical fallacy I was calling you on... It was an indirect ad hom, via an appeal to ridicule, which is something you have been increasingly relying upon.

Quote:
If you read through the test criteria they disregard any equipment with significant artifacts, distortion, etc. They would have to. Otherwise someone could just show up with a busted amp and walk away with the prize money.
No problem with them rejecting equipment not having a reasonable bandwidth of usability, and being fully functional. But, that does not explain the equalization.

Quote:
How does other's mediocrity excuse your own laziness?
First demonstrate that I was lazy. Second, demonstrate (if you can) that it is relevant to my observation that people by and large are both gullible and tolerate mediocrity the vast majority of the time.

Quote:
Great! Thats a real claim, something that could be debated and discussed. Now provide test results showing that to be the case and we'll do so.
Seriously? You are going to dispute that an audio device pushed to the point of failure will exhibit audible results of that failure?

Quote:
Clearly its not since you didn't even bother replying to my post explaining your misconceptions.
I didn't see enough in that post that was independent of your previous posts that I disputed.

Quote:
You claimed that the current as well as the 103 VAC open circuit voltage "mattered" in order to calculate the resistive losses. In fact, the current alone is sufficient for a given piece of wire:
Not really... Voltage is what is dropped on resistive loads in a series circuit. Current is only dropped in parallel circuits.

Quote:
P = I^2*R_wire
Cool! You at least understand the first step in how I determined the voltage and current from the power and resisitive load.

Quote:
(note that I can solve for the losses per foot that without knowing the open circuit voltage nor the length of the wire, and given the open circuit voltage I cannot solve that equation without knowing more information, specifically the load impedance and the total length of wire. Thus, the open circuit voltage is both irrelevant and insufficient)

Its a minor mistake, but its the kind of thing most people who do this stuff realize without even thinking. At least when you mentioned the total circuit voltage the first thing I thought was "who cares, its the current that determines how hot a wire gets". At the time I assumed you just misspoke, now I see that you don't really understand what the difference is. Which makes it kind of galling to hear you talking down to other people about electronics.
Since I knew (and posted) what the load on the circuit was (5 ohm), the power on that circuit (1KW), I extrapolated the voltage and current from the formula that you cited earlier.

See above. Resistance in a series circuit causes voltage drops, it is that voltage drop that is converted into heat. Current is constant in a series circuit. That is why wire in a series circuit without a resistive load on it (shorted) will heat up faster (since it is carrying the entire load).

The higher the proportion of the total resistive load (between the wire and the speaker) that the wire carries (and subsequently the amount of voltage dropped on it), the greater level of heat will be generated (by the voltage being converted to heat).

Do you still wish to continue this ludicrous position of yours?
Reply With Quote

  #33  
Old 05-07-2012, 06:01 PM
jupitreas's Avatar
jupitreas jupitreas is offline
Member
 
Join Date: Dec 2009
Location: Warsaw, Poland
Posts: 380
Default

Don't delete anything please! This is such a fun thread and mutescream is truly one of the most amazing trolls I've ever seen.

Also, there is lots of great information in saratoga's posts that I believe can clarify a lot of things for a lot of people who are not as knowledgeable as he is in this field. I certainly enjoyed reading this whole exchange and would love to see it preserved for posterity's sake
Reply With Quote

  #34  
Old 05-07-2012, 07:23 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

Realistically, for the resistive load of a 16gau wire to exceed 5% of a 5ohm load of a Magnepan wall speaker is going to be about 30'. At this point it will have audible effects. Which will become an issue prior to heat becoming an issue.

http://en.wikipedia.org/wiki/Speaker_wire#Resistance

Last edited by mutescream; 05-07-2012 at 07:36 PM.
Reply With Quote

  #35  
Old 05-07-2012, 08:20 PM
saratoga saratoga is offline
Rockbox Developer / Moderator
 
Join Date: Apr 2007
Posts: 3,627
Default

Quote:
Originally Posted by mutescream View Post
Because you seemed unfamiliar with the concept entirely, and I wanted to pick one that would be more universally accessible? FFS, the laws of thermodynamics have even been discussed on the Simpsons.
So you picked a dumb way to explain it because you thought I didn't understand it? How does that even make sense? Were you just trying to be ass?

Quote:
Originally Posted by mutescream View Post
The wire test states that no one can tell the difference between "fancy" speaker wire and 16gau ripcord.


So basically, that doesn't say at all what you said it did. Would it be fair to state that you made up that accusation unfairly against the author?

Quote:
Originally Posted by mutescream View Post
Conversely, please show me anywhere in that article that they admitted a limitation in applicability.
I don't think I ever claimed there was one, but just skimming it now the author explicitly states in the last paragraph that the results are not universally generalizable. Check the paragraph beginning "Can these results be extrapolated to interconnects.."

Anyway, since it seems like you really jumped on this guy (and several other people in this thread) without taking the time to hear him out, I think an apology is in order.


Quote:
Originally Posted by mutescream View Post
You claim here that I don't know what I am talking about, prove it. I've conceded that from the beginning that this would only be relevant for large signals being sent through rather long wire runs. I've also stated that this is directly proportional to the length of the wire for applicability.
How long? How much current? Length of a room, length of a state? Diameter of the Earth?

Quote:
Originally Posted by mutescream View Post
You are the one that has claimed that an equalizer cannot compensate for harmonic differences between amplifiers, not I.
Yes I absolutely am. There is no way to compensate for harmonic distortion with an EQ. This is trivially provable using linear systems theory.

I didn't even realize you disagreed?


Quote:
Originally Posted by mutescream View Post
Actually, I admit to being slightly off there on the logical fallacy I was calling you on... It was an indirect ad hom, via an appeal to ridicule, which is something you have been increasingly relying upon.
It was simply you being dishonest. Don't try to dress it up.

Quote:
Originally Posted by mutescream View Post
No problem with them rejecting equipment not having a reasonable bandwidth of usability, and being fully functional. But, that does not explain the equalization.
See my post above about equalization. Its very important you understand exactly what it does and does not do. Given your comments about harmonic distortion, its clear that you're not quite up to speed with the terminology, and if you don't understand that you will not be able to discuss this with the rest of us (who are quite familiar with the working of EQ).

Quote:
Originally Posted by mutescream View Post
First demonstrate that I was lazy. Second, demonstrate (if you can) that it is relevant to my observation that people by and large are both gullible and tolerate mediocrity the vast majority of the time.
You wouldn't consider jumping to the first easy assumption rather then taking the time to understand a concept lazy? I would. Its extremely lazy.

Quote:
Originally Posted by mutescream View Post
Seriously? You are going to dispute that an audio device pushed to the point of failure will exhibit audible results of that failure?
Perhaps I didn't understand what you were trying to say, but no I don't agree that all devices have equal capacity. I do disagree though that you can somehow equate capacity and audibility.


Quote:
Originally Posted by mutescream View Post
I didn't see enough in that post that was independent of your previous posts that I disputed.
You've been repeating the same misconceptions since the beginning and I've taken the time to try and help you. I expect you to at least take the time to try and learn this stuff. If not, just stop reading and go outside. If you're not interested in learning about audio, you're just wasting your time.

Quote:
Originally Posted by mutescream View Post
Cool! You at least understand the first step in how I determined the voltage and current from the power and resisitive load.
That is actually the only step.

Quote:
Originally Posted by mutescream View Post
Since I knew (and posted) what the load on the circuit was (5 ohm), the power on that circuit (1KW), I extrapolated the voltage and current from the formula that you cited earlier.

See above. Resistance in a series circuit causes voltage drops, it is that voltage drop that is converted into heat. Current is constant in a series circuit. That is why wire in a series circuit without a resistive load on it (shorted) will heat up faster (since it is carrying the entire load).
That actually isn't quite right either:

1kW at the load. 5 ohms. 72V rms. 14a.

Ptotal = Pline+Pload = 72v*14 = 1008w ~~ 1000w.

Pload = 1000w.

Pline = 0.

Pline = Vline*Iline.

Iline = 14a.

Vline = 0.

So we get that there is no voltage drop across the line, which makes sense since above you assumed that it had infinite conductivity. Obviously this is wrong.

The correct way to solve this problem is to take the impedance of the load, the power of the load, and solve for the voltage across it. Then calculate the impedance of the line from its conductance (and if you insist you may integrate its capacitance and inductance but it won't actually matter). Then finally calculate the voltage drop given the impedance.


Quote:
Originally Posted by mutescream View Post
Do you still wish to continue this ludicrous position of yours?
Sure, I'll help you with this until you understand it.
__________________
Interested in Google's Summer of Code ? PM me.
Reply With Quote

  #36  
Old 05-07-2012, 08:26 PM
Enigmatic Enigmatic is offline
Senior Member
 
Join Date: Mar 2008
Posts: 731
Default

Quote:
Originally Posted by mutescream View Post
If you limit the premium product to only being tested within the parameters of the non-premium product, you are most definitely NOT demonstrating that they are equal quality products.
High-quality amplifiers are not being tested within the parameters of low-quality amplifiers. Whenever a high-quality amplifier is ABXed against another high-quality amplifier, no one has been able to hear a difference. But to expand the pool of amplifiers, Clark also allows low-quality amplifiers to also be ABXed if their poor frequency response can be corrected using equalisation.
Quote:
BUT, not all amps will have identical sonic signatures in all situations, such as situations that involve differing resistive loads of various speakers (or headphones) or at increased output levels.
We are talking here about loudspeakers only and non-clipping amplifiers.
Quote:
When you limit the test to only testing the ability of two audio products to the parameters of the weakest link, and have tampered with them to ensure they have identical sonic signatures, of course the outcome is predetermined.
There is no weakest link. A good amplifier is a good amplifier is a good amplifier. It could be $30 000-US monoblocks. It could be a $300 receiver.
Quote:
As for all quality amps having perfect linearity and low distortion, consider the tube amp.
Tube amplifiers can be high quality or low quality.
Quote:
It colors the signal pretty heavily with even order harmonics (which creates a warmer sound).
Not all tube amplifiers deviate from neutral.
Quote:
This is even ignoring the distinct possibility that a particular amp and speaker combo may have a closer sonic signature to what a room requires than another combo, and subsequently require less equalization to achieve the proper corrected sound (the fewer components and connections you introduce, the less distortion).
You do not understand why Clark is allowed to use equalisation. Please see above.
Reply With Quote

  #37  
Old 05-07-2012, 09:39 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

Quote:
Originally Posted by saratoga View Post
So you picked a dumb way to explain it because you thought I didn't understand it? How does that even make sense? Were you just trying to be ass?
No, I was trying to keep it in terms that I hoped were easily accessible, in the hopes of expediting this conversation.

Quote:


So basically, that doesn't say at all what you said it did. Would it be fair to state that you made up that accusation unfairly against the author?
No.

Quote:
I don't think I ever claimed there was one, but just skimming it now the author explicitly states in the last paragraph that the results are not universally generalizable. Check the paragraph beginning "Can these results be extrapolated to interconnects.."
That only indicates that he doesn't believe the results could be honestly used to establish that differences between interconnects are negligible. He never states that his results are not universally applicable to speaker wire.

Quote:
Anyway, since it seems like you really jumped on this guy (and several other people in this thread) without taking the time to hear him out, I think an apology is in order.
I think I have been fairly good natured about it, considering the OP didn't take the time to respond, and a moderator came in to try to intimidate me.

Quote:
How long? How much current? Length of a room, length of a state? Diameter of the Earth?
In a post that was previous to the one I am replying to, I cited a reference that states that when a wire exceeds 5% of the speaker's impedance, it creates audible changes. In the instance of a 5 ohm speaker, and 16gau wire, this is actually only 30'.

Quote:
Yes I absolutely am. There is no way to compensate for harmonic distortion with an EQ. This is trivially provable using linear systems theory.
But deviations in linearity of signal can be attenuated.

Quote:
It was simply you being dishonest. Don't try to dress it up.
Hey, you were playing dirty and don't want to take responsibility for your actions. Don't try to pretend it was something ti wasn't.

Quote:
See my post above about equalization. Its very important you understand exactly what it does and does not do. Given your comments about harmonic distortion, its clear that you're not quite up to speed with the terminology, and if you don't understand that you will not be able to discuss this with the rest of us (who are quite familiar with the working of EQ).
So, if you think I am incorrect, enlighten me... by all means.

Quote:
You wouldn't consider jumping to the first easy assumption rather then taking the time to understand a concept lazy? I would. Its extremely lazy.
There has been enough going in this conversation, that out of context, I am no longer sure WTF is being referred to here. The only laziness I will admit to, is being too damned lazy to go back and figure out what we are arguing about in this. It's kind of pointless, really.

Quote:
Perhaps I didn't understand what you were trying to say, but no I don't agree that all devices have equal capacity. I do disagree though that you can somehow equate capacity and audibility.
Good. I don't quite quite equate them, though. Rather, the limits of a device is defined by its capacity to perform, inherently. Those limits in turn define what it is able to do audibly.

Quote:
That actually isn't quite right either:

1kW at the load. 5 ohms. 72V rms. 14a.

Ptotal = Pline+Pload = 72v*14 = 1008w ~~ 1000w.

Pload = 1000w.

Pline = 0.

Pline = Vline*Iline.

Iline = 14a.

Vline = 0.

So we get that there is no voltage drop across the line, which makes sense since above you assumed that it had infinite conductivity. Obviously this is wrong.
Actually, I assumed the length of the line was 0, initially... Since we hadn't established wire lengths.

Quote:
The correct way to solve this problem is to take the impedance of the load, the power of the load, and solve for the voltage across it. Then calculate the impedance of the line from its conductance (and if you insist you may integrate its capacitance and inductance but it won't actually matter). Then finally calculate the voltage drop given the impedance.
Amazingly enough, it finally sounds like we can agree on something.

*edit* Although, it would probably be easier to just inject a reference signal through the circuit, and measure the loss.

Quote:
Sure, I'll help you with this until you understand it.
I'm willing to agree with you, when you say something correct (see above).

Last edited by mutescream; 05-07-2012 at 10:19 PM. Reason: typo
Reply With Quote

  #38  
Old 05-07-2012, 09:50 PM
mutescream mutescream is offline
Banned
 
Join Date: May 2012
Posts: 56
Default

Quote:
Originally Posted by Enigmatic View Post
High-quality amplifiers are not being tested within the parameters of low-quality amplifiers. Whenever a high-quality amplifier is ABXed against another high-quality amplifier, no one has been able to hear a difference. But to expand the pool of amplifiers, Clark also allows low-quality amplifiers to also be ABXed if their poor frequency response can be corrected using equalisation.
Actually, he allowed the testee (not to be confused with teste) to determine which amplifier was equalized to match the other. They (both amplifiers) are also being limited to the performance limitations of the weaker amplifier (otherwise they could push the weaker amplifier to clip earlier than the stronger one).

Quote:
We are talking here about loudspeakers only and non-clipping amplifiers.There is no weakest link.
The amplifiers were tested below the clipping point of the weakest amplifier. It's stipulated in the test.

Quote:
A good amplifier is a good amplifier is a good amplifier. It could be $30 000-US monoblocks. It could be a $300 receiver.
The problem is that the $300 receiver will not have as strong of an amplifier as a "decent" stand alone separates amp. Which in turn will not perform as well as a pair of monoblocks.

Quote:
Tube amplifiers can be high quality or low quality.Not all tube amplifiers deviate from neutral.You do not understand why Clark is allowed to use equalisation. Please see above.
I suppose a tube amplifier could use negative feedback, and cancel the even order harmonics out... The question would be why do that in the first place? It kind of defeats the purpose of using a tube amplifier over a solid state.
Reply With Quote

  #39  
Old 05-07-2012, 10:25 PM
saratoga saratoga is offline
Rockbox Developer / Moderator
 
Join Date: Apr 2007
Posts: 3,627
Default

Quote:
Originally Posted by mutescream View Post
No.
Yes actually. This is simply a fabrication:

Quote:
Originally Posted by mutescream View Post
The wire test states that no one can tell the difference between "fancy" speaker wire and 16gau ripcord.
And quite a few other statements. You have some corrections to make in you previous posts as you have made some really unfair statements about other people. The correct thing to do is retract them. The not correct but acceptable thing to do is to stop talking about it. Pick one that suits your character.

Quote:
Originally Posted by mutescream View Post
Conversely, please show me anywhere in that article that they admitted a limitation in applicability.
Quote:
Originally Posted by mutescream View Post
That only indicates that he doesn't believe the results could be honestly used to establish that differences between interconnects are negligible.
Ugh this level of dishonesty on your part is ridiculous. You made a mistake, either own up or shut up. Don't bullshit the rest of us. We have better things to do then this.

Quote:
Originally Posted by mutescream View Post
In a post that was previous to the one I am replying to, I cited a reference that states that when a wire exceeds 5% of the speaker's impedance, it creates audible changes. In the instance of a 5 ohm speaker, and 16gau wire, this is actually only 30'.
Aren't you mixing this up? I thought you were talking about capacitance, but the 5% rule assumes zero capacitance. That said, I agree completely, capacitance is negligible in this situation. But are you sure thats what you mean?

Quote:
Originally Posted by mutescream View Post
But deviations in linearity of signal can be attenuated.
Actually, no they cannot be. I'll try this intuitively:

Basically, an ideal amp should do this:

out(t) = in(t)*gain

But all real amps actually do this:

out(t) = [in(t)^n]*gain for n != 1

(technically this isn't exactly true since the function will be much uglier, but if you're mathematically inclined just picture a Taylor series expansion and the math will work just as well as what I am assuming)

This is a problem since the nonlinear term (the power of n) will result in a redistribution of energy across frequencies. Basically, if you put a pure tone in, you will get several (impure) tones out.

For a concrete (if trivial example), think cos(2000*2pi*t) for n =2. Freed that through and you'll get cos(2000*2pi*t)^2 = 0.5*[1+ cos(4000*2pi*t)]. Basically, you'll lose the original frequency entirely, and get a new one back. Obviously no EQ can fix that!

But what can an EQ do?

Now with EQ, what you do is assume that n==1, and then do something like this:

out= [in(t) X EQ_impulse]*gain where 'X' is a convolution operation.

Which will work fine in that case since n==1. But for n != 1, then:

out= [[in(t) X EQ_impulse]^n]*gain where n != 1.

Now this is a huge problem, since there is nothing you can convolve in(t) with that will cancel the power of n outside the convolution. (if you aren't sure why, just think of convolution as a special type of multiplication, then the result is the same, you can't multiply a linear function to cancel out a parabolic function for instance)

So thats really the core problem here: EQ is just a scaling of the linear term of an equation. It cannot help you with the nonlinear part (the terms generated when exponent is much different then 1). So while you can try and 'fix' things by picking an EQ impulse function that will give you a good sound, you'll actually be unable to correct for the nonlinear part, and in fact if its non-negligible you'll probably make the amp even less accurate. This is why doing a test with an EQ is valid: it helps reveal fundamental limitations in the amplifier.

And its really this nonlinear term is the fundamental limitation that we care about. When you talk about the capacity or load of an amp, you're really talking about the magnitude of that nonlinear term when presented a given load. If its small, we say that the amp isn't overloaded. If its large, we say that the amp is probably overloaded.

So, does that make sense to you?
__________________
Interested in Google's Summer of Code ? PM me.
Reply With Quote

  #40  
Old 05-07-2012, 10:27 PM
Enigmatic Enigmatic is offline
Senior Member
 
Join Date: Mar 2008
Posts: 731
Default

Quote:
Originally Posted by mutescream View Post
Actually, he allowed the testee (not to be confused with teste) to determine which amplifier was equalized to match the other.
Only if one of the amplifiers is of low quality.
Quote:
They (both amplifiers) are also being limited to the performance limitations of the weaker amplifier (otherwise they could push the weaker amplifier to clip earlier than the stronger one).
What is the point of ABXing a clipping amplifier against a non-clipping amplifier?
Quote:
The amplifiers were tested below the clipping point of the weakest amplifier. It's stipulated in the test.
What is the point of ABXing a clipping amplifier against a non-clipping amplifier?
Quote:
The problem is that the $300 receiver will not have as strong of an amplifier as a "decent" stand alone separates amp. Which in turn will not perform as well as a pair of monoblocks.
What is the point of ABXing a clipping amplifier against a non-clipping amplifier?
Quote:
I suppose a tube amplifier could use negative feedback, and cancel the even order harmonics out... The question would be why do that in the first place? It kind of defeats the purpose of using a tube amplifier over a solid state.
This was covered in the first post of this thread in a link.
Quote:
Originally Posted by mutescream View Post
I think I have been fairly good natured about it, considering the OP didn't take the time to respond, and a moderator came in to try to intimidate me.
I did not take the time to respond to what specifically? I stand by everything I wrote in the first post of this thread.
Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump



All times are GMT -5. The time now is 05:13 AM.