-
What makes a good Power Amplifier?
Wattage?
Class?
Ohms?
Resistors?
Capacitors?
Type?
Power transformer?
Bill
-
-
A good power amplifier always starts out with a good power supply. One that can easily supply all of the current and voltage necessary to deliver the rated power to the load, one that is fast and quickly recovers from momentary overloads, one that is stable and relatively immune to changes in line voltage including noise, and one that comes as close as possible to the ideal of a battery. No matter what type of amplifier you like or use, it cannot be better than its power supply. That is the heart and soul of every one of them.
-
I'd say what makes a good power amp is one that works when the switch is in the ON position.
-
That only makes one that functions, not necessarily one that performs well.
-
Quote:
Originally Posted by BillB
Wattage?
Class?
Ohms?
Resistors?
Capacitors?
Type?
Power transformer?
Bill
The answer to this depends on how you define "good". To me, a good power amp delivers lots of amps for not much money (<$500) money. I have a heavy amp, but I would prefer a small light one if it could perform similarly.
Obviously, we want distortion to be below audible levels into various impedances (or at least into my speakers' impedance) and at all frequencies in the audible band (20Hz-20KHz).
Good binding posts for speaker wire would be nice.
Who cares what kind of capacitors they use? I'd like my amp to last a long time, so I want some build quality. Such quality might get me to cough up more money, but it depends on how MUCH more money. For the price of some expensive amps, I could buy a new reasonably priced one every few years and still save money.
One must always justify why it wouldn't be better to buy the cheaper amp and spend the extra money on better speakers.
-
listen..listen.....and then decide
I would like to add something to what skeptic said.
1.Whatever he mentioned is the ideal for a power amp. But you need to know the fact that you will be looking at the spec sheet to check this out. But from my experience, you should listen and find if the spec sheet is lying. This is the case with most mass brand amp manufacturerers. I can tell you lots of stories of how the customer has been taken for a ride.Under ideal conditions (this condition exists only in utopia), all amps with the same specs should behave the same and show similar tonal characteristics. But nothing is further from the truth.LISTEN, LISTEN... and decided on what you like.
2.You need to mate the right amp with your speakers. A bright sounding mega watt power amp with a pair of klipcsh(horn loaded high sensitivity speakers) will give you ear fatigue. The same amp will work well with low sensitivity,laid back sound speakers (example - dynaudio).
3.One good way to avoid bad amps is to avoid most of the neon bulb brands. Have fun.
-
Quote:
Originally Posted by BillB
Wattage?
Class?
Ohms?
Resistors?
Capacitors?
Type?
Power transformer?
Bill
You asking the wrong questions...Just listen to some amps that you think are the one. Then make your choice of amps....believe your ears...thats the most important thing in our hobby. ;)
Greetz MF
-
BillB
Unless you are using Magnepan speakers, I would not worry about this very much beyond the "good power supply design" comments by Skeptic combined with adequate power at reasonably low distortion levels... which are easily found in many, many amplifiers today.
I have found that cone speakers and electrostatic speakers sound the same when used with ANY reasonable choice of amplifier. The only exceptions to my ear are Magnepan speakers whose sound can be greatly altered by your choice of amp. But even tube amps for use with Magnepans should be selected with care. I never cared for Audio Research amps with Tympani- the resulting sound was agressive and not at all natural for my ears.
Of course, if you do not religeously attend live recitals, you will never develop the "ear" that will allow you to obtain a realistic sound, and this will make your amp quest completely moot.
-
Quote:
Originally Posted by Mash
Unless you are using Magnepan speakers, I would not worry about this very much beyond the "good power supply design" comments by Skeptic combined with adequate power at reasonably low distortion levels... which are easily found in many, many amplifiers today.
Do tell me why you think this is so. They are nearly a purely resistive load.
Quote:
Originally Posted by Mash
I have found that cone speakers and electrostatic speakers sound the same when used with ANY reasonable choice of amplifier.
On which cone speakers and which electrostats is your opinion based?
rw
-
Electrostatic speakers have coupling transformers which consist of a pair of coupled wire coils. A coil of wire is an inductance, and an inductance is the equivalent of a mass in a spring-mass system, i.e. electrical inertia. The voice coil in a cone driver is likewise an inductance, and again an inductance is the equivalent of a mass in a spring-mass system, i.e. electrical inertia. Basic physics. This is why the people who conduct DBT on amplifiers always claim that all amps sound the same when those amps are used within their limitations for power and load handling capabilities....because invariably amp DBT were conducted with cone speakers. Try a DBT using Tympani or other Magnepan speakers and the conclusions will be different.
Remember that filters and cross-over networks are composed of inductors and capacitors.
True, Magnepan speakers are simple resistances, particularly when directly tri-amped, because their voice-wires are arranged into straight lines up and down the proprietary membrane drivers, i.e. there are no wire voice-coils to add inductance and so there is no inductance created. Furthermore, those proprietary Magnepan membrane drivers are designed (tuned, if you will- refer to a good book on frequencies & mode shapes; I can recommend one if you want to learn the subject) for specific response ranges rendering extensive LC cross-over filters unecessary. Cone speakers present a very different set of physics and so powerful LC cross-over circuits must be used to control and blend their output ranges.
You do not have to investigate every duck on earth to cofirm that ducks quack.
-
Quote:
Originally Posted by Mash
You do not have to investigate every duck on earth to cofirm that ducks quack.
No, but there is quite a variance among ducks. :)
So the presence or lack of inductance is a primary source for audible differences among amplifiers. Is that the sole determining factor?
rw
-
When you get beyond amplifier distortion products versus available output power versus needed power, load sensitivity, and so forth, the inducive and capacitive qualities of a type of loudspeaker will limit what you can hear from that speaker type. Some loudspeakers of a given type will be better than other loudspeakers of that given type, but observe that even the very best and most expensive auto on the market today makes a lousy boat. Note again that electrical filters are composed of LC elements.......
One can build cone drivers combining heavy ceramic magnets with a high magnetic density with a large voice coil intended to drive a well-stiffened, and therefor relatively more heavy, cone but at the same time voice-coil inductance has been increased. Not to mention that the speaker cone basket must be stiffened to keep the heavier magnet from vibrating at an audible frequency. The heavier magnet and basket structure means the speaker enclosure mounting board must be stiffer and thicker, and therefore heavier, to again keep the heavier cone driver assembly from vibrating at an audible frequency. This is a loosing battle in every sense of the word.
Remember that natural frequency equations can be complicated but they are all a variant of
natural frequency = (a constant) x Square Root (System Stiffness / Mass )
So if you double the stiffness without adding any mass, you raise frequency by 41%. Wow!!
But then you will have to double that stiffness without changing mass by using UnObtanium that has a Young's Modulus of 30,000,000,000,000 psi and a density of 5 nano-pounds per cubic yard.
One can consider some real-world structural metals:
Steel: Young's Modulus = 30,000,000 psi and density = 0.3 pounds/cu in.
Titanium: Young's Modulus = 16,000,000 psi and density = 0.16 pounds/cu in.
Aluminum: Young's Modulus = 10,000,000 psi and density = 0.1 pounds/cu in.
Notice anything??
-
Quote:
Originally Posted by Mash
When you get beyond amplifier distortion products versus available output power...
Have we?
Quote:
Originally Posted by Mash
One can consider some real-world structural metals:
Steel: Young's Modulus = 30,000,000 psi and density = 0.3 pounds/cu in.
Titanium: Young's Modulus = 16,000,000 psi and density = 0.16 pounds/cu in.
Aluminum: Young's Modulus = 10,000,000 psi and density = 0.1 pounds/cu in.
Notice anything??
Why don't you tell us.
rw
-
Ok, I've learned that all ducks quack and that a good car still won't float. Is it a witch? Burn it?
I'm not following. To paraphrase my impressions. Mash, I think your argument is that Amp output doesn't vary much from unit to unit, model to model, operating with in the specified range. And that what you hear is speaker dependent. I'm not sure i followed the speaker physics lesson except to note that surely what you hear must also depend as much on acoustical design and material in the speaker as on magnets and inductance. I don't think physics is the whole answer.
I have no idea what the real world structural metal example is supposed to indicate. I don't plan on using my speakers as a floatation device, just listening to them. It sounds like you describe that the speaker/amp pair set up an electrical oscillation at a frequency that may or may not be supported by the material components of the speaker. But, since most big, great speakers out there obviously aren't audibly humming (or people would not be buying them), I don't see where this point is going?
I could use an synopsis in english...
noddin0ff
-
Quote:
Originally Posted by noddin0ff
I don't think physics is the whole answer.
I could use an synopsis in english...
noddin0ff
The advice of "just listen" can't hurt, but understand that you would have to listen on the same speakers in the same acoustic space (ideally, your listening room). You also have to level match or you will likely prefer the amp that is playing louder.
As nobody (I have every heard about in over 20 years) has been able to show that they hear a difference, good luck. The bottom line is that you have to have some really abnormal (weird impedance) speakers, such as some electrostatics, for most amps to have a problem.
If physics (electronics and acoustics) isn't the whole answer, what is? The only logical response is psychology, which indicates that people will think they hear differences based on their expectations. This is a long established principle that is completely ignored by subjectivists as it iindicates that their keen perceptions are illusions.
-
The data on the three (most popular) structural metals was provided to allow you to notice that for all three metals the result of [Yonngs Modulus / Density] is the same. The structure's mass is volume x density, and the structure's stiffness is a direct function of Young's Modulus. Hence simply changing materials will not necessarily get you a change in natural frequency. Note that enclosure vibratory response can often be perceived as a reduction in clarity rather than being heard as "buzzing".
Every amp DBT I have ever read about used cone speakers for listening, and every time the conclusions, as I recall, were that the amps were DBT indistinguishable one from another. I believe the inductive electrical characteristic of the coned speakers used blurred the discinctions among the amps while the essentially resistive electrical characteristics of Magnepans would allow the discinctions among the amps to be audible. Its a consideration for those wishing to hold another amp DBT.
Almost all electrostatic speakers in existance have coupling transformers, and amps also used to have coupling transformers but coupling transformers were eliminated from amps for improved sonic performance. One exception I remember is a high-voltage electrostatic speaker featured in The Audio Amateur c1975 which was direct-connected to a dedicated high-voltage tube amp. As I recall the high-impedance membrane was charged to 3000 volts. This speaker was described as being a killer in every sense of the word.
-
Quote:
Originally Posted by RobotCzar
As nobody (I have every heard about in over 20 years) has been able to show that they hear a difference, good luck.
Tests are entirely valid for that which they test. For what they don't test, they stand mute.
Quote:
Originally Posted by RobotCzar
The bottom line is that you have to have some really abnormal (weird impedance) speakers, such as some electrostatics, for most amps to have a problem.
You presume that audible differences only occur upon distress. Differences in amplifier design, implementation and construction have zero effect whatsoever, right?
Quote:
Originally Posted by RobotCzar
If physics (electronics and acoustics) isn't the whole answer, what is?
You will only know that to be the case for when the testing encompasses all that which is available. I agree that there is little difference between a Buick and an Oldsmobile.
rw
-
Quote:
Originally Posted by Mash
Every amp DBT I have ever read about used cone speakers for listening, and every time the conclusions, as I recall, were that the amps were DBT indistinguishable one from another.
Do share with us the specific amplifiers included in the tests to which you refer. Please don't resort to the circular reasoning fallacy that such does not make any difference.
Quote:
Originally Posted by Mash
One exception I remember is a high-voltage electrostatic speaker featured in The Audio Amateur c1975 which was direct-connected to a dedicated high-voltage tube amp. As I recall the high-impedance membrane was charged to 3000 volts. This speaker was described as being a killer in every sense of the word.
You really do need to catch up on history a bit. The Acoustat X design, released in 1976 also ran high voltage tube amps directly to the plates of electrostatic panels at slightly over 3kV. Later, they replaced the tube amps with bi-formers increasing the drive voltage to around 5 kV. The current Sound Labs electrostats run the bias higher still. I seem to have lost your point with your dissertation about a product for which I have used for more than a quarter century.
rw
-
Other amp characteristics.
Quote:
Originally Posted by RobotCzar
The answer to this depends on how you define "good". To me, a good power amp delivers lots of amps for not much money (<$500) money. I have a heavy amp, but I would prefer a small light one if it could perform similarly.
Obviously, we want distortion to be below audible levels into various impedances (or at least into my speakers' impedance) and at all frequencies in the audible band (20Hz-20KHz).
Good binding posts for speaker wire would be nice.
Who cares what kind of capacitors they use? I'd like my amp to last a long time, so I want some build quality. Such quality might get me to cough up more money, but it depends on how MUCH more money. For the price of some expensive amps, I could buy a new reasonably priced one every few years and still save money.
One must always justify why it wouldn't be better to buy the cheaper amp and spend the extra money on better speakers.
In "How to Be a Sophisticated Audiophile and Resist Trendy Stupidities," in The Audio Critic, vol. 25, Winter 1998-99, p. 17, Peter Aczel gave the reasons why he thinks there is no reason why good mid-priced solid state preamps and amps should sound different from well designed higher priced stuff:
"The midpriced equipment also has high input impedance, low output impedance, flat frequency response, low distortion and low noise--and that is what we can hear. There is no such thing as an effect without a cause, and there's nothing to cause the high-end equipment to sound better."
Of course, The Audio Critic put amplifiers through a set of tests including dynamic power output down to 2 ohms into reactive loads. Some speakers are difficult to drive, and that can be significant under some circumstances.
Yeah, good binding posts are a definite plus--also on the speakers at the other end of the cables!
-
Quote:
Originally Posted by RobotCzar
If physics (electronics and acoustics) isn't the whole answer, what is? The only logical response is psychology, which indicates that people will think they hear differences based on their expectations. This is a long established principle that is completely ignored by subjectivists as it iindicates that their keen perceptions are illusions.
I guess I should have parsed my questions better. I would agree that physics probably describes very well how an amp performs. Signal in, signal out. Mash goes on about the 'physics' of speakers with mass, inductance, stiffness, density,coils, Young's modulus etc. Confusing. However, what he seems to be describing is the physics of energy transfer and of resonance. Which can be accurately measured but do not have a bearing on accoustics and the quality of sound, only the energy imparted to the air. I would parse 'acoustics' separately from 'physics' because there's no way his equations and a super computer could calculate the acoustics, and they certainly aren't amp dependant.
Certainly psychology has a role. And I am a great believer in physics. (I know pounds is not unit of mass, e.g.). But some things can be measured. Sometimes its difficult for the aspiring audio nut to learn how far numbers are meaning full. And at what point they are bullsh't. And, where in the field of audio equipement diminishing returns begins.
So, my question and confusion lies in which parts describe actually pertain to the amp and which to the speaker. I think judging the physical performance of an amp by the perceived sound of a speaker is not quite right, but I suppose if the speaker type is actually altering the physical performance of an amp, that would be interesting, and I'd like to try to understand what he's talking about. A lot of the jargon sounded like a description of speaker physics which doesn't strike me as so critical for amp performance.
Am I off?
noddin0ff
-
Quote:
Originally Posted by Mash
The data on the three (most popular) structural metals was provided to allow you to notice that for all three metals the result of [Yonngs Modulus / Density] is the same. The structure's mass is volume x density, and the structure's stiffness is a direct function of Young's Modulus. Hence simply changing materials will not necessarily get you a change in natural frequency. Note that enclosure vibratory response can often be perceived as a reduction in clarity rather than being heard as "buzzing".
Every amp DBT I have ever read about used cone speakers for listening, and every time the conclusions, as I recall, were that the amps were DBT indistinguishable one from another. I believe the inductive electrical characteristic of the coned speakers used blurred the discinctions among the amps while the essentially resistive electrical characteristics of Magnepans would allow the discinctions among the amps to be audible. Its a consideration for those wishing to hold another amp DBT.
Almost all electrostatic speakers in existance have coupling transformers, and amps also used to have coupling transformers but coupling transformers were eliminated from amps for improved sonic performance. One exception I remember is a high-voltage electrostatic speaker featured in The Audio Amateur c1975 which was direct-connected to a dedicated high-voltage tube amp. As I recall the high-impedance membrane was charged to 3000 volts. This speaker was described as being a killer in every sense of the word.
Imagine this... a world where all amps sound the same...
We have now entered...the Twighlight Zone...
-
Quote:
Originally Posted by Peter_Klim
Imagine this... a world where all amps sound the same...
We have now entered...the Twighlight Zone...
Hey, let's all stay with the good spirit here. :)
rw
-
Gee, E-Stat, it is good to learn that
the Acoustat X design, released in 1976 also ran high voltage tube amps directly to the plates of electrostatic panels at slightly over 3kV.
Later, they replaced the tube amps with bi-formers increasing the drive voltage to around 5 kV.
The current Sound Labs electrostats run the bias higher still.
Did these people buy the rights from the dude who had published the direct-coupled high-voltage electrostat concept in Audio Amateur ? Or did they simply rip him off? Do you know ? Did you research to learn if these speaker companies held valid patents to the concept? You can't get a valid patent on what someone else invents and publishes......
Jim Winey got his first patent in 1972 and that patent caught my eye. I built a small test panel and found the idea worked quite well even if my demo unit had very low output.
Peter Kim: There are two issues:
1. Do all amps sound the same?
2. Will the speakers you use allow you really perceive any differences between amps that may actually exist?
The DBT tests I read said "all amps sound the same". My little test said "They don't" But my test was for my benefit: what to use on Tympani. The "answer" I got is why I have Futtermans.
Noddinoff: Stereophile measures cabinet vibrations and publishes 3-D output charts in their speaker tests on the correct premise that cabinet vibrations includingthose in the cone mounting board that occur in response to the audio being played blurs clarity.
Ever hear the horn of an approaching train at a RR crossing and notice the pitch of the horn drop as the locomotive passes you? That is the (Doppler) effect of a moving source. Or do you race to cross the RR tracks before the train can reach the crossing?
-
Quote:
Originally Posted by Mash
Did these people buy the rights from the dude who had published the direct-coupled high-voltage electrostat concept in Audio Amateur ? Or did they simply rip him off? Do you know ? Did you research to learn if these speaker companies held valid patents to the concept? You can't get a valid patent on what someone else invents and publishes......
Is this it?
<a href="http://hitechnetworks.net/bwaldron/projects/sanders_esl.htm">1975 Audio Amateur estat</a href>
This hybrid TL woofer / electrostatic design was by Roger Sanders (later of Innersound Labs). James Strickland of Acoustat originally began work in 1962 on his full range electrostat following Arthur Janszen's 1953 sheathed conductor design to extend a technology previously limited to tweeters. Acoustat was founded in 1972 and the first model X was in production by 1976. Did he copy Sander's design? Don't think so. It takes quite a while from prototype creation to production. Sanders didn't start producing his own electrostatic speaker until decades later. Strickland was granted patent 4,324,950 submitted 6/25/79 for his direct drive tube amplfier design. Harold Beveridge also hold a slightly earlier patent on another tube amp design. Ironically, both Sanders and Strickland moved on to better transformer designs.
Quote:
Originally Posted by Mash
The DBT tests I read said "all amps sound the same". My little test said "They don't" But my test was for my benefit: what to use on Tympani. The "answer" I got is why I have Futtermans.
Very interesting.
rw
-
Quote:
Originally Posted by E-Stat
Hey, let's all stay with the good spirit here. :)
rw
Sorry, I didn't intend it come off with bad spirit.
-
Correction
Quote:
Originally Posted by Mash
The DBT tests I read said "all amps sound the same". My little test said "They don't" But my test was for my benefit: what to use on Tympani. The "answer" I got is why I have Futtermans.
Well, perhaps you haven't looked up enough on DBTs, for you have rather overstated the matter. A number of them do show differences between amps and this is reflected on the ABX site.
http://web.archive.org/web/200108110...t/abx_data.htm
Now, you could distinguish between tube and solid state amps, but apparently some tube amps sound pretty much like solid state ones. One main difference seems to be the output impedance, which is usually high for tube amps, but could be high for solid state amps. So, I usually distinguish between amps with low and high output impedances. Here are measured results of one such amp, and it happens to be a tube amp: it will sound different into most speaker loads and one would hardly need a DBT to figure that out! Check out the green line for the response into the simulated speaker load in Chart 1:
http://www.soundstagemagazine.com/me...ere_ma1_mkii2/
That is a little extreme, and most tube amps have a less extreme output impedance. Stereophile and Soundstage use a simulated speaker load in their amplifier measurements; so did the old Audio magazine.
-
Quote:
Originally Posted by Mash
Noddinoff: Stereophile measures cabinet vibrations and publishes 3-D output charts in their speaker tests on the correct premise that cabinet vibrations includingthose in the cone mounting board that occur in response to the audio being played blurs clarity.
Ever hear the horn of an approaching train at a RR crossing and notice the pitch of the horn drop as the locomotive passes you? That is the (Doppler) effect of a moving source. Or do you race to cross the RR tracks before the train can reach the crossing?
Mash, I do appreciate the effort at simplicity. But I think you only confuse your points. Doppler shifts don't blur clarity; they change pitch. I don't think you mean to say that cabinet vibrations are changing the pitch of the music, that would be silly. If you are discussing sound waves in air, interference blurs clarity. I still feel we're having two conversations...I'm still attempting to summarize your points, with out appearing to have a high mass to volume ratio. From the above, I now see you were trying to connect mass, stiffeness, etc to the cabinate's structural resonances which of course interfer with the good good good good vibrations of the music. You never exactly addressed whether the natural frequency was in a range of concern, only to say that it wouldn't change much based on stiffness and mass. Still, to me this physics only view doesn't really seem key. a dime-store violin and a Stadivarius probably have similar Young's moduli but they most certainly resonate differently.The non-uniform nature and architecture of the materials have a greater bearing on the ability to transmit frequences than their densities and stiffness.
But back to electrical resonance. Your point, I think, was that some speaker types add wave forms to the electrical signal from the amp that could be perceived.
Quote:
Originally Posted by Mash
I believe the inductive electrical characteristic of the coned speakers used blurred the discinctions among the amps while the essentially resistive electrical characteristics of Magnepans would allow the discinctions among the amps to be audible. Its a consideration for those wishing to hold another amp DBT.
You didn't answer my question as to if this actually affects the amps output. You also imply that this allows diiferences in amps to be heard. Yet you also made the statement that there is really no difference in amps. I see that E-stat already phrased the appropriate question
{QUOTE=E-stat]So the presence or lack of inductance is a primary source for audible differences among amplifiers. Is that the sole determining factor?[/QUOTE]
But it sounds to me like the inductance discussed is in the speaker. So thus, the speaker is the determining factor in differences among amplifier? Logic error there. My mass to volume ratio is climbing...
noddin0ff
-
Quote:
Originally Posted by noddin0ff
{E-stat]So the presence or lack of inductance is a primary source for audible differences among amplifiers. Is that the sole determining factor?
But it sounds to me like the inductance discussed is in the speaker. So thus, the speaker is the determining factor in differences among amplifier? Logic error there. My mass to volume ratio is climbing...
A thousand pardons for the understood adjective clause. Here is the corrected version:
So the presence or lack of inductance as presented by the loudspeaker and it's cables is the primary source for audible differences among amplifiers. Is that the sole determining factor?
I think such an assertion to be a vast simplification of real world behavior.
rw
-
Quote:
Originally Posted by Pat D
Have you found anything more current than from the Carter Administration?
rw
-
Quote:
Originally Posted by E-Stat
Tests are entirely valid for that which they test. For what they don't test, they stand mute.
You presume that audible differences only occur upon distress. Differences in amplifier design, implementation and construction have zero effect whatsoever, right? RIGHT!
You will only know that to be the case for when the testing encompasses all that which is available. I agree that there is little difference between a Buick and an Oldsmobile. WRONG!
rw
As this is probably my last post here I will attempt to answer your ignorance before I leave. I really don't think your mind is open, but...
Point 1: Amp construction and materials are irrelevant (for audible differences)
Amps of different design, contruction, materials, etc. do not matter because their job is to simply take a input electrical signal and amplify it. Anything added to or left out of the input signal can easily be characterized; it is distortion (by definition). The difference between the output signal and the input (minus gain) is distortion. We can, of course, make measurements of this distortion. You may choose to disagree with how distortion is measured or how it is reported in spec sheets, but the fact remains that distortion can be measured.) The point is: how the amp is designed or what materials are used may or may not affect distortion--it really doesn't matter because we can measure the distortion. If two amps of completely different circuit design or materials have the same distortion, then sound generated from thier output will be the same. This is self-evident and cannot logically be debated.
Now, measured distortion may be found to be very low, so low in fact that people can't hear differences in distortion at that level. This is also an elementary principle that cannot be denied. You can't, for example, see things below a certain resolution. Differences in resolution below that level do not matter because you cannot see them. (Note that not all people have be tested, but enough have for use to know the limits of human perception on statistical grounds.) If you do not accept that there is some minimum in detectable distortion then you are being illogical. Differences in distortion below the detectable threshold do not matter (to our ears, but can be detected by instruments). One CAN (barely) debate what that minimum level of distortion is, but amp distortion is way below the commonly accepted levels so this is a fairly specious debate.
Need I repeat the statement that the distortion of most (even cheap) home audio amps is well below audibility?
Point 2: No need to test every case to establish facts and truth
Scientific proof does not require testing of every case as such testing is ridiculously impossible. Instead science takes positions such as: science doesn't prove negaitves, assertions must be demonstrated--not proven not to be true. The use of the null hypothesis (used in ALL psychological and preceptual testing) is also designed to deal with this problem by posing scientific questions in a way that does not require all cases to be tested (which is impossible for most scientific principles).
Not all matter in the universe has be tested to see if it follows Newton's Law of UNIVERSAL Gravitation, but scientists and reasonable people hold the law to be true for all matter. We have not tested all people to see if they can jump to the moon, but we have plenty of good reasons to believe that nobody can (including the data that nobody has).
There are plenty of good reasons to believe that people cannot distinguish properly performing amps (such as the measured distortion and the fact that nobody has). There are also plenty of good scientific reasons NOT to believe the reports of those who claim they have. (For example, some who claim to easily hear differences have failed to demonstate that they can.)
Ignorance of logic and scientific method is very common. Everyone can choose to simply believe what they wish to believe; or they can choose to educate themselves. Many so-called audiophiles choose the former--it is a lot easier and offers ego rewards. Those how following a logical path will get better home audio results for a lot less money.
-
Quote:
Originally Posted by E-Stat
A thousand pardons for the understood adjective clause. Here is the corrected version:
So the presence or lack of inductance as presented by the loudspeaker and it's cables is the primary source for audible differences among amplifiers. Is that the sole determining factor?
I think such an assertion to be a vast simplification of real world behavior.
rw
To complete my arguments of the previous post I need to address the assertion that different load impedances (of speakers) determine differences in what we hear.
First off, nobody denies that speakers sound different and are (somewhat) easily differentiated in blind tests (no matter how nervous the listeners are). Gee, I wonder why DBT works for these components and not electronics? (Actually, I don't wonder, I know why).
Second, nobody denies that amps have differing output into different loads (impedances). Some are "better" than others (not always predictable based on cost and certainly not based on materials used). But, are speakers presenting loads that are outside the abilities of cheap amps? Where is the evidence for this assertion? E-stat and his buddies simply take this as a given without demonstrating it.
E-stats also doesn't seem to understand the basics of how electrical factors interact, He views the "real world" as always too complex to describe or measure accurately (except, ironically enough, with his ears). He does not seem to get that "complex" electrical signals are really simply combinations of simple signals. This is a key lack of understanding. He cannot say why tests of sine wave signals are somehow different or not "complex" enough even though it is a scientific fact that music signals are simply superpositions of multiple sine waves.
E-stat, and many of his ilk, simply hand wave and make statements like "iinductance as presented by the loudspeaker and it's cables is the primary source for audible differences among amplifiers" Does he think that we should test amp output into different inductances? No, the real world is too complex. Based on what expertise does he make these claims? None beyond what he thinks he hears!
Does anyone really think E-stat knows how inductance really works? Please tell us, Mr. E, does various load inductance affect the frequency response and distortion of the amp output? Can't we just measure that by placing different inductors across the electrical output and measureing frequency response and distortion? Is the performance of your high-priced amps better into various inductances?
I predict you will say that speaker loads are too "complex" in the "real world" to do that. Why else is he not calling for this kind of testing or quoting specs from his prized amps? If you think that an amp's perfomrance cannot be tested by specific inductance loads, then why not say so? I don't think you are going to get much support from EEs or physists.
You don't get the basics, as I have told you many times before. Why not find out?
-
Quote:
Originally Posted by RobotCzar
Point 1: Amp construction and materials are irrelevant (for audible differences)
Sorry, connector quality, wiring quality, and internal component composition do matter.
Quote:
Originally Posted by RobotCzar
Need I repeat the statement that the distortion of most (even cheap) home audio amps is well below audibility?
You may repeat your claim any number of times as desired.
Quote:
Originally Posted by RobotCzar
Point 2: No need to test every case to establish facts and truth...
Scientific proof does not require testing of every case as such testing is ridiculously impossible
You do, however, have to test a representative sample of that which is available. You continue to plead a case with zero substantiation. Is your opinion based upon the same quarter century old data as used by Pat D.?
rw
-
Objection
Quote:
Originally Posted by E-Stat
Sorry, connector quality, wiring quality, and internal component composition do matter.
You may repeat your claim any number of times as desired.
You do, however, have to test a representative sample of that which is available. You continue to plead a case with zero substantiation. Is your opinion based upon the same quarter century old data as used by Pat D.?
rw
My most recent referral to the ABX data was to counter the assertion that DBTs never find any differences. They are also easy to link. Not everything on the site is that old, BTW. There have been others which I have and mtry has a bunch, but I am not aware of links to them on the net except as bibliographical references.
I also pointed out an amp which will sound different, as it is quite clear that with many speaker loads, its FR will differ from flat by a clearly audible amount (using the ABX site's matching criteria graph, BTW).
Oh well, you are not the only person here that wants to misrepresent my opinions.
-
Quote:
Originally Posted by RobotCzar
E-stat, and many of his ilk, simply hand wave and make statements like "iinductance as presented by the loudspeaker and it's cables is the primary source for audible differences among amplifiers"
No, that would be Mash's assertion. Catch up on the thread. I was asking him if that is actually what he meant.
Quote:
Originally Posted by RobotCzar
You don't get the basics, as I have told you many times before. Why not find out?
Do you know of any world class automobile drivers who are automotive engineers? Why do you think that is? Don't you have to be an engineering geek to understand how a car corners at 4 lateral Gs? Evidently not. Exclusively not.
Do you know of any world class musicians who are acousticians and/or cabinet makers? Why do you think that is? Do you get the pattern here?
The answer is obvious. You see oscilloscope plots. I and many others like me hear the emotion in a voice or a cello. Sure Redbook CD is audibly perfect as are Best Buy sourced receivers. :)
rw
-
Ok, this is becoming a flame war... As a someone trying to learn more about audio, may I offer my 'neutral' opinion.
So far all the arguments for differences in amp performance made in this thread cite the load presented by the downstream components. That to me implies that it is the downstream components that are different. Not the amp output. Since this thread has been using many cheesy analogies, here's mine.
If you take a fine sports car engine and place it in the body of a well constructed sports car, you will have nimble, responsive performance. Put it in a school bus, it will be lacking in response and a bad driving experience. Put it in a motorcycle and you have an accident waiting to happen. The engine performs to spec (x gas converted to y acceleration).
Do you blame the engine? NO! You blame the body for the difference. Physics do not lie. Electrical signals can be very very accurately measured. The functioning of almost every aspect of modern life depends on accurate measurement of electical signals. This is one thing humans do extremely well. Humans do this much better than accurately recalling experiences. If your view is that we still rely on visual inspections of oscilloscope patterns, you need to fast forward about 100 years. If an amp's measurable distortion is beyond our ability to discriminate, then it IS beyond our ability to discriminate. But this assumes an ideal speaker downstream. And, distortion is not the same as power output.
As far as emotional content, that has a lot to do with volume, resonance, tonal coloration, your mood, who's sitting next to you, and much more. If you find emotional content to vary from set up to set up, it's likely downstream of the amp.
I'll believe that well characterized amps performing within reasonable specs on an ideal speaker/cable have no sonic difference. I'll also believe that Mash and E-stat hear differences. However, I'm inclined to believe these differences are due to downstream components and not the 'quality' of the amp. I say this based on the nature of the physics lessons above. If the only difference is the system is an amp swap, then the audible difference is either A) between the listeners ears, or B) due to power differences and not due to the quality of the amplification per se.
Yes?
noddin0ff
-
Glad you have come to see the light!
[QUOTE=E-Stat] Sure Redbook CD is audibly perfect as are Best Buy sourced receivers./QUOTE]
Glad you have come to see the light.;)
-
Bush Senior.
Quote:
Originally Posted by E-Stat
Have you found anything more current than from the Carter Administration?
rw
E. Brad Meyer, "The Amp/Speaker Interface: Are your loudspeakers turning your amplifier into a tone control?", in Stereo Review, June 1991, p. 53-56.
E. Brad Meyer shows some measurements using a solid state amplifier and as tube amplifier. With his own speakers, the difference was just slightly audible using pink noise but did not show up on music; with a speaker offering a more difficult load, the differences were more audible. The solid state amp maintained a very flat frequency response into the difficult speaker load while the tube amp showed some dips which were fairly audible on pink noise.
I am not sure of the history, but you will no doubt have noticed that the old Audio magazine, Stereophile, and also Soundstage, not only measure the response of amplifiers into a resistor but also into simulated speaker loads (these have been standardized, I think). So I think his conclusions were (mostly tacitly) widely accepted.
-
Quote:
Originally Posted by Pat D
E. Brad Meyer, "The Amp/Speaker Interface: Are your loudspeakers turning your amplifier into a tone control?", in Stereo Review, June 1991, p. 53-56.
E. Brad Meyer shows some measurements using a solid state amplifier and as tube amplifier. With his own speakers, the difference was just slightly audible using pink noise but did not show up on music; with a speaker offering a more difficult load, the differences were more audible. The solid state amp maintained a very flat frequency response into the difficult speaker load while the tube amp showed some dips which were fairly audible on pink noise.
I am not sure of the history, but you will no doubt have noticed that the old Audio magazine, Stereophile, and also Soundstage, not only measure the response of amplifiers into a resistor but also into simulated speaker loads (these have been standardized, I think). So I think his conclusions were (mostly tacitly) widely accepted.
Thanks for the reference, but it sure would help to have the kind of specific information that Bert Whyte, Bascom King, et. al. with Audio Magazine would provide. In this case, the audible difference could simply be due to the inherently low damping factors found with most tube amps. Such could result in frequency variations on the speaker side if it had a roller coaster impedance curve like an Advent. If Brad were using a typical Stereo Review mid-market receiver, then any results would hardly be "conclusive" to the absolute assertions made by some folks.
rw
-
Quote:
Originally Posted by E-Stat
Thanks for the reference, but it sure would help to have the kind of specific information that Bert Whyte, Bascom King, et. al. with Audio Magazine would provide. In this case, the audible difference could simply be due to the inherently low damping factors found with most tube amps. Such could result in frequency variations on the speaker side if it had a roller coaster impedance curve like an Advent. If Brad were using a typical Stereo Review mid-market receiver, then any results would hardly be "conclusive" to the absolute assertions made by some folks.
rw
Well, a high output impedance translates into a low damping factor, since the damping factor is simply the nominal impedance of the speaker divided by the source impedance. This would in actual use include the resistance of the speaker cables, as well. An amp with a 0.1 output impedance would have a damping factor of 80 into an 8 ohm speakers.
Now, tube amps often have a high output impedance (low damping factor, if you wish), usually 1-4 ohm and I linked above to one with an output impedance of about 10 ohms. That will affect the frequency response into the varying impedance vs. frequency load of many speakers, often enough to be audible. Such data can be found at the Soundstage and Stereophile sites for a number of different amplifiers. What are you looking for?
Brad Meyer did not use a receiver (your remark indicates a prejudice against Stereo Review :eek: ) but power amplifiers, but he did not specify which ones. The SS one was under a grand, the tube amp was considerably more expensive, that's all he said. I think his point was to illustrate a test method.
|