Amplifier Sound

[mtrycrafts]

I can't imagine a more golden opportunity without having to put any cash on the table. Just go and collect your $10,000.

Certainly cannot do it in Vegas, they need money up front

 

[mtrycrafts]

It's unbelievable how many people have already tried and failed in that test. Last time I checked, no one had even got much over 50% right. It's doubtful anyone will ever win that money, but many have tried.

I must have missed the numbers how many have taken the challenge?

He can hook you up with enough scientific papers, journals & test results to choke a horse.

I think he is kinder to horses

 

[NewYorkJosh]

It's unbelievable how many people have already tried and failed in that test. Last time I checked, no one had even got much over 50% right. It's doubtful anyone will ever win that money, but many have tried.

NYJ, if you're honest in wanting to know, shoot WmAX a PM. He can hook you up with enough scientific papers, journals & test results to choke a horse.

Thanks Rob! I've gone ahead and sent WmAX a PM. I'll definitely look at the evidence. I've read the $10,000 challenge thread on AVSforum.com. I don't see anything there about how many people have taken the test, and what the results are. Did you get the results from Richard Clark? Or are they posted somewhere? (If so, please provide a link).

While I'd love to win the challenge and crow it over all of you, the reality isn't so simple:
I'll quote directly from the AVSforum thread:

12. Although anyone is welcome to take the test, only subjects employed in the car audio industry or Car Sound subscribers are eligible for the $10,000.00 prize.
13. Cost to take the test is $100.00. $300.00 for people representing companies. Payable in advance, scheduled appointments only. Done correctly the test takes several hours and I don't have the time if you aren't serious.

So - in truth - for me to take the challenge I'd have to disconnect my gear, transport it (i.e. drive) about 800 miles, pay $100, and then not be eligible to actually win the $10K after all. There are some additional issues:
1) Hostile environment - I don't know if an adversarial challenge environment is condusive to the state of relaxed concentration I usually do my listening in.
2) Equalization networks - The challenge specifies a (probably) low-fi equalization network goes on. High-end sound is pretty fragile. I'd wager that this equalization would make things sound unfamiliar to me.
3) Room effects and overall system matching: I've found that amplifier sound is definitely a product of a system. A given amp can sound differently depending on the impedence of the speakers it drives, and the capacitance (and inherent tonal qualities) of the cables used. [Please don't flame me on the cable issue - I'm not defending the idea of $10,000 cables (at least here)]. Also the listening room setup has a great influence on how speakers sound. I have trouble believing that a novel setup will be comfortable or familiar enough for me to do meaningful listening in the short period allotted. These are a lot of my criticisms of DBT amp testing in general as currently practiced. Note - I'm not rejecting DBT as a methodology. I simply argue that it hasn't been properly applied because what makes amp differences audible hasn't been adequately studied.

 

[NewYorkJosh]

What proof will you accept? What will satisfy you or change your mind?

Oh, you have convincing evidence that they sound different within their design limits?

The proof I need is in the testing methodologies used.
1) I need to see that amps of different designs are tested in a system of sufficient quality to reveal the differences. A part of this is the room needs to have been set up with enough care to make room interaction an inconclusive factor.
2) I need to see that listeners could sit in the stereo sweet spot and have enough time to do meaningful listing.

I do not have proof of the audibility of differences between amplifiers. I have my own experience over the decades I've been an audioholic. I'm not saying that my experience is scientific - or that it should convince anyone. However, it makes me personally suspicious of the claim of inaudibility. I require the evidence to even consider changing my mind. I also need to participate in some testing.

I aknowledge the placibo effect exists. I can't deny that some measure of it may play a part in my experience. However, my experiences in extended listing are so consistent and real to me that I have a lot of trouble accepting that this is all. I just swapped amps in an otherwise identical system yesterday (an ARC Classic 60 tube amp for a Yamaha 5760 driving ProAc response 2s - the rest of the system high-end) and immediately heard a dramatic difference. To be fair the Yamaha is still breaking in. I'll continue to do this test over the Yamaha's next hundred hours and if I still find the difference dramatic will so what I can do to find a way to implement a DBT. If I'm unable to tell then I'll be very happy to change my position. If I can tell I'll be happy to present my evidence.

 

[Unregistered]

1) I need to see that amps of different designs are tested in a system of sufficient quality to reveal the differences. A part of this is the room needs to have been set up with enough care to make room interaction an inconclusive factor.
2) I need to see that listeners could sit in the stereo sweet spot and have enough time to do meaningful listing.

These are the typical 'caveats' mentioned every time by someone who believes as you do. If you take a DBT and fail you can always fall back on the notion that the system wasn't of 'sufficient quality' or that you didn't have enough time to thoroughly evaluate it. The whole point is that if you were to take such a test you would likely find, as have so many others, that there is in fact little to no difference in 'quality' whatsoever.

 

[NewYorkJosh]

If they are typical excuses perhaps there's a logical reason why. In the meantime it's clear I'm going to need to do some extensive blind testing to get this issue straight in my own mind. I'll get back to you.

These are the typical 'caveats' mentioned every time by someone who believes as you do. If you take a DBT and fail you can always fall back on the notion that the system wasn't of 'sufficient quality' or that you didn't have enough time to thoroughly evaluate it. The whole point is that if you were to take such a test you would likely find, as have so many others, that there is in fact little to no difference in 'quality' whatsoever.

 

[Rip Van Woofer]

Be aware that A/B testing, blind or otherwise, don't mean squat unless the levels are VERY precisely matched to less than 0.15db; 0.1dB is the standard. Yes, the decimal is in the right place! It takes time, futzing, and some good instrumentation to do it right. Any greater level change can be perceived (though it will not be obviously "louder") and the louder one will usually "win".

 

[Karp]

...The amps are tested to find their linear range (i.e. where they operate w/o distortion or clipping).
The linear range of the less powerful amp is selected (so that 10-watt tube amp isn’t driven into clipping while going up against the $10,000 SS amp) Adjustments are made for the input sensitivity of each amp. If necessary, the amps are EQ’ed to sound the same, apparently some amp makers boost/cut certain frequencies to give their amp a distinctive sound...

Okay, I understand why the test would occur at a db level where the "lesser" amp does not clip. It also makes sense to use a db meter with white noise to get the two amps at the same level. What makes no sense to me is the EQ'ing part. If they have to be EQ'd, then it is obvious that the two amps do not sound the same. Period.

I have heard differences in receivers/amps although others say there are no differences in sound. My point has always been that you can take two receivers and db them with white noise so that they match, but then db them with test tones and they vary. If they vary with linear test tones, they certainly sound differently with music. Whether the manufacturers are boosting/cutting frequencies on purpose or if it because of differences in design is a moot point. The bottom end is that the two units DO sound differently.

 

[WmAx]

Many of the terms of that challenge have been relaxed over the last couple of years. If you are serious about subjecting yourself to such a test, contact Mr. Clark directly and find out what he is willing to accomodate for you.

As for E.Q., it is applied to the control amp, not both amps. FOr example, if one presents him with their amplifier and it measures non-flat, he will apply E.Q. to the 'low fi' amp it's being compared to, not to both amps.

The point of his test is to demonstrate that amplifier topology used is irrelevant as long as their measuable parameters are below known human thresholds of detectability as established by perceptual research.

-Chris

Thanks Rob! I've gone ahead and sent WmAX a PM. I'll definitely look at the evidence. I've read the $10,000 challenge thread on AVSforum.com. I don't see anything there about how many people have taken the test, and what the results are. Did you get the results from Richard Clark? Or are they posted somewhere? (If so, please provide a link).

While I'd love to win the challenge and crow it over all of you, the reality isn't so simple:
I'll quote directly from the AVSforum thread:

So - in truth - for me to take the challenge I'd have to disconnect my gear, transport it (i.e. drive) about 800 miles, pay $100, and then not be eligible to actually win the $10K after all. There are some additional issues:
1) Hostile environment - I don't know if an adversarial challenge environment is condusive to the state of relaxed concentration I usually do my listening in.
2) Equalization networks - The challenge specifies a (probably) low-fi equalization network goes on. High-end sound is pretty fragile. I'd wager that this equalization would make things sound unfamiliar to me.
3) Room effects and overall system matching: I've found that amplifier sound is definitely a product of a system. A given amp can sound differently depending on the impedence of the speakers it drives, and the capacitance (and inherent tonal qualities) of the cables used. [Please don't flame me on the cable issue - I'm not defending the idea of $10,000 cables (at least here)]. Also the listening room setup has a great influence on how speakers sound. I have trouble believing that a novel setup will be comfortable or familiar enough for me to do meaningful listening in the short period allotted. These are a lot of my criticisms of DBT amp testing in general as currently practiced. Note - I'm not rejecting DBT as a methodology. I simply argue that it hasn't been properly applied because what makes amp differences audible hasn't been adequately studied.

 

[Leprkon]

1) I need to see that amps of different designs are tested in a system of sufficient quality to reveal the differences. A part of this is the room needs to have been set up with enough care to make room interaction an inconclusive factor.
2) I need to see that listeners could sit in the stereo sweet spot and have enough time to do meaningful listing

so you need $ 10,000 worth of ADDITIONAL equipment, another $10,000 worth of room mods and a couple of hours to determine the difference between a Sony and a Sunfire ??

It sounds like there's not that much doggone difference. I (and every girl at the local Hooters) can tell the difference between a twenty grand Miata and a forty thousand dollar Thunderbird in nothing flat... which one of us is spending our money more wisely ???

 

[WmAx]

...
I have heard differences in receivers/amps although others say there are no differences in sound. My point has always been that you can take two receivers and db them with white noise so that they match, but then db them with test tones and they vary. If they vary with linear test tones, they certainly sound differently with music. Whether the manufacturers are boosting/cutting frequencies on purpose or if it because of differences in design is a moot point. The bottom end is that the two units DO sound differently.

The point of this challenge was to address the assertions(dare I call it theory?) by audiophiles that NON-MEASURABLE parameters contribute to amplifier sound signature. Mr. Clark merely insures that both amps measure frequency vs. amplitude equally and that the amps are not used into clipping region(s). If non-measurable parameters are responsible for sound signature, then the audiophile theory should be proven true by subject(s) being able to achieve a statistically significant score in a controlled blilnd test. If not, then the audiophile theory has failed to be proven true.

-Chris

 

[NewYorkJosh]

so you need $ 10,000 worth of ADDITIONAL equipment, another $10,000 worth of room mods and a couple of hours to determine the difference between a Sony and a Sunfire ??

It sounds like there's not that much doggone difference. I (and every girl at the local Hooters) can tell the difference between a twenty grand Miata and a forty thousand dollar Thunderbird in nothing flat... which one of us is spending our money more wisely ???

Who said I was going to spend $10,000? I can put together a really musically satisfying stereo system for less than $3,000 (if I can buy used equipment). That should be enough to do a decent amp test. In any case I'd further argue that certain amps mate particularly well with certain speakers - another subjectivist "heresy". The end result being that some combinations work a particular sonic magic. If the goal is getting laid at the local hooters - well go get 'em and power to you. I've found that fancy stereo equipment doesn't do much for the ladies. If the goal is to really have a good time listening to music - you know hearing amazing detail and boogy factor that makes you want to get up and move and then pull out all those old favorites and listen to them again so that you can hear all the music in them that you've never heard before - well... maybe the money is well spent on the stereo after all. I don't think you're going to get there with just any amp and any pair of speakers.

I just need to find a way to put this in a scientific way - with objectively verifiable and testable attributes to satisfy the particular focus of this forum. Thus the interest in DBT.

 

[NewYorkJosh]

The point of this challenge was to address the assertions(dare I call it theory?) by audiophiles that NON-MEASURABLE parameters contribute to amplifier sound signature.-Chris

That's the idea, Chris. But I'd amend it to say "not commonly measured", not "non-measurable". My hypothesis is that the current state of audio science is missing something - not that the thing itself is unknowable. I'm getting into deep waters here - I'm just an audiophile who stumbled on this site while looking for info about HT receivers (a topic new to me) and found myself getting into the middle of an age-old arguement. I'm not the one to blaze the resolution - but I will attempt some blind testing to see if my sacred cows are easily slaughtered.

 

[mtrycrafts]

The proof I need is in the testing methodologies used.
1) I need to see that amps of different designs are tested in a system of sufficient quality to reveal the differences. A part of this is the room needs to have been set up with enough care to make room interaction an inconclusive factor.
2) I need to see that listeners could sit in the stereo sweet spot and have enough time to do meaningful listing.

Back to my initial post.
You claim that you need all these fancy systems, yet every golden ear Tom, ****, and Harry has no problem hearing differences in any system, in any room, with any material. YOU cannot have it both ways. If these systems didn have the resolving capability, music would sound like even level skips on a stairstep. How could it resolve any small increments in a continuous, fluid manner?

You either need a system as you describe and that is the only system to reveal audible differences, or, it is just another lame excuse for failure.
But then, I don't suppose you have evidence for a need for this system? That is yet another one of your claims and burdens to prove.

"The Great Ego Crunchers: Equalized, Double Blind Testing", Shanefield, Daniel, Hi-Fidelity, Mar 80, pg 57-61.


This is a paper on long term listeing, taking months. Guess what the outcome was?
I thought you might come up with some original ideas but just recycled excuses, nothing new.

I do not have proof of the audibility of differences between amplifiers.

I didn't think you had or it would be posted, I am sure.


I have my own experience over the decades I've been an audioholic. I'm not saying that my experience is scientific - or that it should convince anyone. However, it makes me personally suspicious of the claim of inaudibility.

Of course. No difference as people been exposed to psychics. Nothing will convince them that they faces. Their experience tells them otherwise. Alien abductees? Same. They will never be convinced. After all, even Juhn Mack, a Harvard professor supports them.
How about the ones who use homeopathic medicines? Or that Airborne cold stuff. That is so powerful that it work just by carrying it along your trip in the luggage. I tried it. It works. It even works when I leave it home. Magical. I cannot explain it to you but it works. I have the experience. You cannot convince me otherwise.


I require the evidence to even consider changing my mind.

You then, you have the required system or sufficient resolution, room setup to conduct your testing.

I also need to participate in some testing.

Now that is the crux of it, you need to sit down and do it properly.

I aknowledge the placibo effect exists. I can't deny that some measure of it may play a part in my experience.

There is hope yet

However, my experiences in extended listing are so consistent and real to me that I have a lot of trouble accepting that this is all.

But all your listeing is under biased conditions, conditions you have no control over, no on/off switch. So, no matter how long you do unreliable protocols, in the end, it is still a whole lot of unreliable listeing, one day or ten years.



I just swapped amps in an otherwise identical system yesterday (an ARC Classic 60 tube amp for a Yamaha 5760 driving ProAc response 2s - the rest of the system high-end) and immediately heard a dramatic difference.

Maybe yes, maybe no. No way of knowing as you compared it. Levels matched to 0.1 dB spl? Level matched at a number of frequencies? Maybe one, the tube amp is broken, or designed euphonic? Certainly not conducted under bias controlled conditions. Nothing to conclude.

To be fair the Yamaha is still breaking in.

No such thing. Yet another audio myth, voodoo, urban legend. No evidence.

I'll continue to do this test over the Yamaha's next hundred hours

Waste of time.


and if I still find the difference dramatic will so what I can do to find a way to implement a DBT. If I'm unable to tell then I'll be very happy to change my position. If I can tell I'll be happy to present my evidence.

Without a credible comparison, 3rd party reporting, you will have another anecdote, unreliable and unconvincing.

But, I appreciate the exchange

 

[mtrycrafts]

Be aware that A/B testing, blind or otherwise, don't mean squat unless the levels are VERY precisely matched to less than 0.15db; 0.1dB is the standard. Yes, the decimal is in the right place! It takes time, futzing, and some good instrumentation to do it right. Any greater level change can be perceived (though it will not be obviously "louder") and the louder one will usually "win".

How true.

One needs a test disc with 100Hz, 1kHz and 10kHz. That is where Nousaine level matches
And, you need a voltmeter that can resolve 1% at 2.83V. That is .03V at these frequencies. This is .1dB spl
If the amps have level differences at these frequencies,

That is the technical aspect. Then the blinding aspect and statistics.

All the claims for resolution in a system. WOW, if a system cannot reproduce at these levels, it will sound horrible with flat response all over the place. But, it is just excuses.

 

[mtrycrafts]

1) I need to see that amps of different designs are tested in a system of sufficient quality to reveal the differences. A part of this is the room needs to have been set up with enough care to make room interaction an inconclusive factor.
2) I need to see that listeners could sit in the stereo sweet spot and have enough time to do meaningful listing

so you need $ 10,000 worth of ADDITIONAL equipment, another $10,000 worth of room mods and a couple of hours to determine the difference between a Sony and a Sunfire ??

It sounds like there's not that much doggone difference. I (and every girl at the local Hooters) can tell the difference between a twenty grand Miata and a forty thousand dollar Thunderbird in nothing flat... which one of us is spending our money more wisely ???

No, they don't need all that, just an excuse. Why can they hear differences with their eyes wide open with their system, no such equipment in place and then nothing when their eyes are shut? Oh, yes, the eyes do the hearing, not their ears. Curious.

 

[mtrycrafts]

What makes no sense to me is the EQ'ing part. If they have to be EQ'd, then it is obvious that the two amps do not sound the same. Period.

Yes, then one is designed to be euphonic if it is above the threshold of detection.

My point has always been that you can take two receivers and db them with white noise so that they match, but then db them with test tones and they vary.

Pink/white noise is random over the entire band. It cannot level match the component. Single frequency test tones are needed.

If they vary with linear test tones, they certainly sound differently with music.

Only if the variation is above the threshold of detection. With music, it is a lot. But, that is why historically most are null results and the ones that are not, we know why.

Whether the manufacturers are boosting/cutting frequencies on purpose or if it because of differences in design is a moot point. The bottom end is that the two units DO sound differently.[/]

No, not a moot point. One can always design a euphonic amp. And, that will show up on the spec sheet, especially the frequency response. No, they don't design amps with a cut in only one band, or a boos in one band. When it happens, it is over a broad band of frequencies, like a roll off above 1kHz, or 10kHz?

 

[surveyor]

I would like to know the Audioholics staff opinion?

To the staff members of Audioholics:
Do all amps sound alike?
Does the DBT information discussed in this series of posts hold water?
Please let the staff members answer my question?

Respectfully,

Surveyor

 

[Rip Van Woofer]

...I (and every girl at the local Hooters) can tell the difference between a twenty grand Miata and a forty thousand dollar Thunderbird in nothing flat... which one of us is spending our money more wisely ???

The one with the twenty grand Miata! And who eats better food elsewhere for the same price as Hooters, albeit without the "entertainment".

 

[miklorsmith]

NewYorkJosh - I agree with your assessments of this problem and I would take the argument a large step further. Why would anybody concerned enough about HiFi to monitor this forum argue that all amplifiers sound the same? This position is utterly ludicrous, at least for anybody that knows anything about anything.

This statement requires acceptance of all the following: 1) Cables (all) make no difference; 2) Capacitors, resistors, power supplies and other functional internals are of no sonic consequence; 3) Internal topology is meaningless, solid state, tube, or otherwise (I have a digital-switching amplifier); 4) Construction quality, including soldering, case, and internal wiring, are esoterica; 5) Speaker/amplifier matching is an audiophile justification; and 6) Radio Shack produces the same ultimate product as Krell.

There are doubtless more implicit assumptions not listed above. C'mon, people! Why are you in this hobby if this is what you believ?. Your ears don't hear a spectrum graph. Measurements aren't everything. The CD player in my garage is 15 years old and basically produces a flat signal from top to bottom. Does it sound the same as the heavily modified player in my main system? Not even close!

I defy any of the "Don't tell me what you hear without backup data" people to even explain what measurements they might attribute to the sense of ease, dimension, and clarity of any fine component.

Setting up a true DBT for anything is exceptionally difficult. Equalizing the variables, especially without degrading the signal path, is critical but nearly impossible. HOWEVER:

One of biggest problem in DBT testing and one that keeps the professionals wary is one of the human mind: Aural memory is exceedingly short. The brain adapts quickly to incoming signals from the ears. As the professionals acknowledge, differences between, say, two interconnects are slight. Their comparisons never are between a Sony receiver and a Mark Levinson amp because, like I said, c'mon.

Regardless of what component is under discussion, they're all made of the same basic parts. If it's true for amps, your other components should be subject to your same theory. Amps, CD players, preamps, what's the difference? I guess speaker selection must be the only thing that makes any difference in any system.

Those of you who say there is no difference between amplifiers should proceed like this: Buy a pair of speakers you like, find the cheapest integrated amplifier you can find with sufficient power for your liking, buy the cheapest source components available and plug it all together with the RCA's and power cables that come with those new gems. Get a 400' spool of lamp cord to use for speaker cable. Can you believe people pay for wires?! What a scam.

Then, hang it up. Quit thinking about your system. You have everything you'll ever need, because upgrading is just an immeasurable figment of other people's imaginations.

Me? Can't wait to upgrade my interconnects to silver.