The way towards a good hifi is clouded by myths and lies. The true high ender is always looking for another voodoo accesoire to solve "problems" in his system.
De-magnetising cd's, special feet for your components to absorbe unwanted energy, coins you have to stick on the wall to improve acoustics, cable lifters, netconditioners, audio-racks etc.
Ask a recording engineer, ask most women or anybody else and they'll say you would be crazy to buy any of those things. Ask the regular audiophile, and he will say those things can turn your hifi-setup from a reasonably good one into something special!
Our audiophile minds are poluted with false "knowledge" of high end hifi. We know that the expensive Gryphon pre and power amp, combined with that great Mark Levinson cd-player, and wired with siltech and NordOst sounds wáááy better than the el cheapo Sony cd-player with Yamaha integrated and wired with the cables that came with the equipement.
WRONG. The sony and Yamaha may sound exactly the same as the Levinson Gryphon combo. I've never come across an amp that didn't sound good (better not go for tubes: it is more expensive to build a good transistor than to build a good tube amp. Most designs sound practically the same as a transistor amp, but don't deliver as much power). Neither have I ever heard a cd-player that didn't sound almost perfect. Cables are probably even less important. Feet to absorb negative energy? What should the negative energy in your equipement consist of?
If sound-quality is the only reason for you buying expensive equipement, buy cheaper. Select you amp on the power it has, and on the features you need. A goodsounding cd-player can be bought for as little as $70.
There are 3 things that can make a difference to the sound that you hear from your hifi: the quality of the recording, the speakers and the acoustics of the room they are playing in. We can not invluence the first thing. Most of our energy (and money) should go to the speakers. The best speakers in the world will distort the sound more, than will the sony yamaha combo with its supplied cables.
fire at will
De-magnetising cd's, special feet for your components to absorbe unwanted energy, coins you have to stick on the wall to improve acoustics, cable lifters, netconditioners, audio-racks etc.
Ask a recording engineer, ask most women or anybody else and they'll say you would be crazy to buy any of those things. Ask the regular audiophile, and he will say those things can turn your hifi-setup from a reasonably good one into something special!
Our audiophile minds are poluted with false "knowledge" of high end hifi. We know that the expensive Gryphon pre and power amp, combined with that great Mark Levinson cd-player, and wired with siltech and NordOst sounds wáááy better than the el cheapo Sony cd-player with Yamaha integrated and wired with the cables that came with the equipement.
WRONG. The sony and Yamaha may sound exactly the same as the Levinson Gryphon combo. I've never come across an amp that didn't sound good (better not go for tubes: it is more expensive to build a good transistor than to build a good tube amp. Most designs sound practically the same as a transistor amp, but don't deliver as much power). Neither have I ever heard a cd-player that didn't sound almost perfect. Cables are probably even less important. Feet to absorb negative energy? What should the negative energy in your equipement consist of?
If sound-quality is the only reason for you buying expensive equipement, buy cheaper. Select you amp on the power it has, and on the features you need. A goodsounding cd-player can be bought for as little as $70.
There are 3 things that can make a difference to the sound that you hear from your hifi: the quality of the recording, the speakers and the acoustics of the room they are playing in. We can not invluence the first thing. Most of our energy (and money) should go to the speakers. The best speakers in the world will distort the sound more, than will the sony yamaha combo with its supplied cables.
fire at will
don't mean to offend anyone. I once believed a lot off bull**** too. I now have spare cables that once costed me 800 euros! And a Purist Audio cd with a lot of noise on it, intended to "clean up" the system. I have also had a few magic feet. Managed to sell them again.
I once did a first double blind test between amps. I concluded that I could'nt hear the difference. Same for cd-players. The cables I use are very short. My speakercable is 5 feet long. On such short piece of wire I don't believe it makes any difference what type of cable u use. Record players dó sound different!
I once did a first double blind test between amps. I concluded that I could'nt hear the difference. Same for cd-players. The cables I use are very short. My speakercable is 5 feet long. On such short piece of wire I don't believe it makes any difference what type of cable u use. Record players dó sound different!
I got turned off of "Ultra-Fi" after doing listening to a pair of Avantegarde Duos beside some B&W 804S. The BW's were like 1/5 the cost, abeit not as sensitive, they were far more precise in their tonal balance and placement of the soundstage. It was a McIntosh Amplifier, and Mark Levinson CD source, so it's all audiofile quality setup. While the Duo's sound respectable in their own right, when you the B&W's side by side, you realize you're paying Lamborghini price for a Corvette.
Another thing that turned me off of Ultra-Fi was a review of some amplifier in Stereophile. I forget the details of the system, but the reviewer was going on and on about their great recording he had, and all the detail he heard. By coincidence I had the same recording and plugged it into my ancient Technics turntable, Luxman amp, and KEF speakers. We're talking ancient here, this amp had fairly audible hiss with no signal, and you could hear the hum from the transformer in it. I turned it up, and I heard everything he paid like 100,000 to hear.
By another coincidence I came accross the schematic for the Luxman amp I had. It bears a strikinng resemblance to a schematic that was floating around on DIY Audio claiming to be the best amp design ever. Luxman got a lot of mileage out of that design, as far as I know they didn't change it for thirty years.
Another thing that turned me off of Ultra-Fi was a review of some amplifier in Stereophile. I forget the details of the system, but the reviewer was going on and on about their great recording he had, and all the detail he heard. By coincidence I had the same recording and plugged it into my ancient Technics turntable, Luxman amp, and KEF speakers. We're talking ancient here, this amp had fairly audible hiss with no signal, and you could hear the hum from the transformer in it. I turned it up, and I heard everything he paid like 100,000 to hear.
By another coincidence I came accross the schematic for the Luxman amp I had. It bears a strikinng resemblance to a schematic that was floating around on DIY Audio claiming to be the best amp design ever. Luxman got a lot of mileage out of that design, as far as I know they didn't change it for thirty years.
Bill Fitzpatrick said:
true. you already know I have recently bi-amped myself, and the effect is very audible. As I can not hear the difference between amps in direct comparison, my guess is that the removal of the passive components caused the effect.
hummhoom said:
Reminded me of a site by the former director of the Aoustic Research departement of McIntosh.
http://www.roger-russell.com/
the most interesting info on the following:
http://www.roger-russell.com/truth/truth.htm
"I have personally completed several blind A-B listening tests over the years between good amplifiers, tube or transistor. Although I thought I could hear a difference each time, my choice was only correct about 50% of the time. I have also conducted blind listening tests for other people. I have learned how important it is to set the amplifier gains to be exactly equal and that the amplifiers should not be seen or identified for the listener. The slightly louder amplifier often is preferred. Comparison must be instantaneous or the listener forgets. If the identity of the amplifiers is known, the listener often gets preoccupied with identifying which amplifier is playing instead of the sound quality. The questions asked of the listener about the sound quality are also very important. I even hide the speakers as well as the amplifiers behind an acoustically transparent curtain."
I think this guy should know!
Konnichiwa,
Especially by the ABX Mafia and those "all equipment sound the same" perachers, who have for deades very effectively and actively prevented any serious investigation in the ACTUALLY EXISTING differences in sound caused by comparably small differences in equipement not covered by traditional measurement.
Much polemic cut.
As a direct result of the efforts of people like you. There are many who do hear real differences. As the mainstream engineering community refuses to aknowledge them and as they and people like you simply deny the existence of what they cannot morally afford to admit exists no systematic investigation takes place.
Thus the door remains wide open for any number of claims, which may result to a given individual in subjectively percieved changes (real or not is irelevant) and given that the mainstream has already pelicanised the whole subject yet Joe Public hears the difference anyway he will reject "your Philosophy" and adopt that pandered by the salesmen.
Actually, WRONG in both cases. The items you note will all sound different to different degrees. What will sound "better" to a given person is a question of preference.
That is your experience, it makes it far from universally applicable and does not invalidate any experiences to the contrary others may have had.
Also, have you considered that you where decieved by your own expectations, meaning you expected that all gear sounds the same and thus it did (BTW, this will also hold up under blind conditions)?
Why, should I not buy within my budget whatever sounds best to me? Have you found that there is a reliable law that the cheaper the gear the better it sounds, if so would you mind publishing the evidence for that?
But you are, just like any other selfrighteous preachers.
And now you choose to believe the opposite. Still just following others instead of making your own mind up.
DB Test and/or ABX testing is subject to many vagaries. I once amused myself by prooving this to an Objectivist and ABX Test advocate. I told him we would blind test Cables. He of course KNEW that cables cannot make a difference.
What I ACTUALLY tested was the polarity reversal in one channel. He was the only one in the test group of 4 people who scored "random", everyone else scored 100%. He simply (subconciously) refused to hear the obvious difference as according to his beliefsystem it should not exist. Therefore he "scored" no significance difference disregardless of the actual situation.
The same effect also works the other way. If you have a strong conviction that a difference exist you will tend to hear even if it does not exist and thus will again score "no significant difference" disregardless of the actual facts.
You may believe whatever you like. I have no room for irrational believes, I prefer to know.
And again, what where your expectations? Did you expect a difference? Did you expect non?
Why? I repeat, blind testing is really problematic usually do statistical problems (small datasets) and the fact that people,tend to know what they "blind test", as a result the actual is NOT blind. Therefore it is poor judgement to place faith in such flawed data.
Sayonara
keyser said:
Especially by the ABX Mafia and those "all equipment sound the same" perachers, who have for deades very effectively and actively prevented any serious investigation in the ACTUALLY EXISTING differences in sound caused by comparably small differences in equipement not covered by traditional measurement.
keyser said:
Much polemic cut.
As a direct result of the efforts of people like you. There are many who do hear real differences. As the mainstream engineering community refuses to aknowledge them and as they and people like you simply deny the existence of what they cannot morally afford to admit exists no systematic investigation takes place.
Thus the door remains wide open for any number of claims, which may result to a given individual in subjectively percieved changes (real or not is irelevant) and given that the mainstream has already pelicanised the whole subject yet Joe Public hears the difference anyway he will reject "your Philosophy" and adopt that pandered by the salesmen.
keyser said:
Actually, WRONG in both cases. The items you note will all sound different to different degrees. What will sound "better" to a given person is a question of preference.
keyser said:
That is your experience, it makes it far from universally applicable and does not invalidate any experiences to the contrary others may have had.
Also, have you considered that you where decieved by your own expectations, meaning you expected that all gear sounds the same and thus it did (BTW, this will also hold up under blind conditions)?
keyser said:
Why, should I not buy within my budget whatever sounds best to me? Have you found that there is a reliable law that the cheaper the gear the better it sounds, if so would you mind publishing the evidence for that?
keyser said:
But you are, just like any other selfrighteous preachers.
keyser said:
And now you choose to believe the opposite. Still just following others instead of making your own mind up.
keyser said:
DB Test and/or ABX testing is subject to many vagaries. I once amused myself by prooving this to an Objectivist and ABX Test advocate. I told him we would blind test Cables. He of course KNEW that cables cannot make a difference.
What I ACTUALLY tested was the polarity reversal in one channel. He was the only one in the test group of 4 people who scored "random", everyone else scored 100%. He simply (subconciously) refused to hear the obvious difference as according to his beliefsystem it should not exist. Therefore he "scored" no significance difference disregardless of the actual situation.
The same effect also works the other way. If you have a strong conviction that a difference exist you will tend to hear even if it does not exist and thus will again score "no significant difference" disregardless of the actual facts.
keyser said:
You may believe whatever you like. I have no room for irrational believes, I prefer to know.
keyser said:
And again, what where your expectations? Did you expect a difference? Did you expect non?
keyser said:
Why? I repeat, blind testing is really problematic usually do statistical problems (small datasets) and the fact that people,tend to know what they "blind test", as a result the actual is NOT blind. Therefore it is poor judgement to place faith in such flawed data.
Sayonara
konnichiwa to you too!
We obviously look at things differently. It really is not my intention to offend you. If I did, I am sorry. Offending anyone is not my reason to write here. My intention is to start a dialogue about what is audible, and what is not. Rather dialogue than discussion.
In my experience well designed equipment sounds the same. No, I have not tested all these things I do indeed "believe" by myself. I believe in science. I believe what the experts test, if they seem reliable and they know what they are talking about. I believe those things, as I believe that if I drop something it will fall to the ground. I have tested that by myself. But I also believe that satelites circle around the planet, even though I've never been there to look at them myself.
What do you think makes you hear differences between amps? Is it the slightly different THD, IMD or signal to noise ratio? Is it anything else that can be measured? Or do you think our ears are so sensitive that we hear things that are immeasurable?
How do you think Roger Russel could have ever designed amplifiers, the one sounding even greater than the other, if he himself was incapable of hearing the difference? Many people think McIntosh amplifiers are among the best sounding amps ever made. The designer can not hear it...
We obviously look at things differently. It really is not my intention to offend you. If I did, I am sorry. Offending anyone is not my reason to write here. My intention is to start a dialogue about what is audible, and what is not. Rather dialogue than discussion.
In my experience well designed equipment sounds the same. No, I have not tested all these things I do indeed "believe" by myself. I believe in science. I believe what the experts test, if they seem reliable and they know what they are talking about. I believe those things, as I believe that if I drop something it will fall to the ground. I have tested that by myself. But I also believe that satelites circle around the planet, even though I've never been there to look at them myself.
That is a good point. Then perhaps the only accurate DBT would be to let "believing" audiophiles test cables or equipment. Their prejudices would have little to no effect on the test.
What do you think makes you hear differences between amps? Is it the slightly different THD, IMD or signal to noise ratio? Is it anything else that can be measured? Or do you think our ears are so sensitive that we hear things that are immeasurable?
How do you think Roger Russel could have ever designed amplifiers, the one sounding even greater than the other, if he himself was incapable of hearing the difference? Many people think McIntosh amplifiers are among the best sounding amps ever made. The designer can not hear it...
To return to the original question, as posted by keyser, I am of the thought that if you are interested solely in the sound and are not taken in by the visual aspect nor hearsay then spend half your money on speakers and half on the rest.
My two sense (sic)
Thorsten: I believe you are correct in the points you made to keyser but I also conclude you have a lot of time on your hands. Sayonara.
Cal
My two sense (sic)
Thorsten: I believe you are correct in the points you made to keyser but I also conclude you have a lot of time on your hands. Sayonara.
Cal
Konnichiwa
So do many other people.
I am really not offended. I do not get offended when others peddle their faithbased worldviews. I merely debunk them... ;-)
You started a "dialogue" by making ex cathedra pronouncements as to what is audible and what is not. You pronounced that people should just buy cheap stereo gear as it all sounded the same. You left no room for discussion or dialogue, excet allowing agreement or diagreement, sort of like in this little "discussion":
K: "The moon is made of green Cheese!"
KYW: "No, the moon is not made of green chesse. It is a theory often advanced but rather wrong (goes on to list objections)..."
K: "The moon is made of green Cheese!"
It usally plays on from here something like this:
Loop until KYW gives up
KYW: "Now look, there are loads of perfectly good arguments and why the moon is not made of green chesse. True, we will not know for sure untill we fly up there personally and one may very well question that the Nasa Monn Flights ever took place, but really, in the larger scheme of things a moon made of green cheese makes little sense."
K: "The moon is made of green cheese!"
Do Loop
Well, in the experience of many others it does not.
Well, you are welcome to ytour religion, but quite frankly, I would not call it science, not even christian science. It is what I call "purpose opbjectivism", eg. something that pretends an objective view for a purpose, but is highly counterfactual in nature.
So, if I where to demonstrate to you that your believes where indeed inaccuratly in one specific item, you would flip again and become a "audio voodoo believer" again?
Do you realy lack any understanding so profoundly?
Okay, here the simple version:
Subject 1 is a Disbeliever - he knows there cannot be an audible difference
Subject 2 is a Believer - he knows there must be an audible difference
The likelyhood of either of them scoring "random" in any ABX or DB Test is practically 100%, because each will hear what they want to hear. Subject 1 will hear no difference if one exists and subject 2 will hear a difference even if non exists.
If our experimenter had more "different" trials in his test than "same" in fact the data will show that an audible difference exists with Subject 2 and that Subject 1 has tinears.
If he had more "same" trials in his test than "different" his results will show that that Subject 1 hears correctly no difference and Subject 2 is delusional.
Note, all of this will happen COMPLETELY DISREGARDLESS of any actual audibility.
Therefor any ABX or DB Test can only be considered of ANY relevance is such bias is eliminated, in other words if the subjects are unaware of the test itself.
I know in many cases what makes the diofference. In others I'm guessing.
The issue is neither that the ear is more sensitive than the best measurement gear or not. Nor is it if what is audible can be measured or not. Potentially ANYTHING (even unreal items) can be quantified and thus measured.
What is key to know is that the way the human hearing works is very complex, very non-linear and in peculiar ways, or in other words the way we HEAR is very different from the way we MEASURE.
Most of the commonly quoted measurements (not just in audio) as a measure of quality are one-dimensional numbers, taken under certain artificial conditions.
A non Audio Example are such items as Break-Horsepower, Miles per Gallon and Top Speed in cars. One might expect that car with the highste quooted BHP and top speed for the highest MPG would be the best car overall. It rarely is, even with rigidly standardised and independent tests for said measurements.
And of course, non of these numbers remotely express the fun you get from driving a poor MPG, low BHP and low Topspeed old British Sportster or a Gas Guzzling Turbo charged racing trim Trans Am, when compared to the lastet dullard car from mitsuyota or toyobishi et al....
Well, according to you quoting him him he did not hear any difference under blind conditions so there cannot have been one, surely? So therefor all MacIntosh Amplifiers sounded the same.
Only the very early ones (with Valves and such), actually.
Finally, I will agree that most HiFi gear sounds rather similar to me, similar bad and unable to bring music alive. I find generally that there is NO correlation whatsoever between subjective sound quality and price (eg $ 10,000 CD Player better than $ 100 one) similar or between subjective sound and fundamental technologies (tubes vs transistor etc).
I do find very much that the differences accrue from specific implementation details which are usually driven from a good understanding of what makes measurements correlate poorely with "good sound". Some of these differences are easily measured, others are not "caught" with traditional measurements but can be quantified as well.
Sayonara
keyser said:
So do many other people.
keyser said:
I am really not offended. I do not get offended when others peddle their faithbased worldviews. I merely debunk them... ;-)
keyser said:
You started a "dialogue" by making ex cathedra pronouncements as to what is audible and what is not. You pronounced that people should just buy cheap stereo gear as it all sounded the same. You left no room for discussion or dialogue, excet allowing agreement or diagreement, sort of like in this little "discussion":
K: "The moon is made of green Cheese!"
KYW: "No, the moon is not made of green chesse. It is a theory often advanced but rather wrong (goes on to list objections)..."
K: "The moon is made of green Cheese!"
It usally plays on from here something like this:
Loop until KYW gives up
KYW: "Now look, there are loads of perfectly good arguments and why the moon is not made of green chesse. True, we will not know for sure untill we fly up there personally and one may very well question that the Nasa Monn Flights ever took place, but really, in the larger scheme of things a moon made of green cheese makes little sense."
K: "The moon is made of green cheese!"
Do Loop
keyser said:
Well, in the experience of many others it does not.
keyser said:
Well, you are welcome to ytour religion, but quite frankly, I would not call it science, not even christian science. It is what I call "purpose opbjectivism", eg. something that pretends an objective view for a purpose, but is highly counterfactual in nature.
keyser said:
So, if I where to demonstrate to you that your believes where indeed inaccuratly in one specific item, you would flip again and become a "audio voodoo believer" again?
keyser said:
Do you realy lack any understanding so profoundly?
Okay, here the simple version:
Subject 1 is a Disbeliever - he knows there cannot be an audible difference
Subject 2 is a Believer - he knows there must be an audible difference
The likelyhood of either of them scoring "random" in any ABX or DB Test is practically 100%, because each will hear what they want to hear. Subject 1 will hear no difference if one exists and subject 2 will hear a difference even if non exists.
If our experimenter had more "different" trials in his test than "same" in fact the data will show that an audible difference exists with Subject 2 and that Subject 1 has tinears.
If he had more "same" trials in his test than "different" his results will show that that Subject 1 hears correctly no difference and Subject 2 is delusional.
Note, all of this will happen COMPLETELY DISREGARDLESS of any actual audibility.
Therefor any ABX or DB Test can only be considered of ANY relevance is such bias is eliminated, in other words if the subjects are unaware of the test itself.
keyser said:
I know in many cases what makes the diofference. In others I'm guessing.
The issue is neither that the ear is more sensitive than the best measurement gear or not. Nor is it if what is audible can be measured or not. Potentially ANYTHING (even unreal items) can be quantified and thus measured.
What is key to know is that the way the human hearing works is very complex, very non-linear and in peculiar ways, or in other words the way we HEAR is very different from the way we MEASURE.
Most of the commonly quoted measurements (not just in audio) as a measure of quality are one-dimensional numbers, taken under certain artificial conditions.
A non Audio Example are such items as Break-Horsepower, Miles per Gallon and Top Speed in cars. One might expect that car with the highste quooted BHP and top speed for the highest MPG would be the best car overall. It rarely is, even with rigidly standardised and independent tests for said measurements.
And of course, non of these numbers remotely express the fun you get from driving a poor MPG, low BHP and low Topspeed old British Sportster or a Gas Guzzling Turbo charged racing trim Trans Am, when compared to the lastet dullard car from mitsuyota or toyobishi et al....
keyser said:
Well, according to you quoting him him he did not hear any difference under blind conditions so there cannot have been one, surely? So therefor all MacIntosh Amplifiers sounded the same.
keyser said:
Only the very early ones (with Valves and such), actually.
Finally, I will agree that most HiFi gear sounds rather similar to me, similar bad and unable to bring music alive. I find generally that there is NO correlation whatsoever between subjective sound quality and price (eg $ 10,000 CD Player better than $ 100 one) similar or between subjective sound and fundamental technologies (tubes vs transistor etc).
I do find very much that the differences accrue from specific implementation details which are usually driven from a good understanding of what makes measurements correlate poorely with "good sound". Some of these differences are easily measured, others are not "caught" with traditional measurements but can be quantified as well.
Sayonara
Hi!
You don't like DB testing. Ok. How do YOU test equipment then? And what difference can you hear between (lets stick to them) amplifiers? Double-Blind tests are done in all fields of science. Why would it not be valid for hifi?
Why are "high end" cables only seen in the world of hifi, and not in for example hospitals, used for the tools surgeons use during surgery.
Can you explain to me why everybody agrees that loudspeakers sound different from each other? In a DB test I could not imagine to ever be wrong if different loudspeakers were being tested. Why are people not sure about other parts in the hifi chain?
One of the first DB tests I ever did, was an amp test. As a "audio voodoo believer" I expected to hear a huge difference between a friends' quite expensive valve-amp (Omak Jewel Three), and my medium priced NAD 216 THX. Why did I not hear a difference?
As I lack any understanding so profoundly!...
What are you actually trying to say? What measurable parameters are important, and what parameters that may perhaps be quantifiable but then again not measurable are important?
Would you please translate your automobile example to an amplifier example?
keyser
You don't like DB testing. Ok. How do YOU test equipment then? And what difference can you hear between (lets stick to them) amplifiers? Double-Blind tests are done in all fields of science. Why would it not be valid for hifi?
Why are "high end" cables only seen in the world of hifi, and not in for example hospitals, used for the tools surgeons use during surgery.
Can you explain to me why everybody agrees that loudspeakers sound different from each other? In a DB test I could not imagine to ever be wrong if different loudspeakers were being tested. Why are people not sure about other parts in the hifi chain?
One of the first DB tests I ever did, was an amp test. As a "audio voodoo believer" I expected to hear a huge difference between a friends' quite expensive valve-amp (Omak Jewel Three), and my medium priced NAD 216 THX. Why did I not hear a difference?
As I lack any understanding so profoundly!...
What are you actually trying to say? What measurable parameters are important, and what parameters that may perhaps be quantifiable but then again not measurable are important?
Would you please translate your automobile example to an amplifier example?
keyser
Konnichiwa,
How do you figure that? I am saying that there are a number of issues one need to observe to make sure the results from a DB test are meaningful. In fact, DB Testing is just another measurement methode, measurements require care to make sure you measure what you want to measure and interpretation.
The way I do. The specifics are of no particular concern, as I aim at specifc things, with specific methodes that relate very littel to the Audio mainstream. Note I am not saying "I am right and everyone else is wrong", only "I do things different from almost anyone else".
DB Tests are valid ONLY if done correctly and within limits inherent to the methode, the size of dataset and many other limitations. This is true for any scientific DB Test outside audio. What in audio makes a pisspoor implemented, inaccuratly analysed and generally biased DB Test less worth of criticism and rejection than in any other field?
Funnily enough, they are. My source for "fancy" ptfe clad silver wire happen to be a Biomed supply house. RF Systems and even computer networks make use and or offer what might be termed "High End" or "premium grade" cables readilly. Take the example of RG-213 and RG-214, electrically they are almost idetical, certainly well below measurement limits yet the RG-214 uses "permium grade" materials and is several times as expensive.
Sure. Because it is 1) generally agreed and undisputed that they do and 2) they sound different - so we have no conflict between reality and dogma, anyone can relax and admit speaker sound different.
YET, if you where told what was tested where $ 10,000 Cables against $ 5,000 Cables but IN REALITY what was tested where different Speakers with similar, but aknowledged different sound? Would you still score 100%? Think carefully. I found that if sufficiently motivated "objectivists" fail to hear reliably the most blatant differences. You just need need to tell them that what they are hearing is something they vehemntly oppose as being possibly audible.
Actually, what you heard (as I pointed out) played no role. Your expectations did. You expect differences so you heard them even if there where non, you looked at your score sheet and voila, no significant difference.
The measurements traditionally standardised and usually quoted are incapable of describing "good sound" in fact, they show no correlation whatsoever once certain (very basic) minimal quality levels are reached.
There are no simple parameters. That is my point. You need to measure more in the manner the human subject hears. In order to do that you need to understand the hearing process. I have no desire to tech you, inform yourself please.
Amplifier 1 has 0.01% THD @ 100W
Amplifier 2 has 1% THD @ 20W
The above numbers have no relation whatsoever as to which one you will prefer and why. If Sony build Amplifier 1 and I did build Amplifier 2, odds are you more people would find Amplifier 2 enjoyable listening.
Sayonara
keyser said:
How do you figure that? I am saying that there are a number of issues one need to observe to make sure the results from a DB test are meaningful. In fact, DB Testing is just another measurement methode, measurements require care to make sure you measure what you want to measure and interpretation.
keyser said:
The way I do. The specifics are of no particular concern, as I aim at specifc things, with specific methodes that relate very littel to the Audio mainstream. Note I am not saying "I am right and everyone else is wrong", only "I do things different from almost anyone else".
keyser said:
DB Tests are valid ONLY if done correctly and within limits inherent to the methode, the size of dataset and many other limitations. This is true for any scientific DB Test outside audio. What in audio makes a pisspoor implemented, inaccuratly analysed and generally biased DB Test less worth of criticism and rejection than in any other field?
keyser said:
Funnily enough, they are. My source for "fancy" ptfe clad silver wire happen to be a Biomed supply house. RF Systems and even computer networks make use and or offer what might be termed "High End" or "premium grade" cables readilly. Take the example of RG-213 and RG-214, electrically they are almost idetical, certainly well below measurement limits yet the RG-214 uses "permium grade" materials and is several times as expensive.
keyser said:
Sure. Because it is 1) generally agreed and undisputed that they do and 2) they sound different - so we have no conflict between reality and dogma, anyone can relax and admit speaker sound different.
keyser said:
YET, if you where told what was tested where $ 10,000 Cables against $ 5,000 Cables but IN REALITY what was tested where different Speakers with similar, but aknowledged different sound? Would you still score 100%? Think carefully. I found that if sufficiently motivated "objectivists" fail to hear reliably the most blatant differences. You just need need to tell them that what they are hearing is something they vehemntly oppose as being possibly audible.
keyser said:
Actually, what you heard (as I pointed out) played no role. Your expectations did. You expect differences so you heard them even if there where non, you looked at your score sheet and voila, no significant difference.
keyser said:
The measurements traditionally standardised and usually quoted are incapable of describing "good sound" in fact, they show no correlation whatsoever once certain (very basic) minimal quality levels are reached.
keyser said:
There are no simple parameters. That is my point. You need to measure more in the manner the human subject hears. In order to do that you need to understand the hearing process. I have no desire to tech you, inform yourself please.
keyser said:
Amplifier 1 has 0.01% THD @ 100W
Amplifier 2 has 1% THD @ 20W
The above numbers have no relation whatsoever as to which one you will prefer and why. If Sony build Amplifier 1 and I did build Amplifier 2, odds are you more people would find Amplifier 2 enjoyable listening.
Sayonara
ABX testing is perfectly sound (pardon the pun) testing procedure. The people who object to it tend to be the ones who can't get results from it. Because you have a 50% chance of picking a certain device, the test must be repeated many times for a pattern to emerge.
Whenever a test fails to identify the component the reviewer wants to flatter, it's claimed that the test was faulty. When's the last time you read a properly conducted review in Stereophile?
The same goes for those who claim that THD, SNR, and IMD don't matter. The only ones who claim that numbers aren't everything are the ones who don't have the numbers. It's kind of like a fat kid trying to use the story of Achilles and the tortoise to claim the class jock can't beat him in a sprint.
The measure of well designed equipment is not that it sounds like other well designed equipment, but that it sounds as close to the oringal recording as possible.
Whenever a test fails to identify the component the reviewer wants to flatter, it's claimed that the test was faulty. When's the last time you read a properly conducted review in Stereophile?
The same goes for those who claim that THD, SNR, and IMD don't matter. The only ones who claim that numbers aren't everything are the ones who don't have the numbers. It's kind of like a fat kid trying to use the story of Achilles and the tortoise to claim the class jock can't beat him in a sprint.
The measure of well designed equipment is not that it sounds like other well designed equipment, but that it sounds as close to the oringal recording as possible.
Konnichiwa.
Excepting you side-swip at "the people who can't get results from it", you correct. ABX Testing per se is fine. You touched on thing though. You need a rather large dataset before you can make any call. Many of the published ABX/DB Tests which claim to have found no significant difference use too small datasets with inapropriate significance levels.
Also, if you use very many trials in a single session with many peopl the attentioon will wander. I recommend no more than 5 Trials per Individual per session and then a break.
Also, few if any ABX experimenters actually test their test (including room, speaker setup etc AND the subject) against known audible phenomenae to ensure at least the minimal sensitivity of the test "setup". If I made measurements with uncalibrated measurement gear and made any claims I would be quite rudely slapped down, why is the same permissable if someone waves the ABX Banner?
Finally, remember my contention that if the actual aim of the study is knonw and has a high potential emotional value for the participants the test will be useless again.
So, unless truely BLIND large scale tests are being done, similar to those used to establish the "degradation severity" for perceptual coding (lossy audio compression) I'm afraid all else is an activity most often inaccuratly attributed to the biblical figure of Onan.
Sayonara
hummhoom said:
Excepting you side-swip at "the people who can't get results from it", you correct. ABX Testing per se is fine. You touched on thing though. You need a rather large dataset before you can make any call. Many of the published ABX/DB Tests which claim to have found no significant difference use too small datasets with inapropriate significance levels.
Also, if you use very many trials in a single session with many peopl the attentioon will wander. I recommend no more than 5 Trials per Individual per session and then a break.
Also, few if any ABX experimenters actually test their test (including room, speaker setup etc AND the subject) against known audible phenomenae to ensure at least the minimal sensitivity of the test "setup". If I made measurements with uncalibrated measurement gear and made any claims I would be quite rudely slapped down, why is the same permissable if someone waves the ABX Banner?
Finally, remember my contention that if the actual aim of the study is knonw and has a high potential emotional value for the participants the test will be useless again.
So, unless truely BLIND large scale tests are being done, similar to those used to establish the "degradation severity" for perceptual coding (lossy audio compression) I'm afraid all else is an activity most often inaccuratly attributed to the biblical figure of Onan.
Sayonara
You don't need a particularly large dataset for the pattern to emerage. A reviewer of "golden ear" calibre ought to be able to choose one device over the other nine times out of ten. That's not a very fatigue inducing number of data points to gather. Mostly ABX testing will show that the revered golden ears are really made of lead.
Konnichiwa,
Please provide proof (statistical that is - lay out your new statistical methode which allows that) of your assertation.
I think I get what you are trying to say, but sadly in statistical temrs what you say has zip relevance. And that is before accounting for the second hidden variable in tests that are not fully blind, namely my point made above that people hear what they expect, not what is actually happening (this one cuts both ways).
Maybe, maybe not. I once had a CDR send to me to test my "golden ears". This contained 8 times the same exceprt of music, in some cases processed via various datarate MP3 in others straight from the wav. This prooved an interesting controvery.
My first conclusion after torturing myself to that point (and after the first 3 excerpts it became EXTREMELY tedious) was that NO excerpt matched the original CD's from which had extracted the original tracks with EAC in paranoid mode. Secondly I gave a "quality" rating where I arbitrarily rated the original (my own CD's) as 10 and then downwards from there.
As I noted that no tracks from the matched the original my challenger declared I had failed the test and could not hear any differences. Interestingly he refused to lay open which track was which.
Unbeknowned to that individual (a German sound guy who is paid by a academic institution for managing their sound system - eg not a very good sound guy) I had taken ALL excerpts from an HDCD encoded disk. As you might know, any digital manipulation of a CD extract via an ASRC or MP3 encoding/decoding effectively strips out the HDCD coding. At the time I had no HDCD capable player so I borrowed a friends and lo and behold, on all the tracks I had marked with a quality rating of 8 the HDCD light cam on.
So, in fact, not only had I not failed my "blind test", but I had found that I could reliably identify the non MP3 mangled copies, also to myself that there was something dodgy in how the CD was burned, as on my own PC copies usually sound the same or slightly better than originals, not notably worse.
Anyway, I found an 8 Trial ABX test rather trying, a 10 Trial one would be more so.
They will show nothing of such nature. If you understood elementary statistics you would know why (hint, extremely high rist of type 2 / type b statistical error due to excessively small sample size and excessively applied significance level).
Sayonara
hummhoom said:
Please provide proof (statistical that is - lay out your new statistical methode which allows that) of your assertation.
hummhoom said:
I think I get what you are trying to say, but sadly in statistical temrs what you say has zip relevance. And that is before accounting for the second hidden variable in tests that are not fully blind, namely my point made above that people hear what they expect, not what is actually happening (this one cuts both ways).
hummhoom said:
Maybe, maybe not. I once had a CDR send to me to test my "golden ears". This contained 8 times the same exceprt of music, in some cases processed via various datarate MP3 in others straight from the wav. This prooved an interesting controvery.
My first conclusion after torturing myself to that point (and after the first 3 excerpts it became EXTREMELY tedious) was that NO excerpt matched the original CD's from which had extracted the original tracks with EAC in paranoid mode. Secondly I gave a "quality" rating where I arbitrarily rated the original (my own CD's) as 10 and then downwards from there.
As I noted that no tracks from the matched the original my challenger declared I had failed the test and could not hear any differences. Interestingly he refused to lay open which track was which.
Unbeknowned to that individual (a German sound guy who is paid by a academic institution for managing their sound system - eg not a very good sound guy) I had taken ALL excerpts from an HDCD encoded disk. As you might know, any digital manipulation of a CD extract via an ASRC or MP3 encoding/decoding effectively strips out the HDCD coding. At the time I had no HDCD capable player so I borrowed a friends and lo and behold, on all the tracks I had marked with a quality rating of 8 the HDCD light cam on.
So, in fact, not only had I not failed my "blind test", but I had found that I could reliably identify the non MP3 mangled copies, also to myself that there was something dodgy in how the CD was burned, as on my own PC copies usually sound the same or slightly better than originals, not notably worse.
Anyway, I found an 8 Trial ABX test rather trying, a 10 Trial one would be more so.
hummhoom said:
They will show nothing of such nature. If you understood elementary statistics you would know why (hint, extremely high rist of type 2 / type b statistical error due to excessively small sample size and excessively applied significance level).
Sayonara
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.