AB Testing

[bhobba]

Bill They would rather have their teeth extracted without anaesthetic than be proved wrong ! (grin)

 

True - oh so true - but some like my mate Kdoot are a lot more honest - but they don't remain strict objectivists for long.

 

Thanks

Bill

[Audio_ELF]

True - oh so true - but some like my mate Kdoot are a lot more honest - but they don't remain strict objectivists for long.

Actually that would be the PROPER definition of a someone with an objective viewpoint on HiFi...

[bhobba]

More power to you... Though to be of relevant proof I would ask you (a) what the equipment used was and (b) what cables were being tested and © what the set up was (i.e. how were you blinded)

 

You see thats what I mean. Rather than accept the result you want to challenge details. Don't you think I am a reasonable enough person that when I was confronted with something I found surprising I would ensure the blind test was done correctly. It was a single blind test with the person conducting the audition changing the cables. Theoretically it is possible I was picking up cues from the person doing the test, however I have done enough blind wine tastings to know in practice it is not an issue.

 

Of course it is quite possible you will retort - invalid. Which just proves my point - the issue is what people are willing to accept as evidence and why each side retreats to their impregnable position - your test is not valid - yes it is and it gets - nowhere.

 

Just for the record the test was with a heavily modified Monachy PC1704 DAC feeding Mac 501's into Lenehan Audio ML1 Reference (fully tricked out with Duelund VSF Copper Capacitors and to reduce resonances the cabinets are lined with steel - they are the deadest things you can imagine and ultra revealing) comparing Lenehan Ribbontek speaker cables to Eichmann speaker cables. The Ribbontek had much more life - the Eichmann sounded dull.

 

Thanks

Bill

[Audio_ELF]

You see thats what I mean. Rather than accept the result you want to challenge details. Don't you think I am a reasonable enough person that when I was confronted with something I found surprising I would ensure the blind test was done correctly. It was a single blind test with the person conducting the audition changing the cables. Theoretically it is possible I was picking up cues from the person doing the test, however I have done enough blind wine tastings to know in practice it is not an issue.

Where did I suggest I wanted to challenge the validity of the test (for the record I actually agree that, to a certain extent, different cables can affect the sound). What I am suggesting is if you are trying to challenge the reasoning of others, you have to step up and answer their questions. Don't automatically get defensive. Most people actually ask further questions because they are interested, not because they are trying to ridicule your ideas - when the person being asked then starts getting defensive or spouting arguments which (to the person asking I will admit) make no sense, that's when the BS filters start kicking in...

 

Of course it is quite possible you will retort - invalid. Which just proves my point - the issue is what people are willing to accept as evidence and why each side retreats to their impregnable position - your test is not valid - yes it is and it gets - nowhere.

Well hopefully you will be pleasantly surprised that your test seems nicely valid.

 

Just for the record the test was with a heavily modified Monachy PC1704 DAC feeding Mac 501's into Lenehan Audio ML1 Reference (fully tricked out with Duelund VSF Copper Capacitors and to reduce resonances the cabinets are lined with steel - they are the deadest things you can imagine and ultra revealing) comparing Lenehan Ribbontek speaker cables to Eichmann speaker cables. The Ribbontek had much more life - the Eichmann sounded dull.

Thank you, now you see (at least to me) saying that is a lot more interesting and reasoned than saying "we compared two cables blind and sighted and they sound different".

 

Eloise

[bhobba]

Thank you, now you see (at least to me) saying that is a lot more interesting and reasoned than saying "we compared two cables blind and sighted and they sound different".

 

Fair enough.

 

Thanks

Bill

[chg]

I'm sorry - but this is the nature of Internet forums...

 

Everyone is entitled to post their views, and repeat them if they feel people are ignoring them or missing the point.

 

Eloise

 

I agree that "everyone is entitled to post their views, but I said "aggressively forcing views".

 

As for "self-professed" facts, when he is talking about (for example) the way MP3 is supposed to work - that isn't a self-professed fact, it's a fact of how the Fraunhofer Society intended MP3 to work when they created it.

 

A bit of a cherry picked example to fit your point. I never criticized his MP3 discussion! Obviously you've not been paying attention when the "facts" were a flyin' in the past.

 

With respect: I'm beginning to think that you just like to reply for the sake of it, or just to get the last word in. Before you just rush to reply, you should maybe contemplate a bit longer. Quality over quantity.

[chg]

Bill - with respect I believe your reply totally missed my point, yet also made it.

 

A test is set up (and I'm talking about sighted tests here) to compare an aspect of an audio system which when looked at from an objective point of view. Everyone agrees that the test should show a difference, the person X says the difference is "night and day" while person Y argues that there should be no difference. The test is carried out on equipment which Person Y chooses. I think you will agree this is a typical situation.

 

Okay, so they do the test, swapping back and forward to compare the two situations. Each time doing a simple A-B; B-A, A-A or B-B comparison and the question is asked "is there a difference?" Person Y listens and each time says if there is a difference or not. The time is spent quite amicably and relaxed.

 

At the end of the test, Person Y feels they haven't heard a difference.

 

There are several possible results to this.

1) There is a difference but Y doesn't notice it because they are deaf or untrained.

2) There is a difference but Y doesn't notice it because the equipment wasn't resolving enough to notice the differences.

3) There is no difference even though X feels he has heard it.

 

In my experience there are very few people with a subjective outlook who will accept that (3) is a possibility. That's all I'm asking for. In every occurrence of sighted testing, you have to accept that it is possible that any difference noticed. Just as the subjectives call for the objectives to accept that they don't have all the answers or can't measure all the differences.

 

Eloise

 

There's a bit of flawed logic here:

 

"Currently science cannot measure the output" from post 117.

 

"3) There is no difference even though X feels he has heard it."

 

In my experience there are very few people with a subjective outlook who will accept that (3) is a possibility.

 

In my experience there are very few people with a objective outlook who will accept that 1 and 2 are a possibility.

 

In my experience there are very few people with a subjective outlook who will accept that (3) is a possibility.

 

I don't know why some think this. I believe that all subjectivists here accept that it's a possibility. But because it's a possibility, does that mean we set up a lab and devote every waking hour trying to determine the validity of every perception? And then there's this bit of a dilemma to deal with "currently science cannot measure the output" which ensures one will never find the answer anyway. The sane person, and those without an agenda, simply goes with what he/she hears, spends a few bucks, and gets on with his/her life.

[chg]

Note *: Though in actual face the true objective accepts that there are something that can't be measured and balances listening testing with scientific theory and measurement - that's what being objective means...

 

If that's the case, the true objectivist is a rare breed around here.

 

That description can also be used to describe a subjectivist with common sense.

[Paul R]

One could far more accurately say you have tunnel blindness in this area.

 

You continue to insist that only information that is inaudible is removed from MP3s and therefore, the theory behind them is correct. When it is pointed out the information removed causes audible changes, you just claim the theory isn't perfect, but is still perfectly correct.

 

It is like being pregnant, either you are or you are not. Either the information is inaudible or it is not. Obviously and quite clearly, the information removed is audible. Just as obviously, that means the confidence in the theory behind MP3s is very small. Or in other words, you can keep repeating you are right and throw nasty cracks all you want. Evidence would be better, as I already said you could convince me with evidence.

 

You do indeed keep claiming the theory works exactly as specified in regard to MP3s.

 

Saying it is imperfect does not somehow magically make the theory "mostly right." It does not work that way- if the theory predicts some results, and this one does, and the results do not match what the theory predicts very precisely, the theory is simply wrong and must be abandoned or modified to produce predictions that agree with the results. Otherwise it isn't science, it is politics. Dishonest politics at that.

 

Lets see some of that evidence you are so keen on. You insist on claiming a flawed theory is perfectly correct, then claim it is imperfect, and have no problem telling other people they are wrong. Stones, glass houses.

 

More accurately, you are embarassing yourself with such sloppy thinking. Very very sloppy thinking, and it appears you are engaging in such sloppy thinking only becase you want this theory to be true, without strong evidence for your case, and in the face of very hard to dispute evidence against it.

 

 

Paul

 

 

I would think you would realize that there is no one explanation for audio phenomena like this. Saying the only operator involved with the audible signature of a MP3 is masking is just plain wrong Dennis.

 

I agree which is why I never said that.

 

Perhaps you know of a study where MP3s have been implemented with perfect reproduction. If so, I'll reconsider my thinking. But you are making an extraordinary claim here, and it will require extraordinary proof. Especially with a preponderance of evidence against it.

 

Perhaps you could read the very passage of mine you quoted here where it says, and I quote myself,"nor is anyone claiming it works perfectly."

 

One could say you were building an extraordinary straw man. But it might be more of the Calico Cat and the Gingham Dog variety. The old Dutch clock is certainly easy to believe in.

 

CatStuff: The Gingham Dog and the Calico Cat

[Paul R]

That is very well said Bill, very well said indeed. -Paul

[Audio_ELF]

That is very well said Bill, very well said indeed. -Paul

Sorry... Who is Bill?

 

Eloise

[Paul R]

3) There is no difference even though X feels he has heard it.

 

In my experience there are very few people with a subjective outlook who will accept that (3) is a possibility. That's all I'm asking for. In every occurrence of sighted testing, you have to accept that it is possible that any difference noticed. Just as the subjectives call for the objectives to accept that they don't have all the answers or can't measure all the differences.

 

Eloise

 

Aiii yiii yii!

 

Um, that is pretty clearly exactly what this is about. Dennis hears a difference, then claims it is not really there even though he is sure he has heard it.

 

Your option 3 is exactly the point.

 

Paul

[Paul R]

Sorry Eloise!

This Bill, and this comment. Sorry, I thought I quoted it. Still very well said indeed, IMO. -Paul

 

Of course. They realize the issue though - if even one person can hear a difference then their argument is kaput. That's why guys like me are attacked relentlessly when we post a blind test detected this or that difference - it must be a DBT, it must be volume matched by oscilloscope, the person involved has commercial interests, - the list goes on. Its in fact for the very heart and soul of Hi Fi - they know it - which is why arguments about it go on and on but get nowhere with each side retreating to their impregnable position.

 

Thanks

Bill

[chg]

Bill

They would rather have their teeth extracted without anaesthetic than be proved wrong ! (grin)

Alex

 

He heard clear difference is USB cables, and while flabbergasted, does not go into denial like objectivists do.

 

When you see this type of objectivist denial behavior, that's when you know an agenda exists.

[Julf]

Sorry... Who is Bill?

 

[Audio_ELF]

It is like being pregnant, either you are or you are not.

Actually its possible for your body to act in a way as if you are pregnant even if you are not physically pregnant - perhaps this makes it a good analogy to HiFi.

 

Just a minute ... shouldn't this be on this thread -- http://www.computeraudiophile.com/f8-general-forum/sexual-preference-audio-13314/

 

Eloise

[4est]

Actually its possible for your body to act in a way as if you are pregnant even if you are not physically pregnant - perhaps this makes it a good analogy to HiFi.

 

Just a minute ... shouldn't this be on this thread -- http://www.computeraudiophile.com/f8-general-forum/sexual-preference-audio-13314/

 

Eloise

 

Talk about extremes... When an example needs to go this far, it makes me wonder about option 3 being at play.

[Audio_ELF]

Talk about extremes... When an example needs to go this far, it makes me wonder about option 3 being at play.

Sorry... I missed a smiley or two and obviously my humour isn't appreciated :-)

[Paul R]

Rotflmao! :)

 

 

 

 

actually its possible for your body to act in a way as if you are pregnant even if you are not physically pregnant - perhaps this makes it a good analogy to hifi.

 

Just a minute ... Shouldn't this be on this thread -- http://www.computeraudiophile.com/f8-general-forum/sexual-preference-audio-13314/

 

eloise

[jhwalker]

One could far more accurately say you have tunnel blindness in this area.

 

You continue to insist that only information that is inaudible is removed from MP3s and therefore, the theory behind them is correct. When it is pointed out the information removed causes audible changes, you just claim the theory isn't perfect, but is still perfectly correct.

 

It is like being pregnant, either you are or you are not. Either the information is inaudible or it is not. Obviously and quite clearly, the information removed is audible. Just as obviously, that means the confidence in the theory behind MP3s is very small. Or in other words, you can keep repeating you are right and throw nasty cracks all you want. Evidence would be better, as I already said you could convince me with evidence.

 

You do indeed keep claiming the theory works exactly as specified in regard to MP3s.

 

Saying it is imperfect does not somehow magically make the theory "mostly right." It does not work that way- if the theory predicts some results, and this one does, and the results do not match what the theory predicts very precisely, the theory is simply wrong and must be abandoned or modified to produce predictions that agree with the results. Otherwise it isn't science, it is politics. Dishonest politics at that.

 

Lets see some of that evidence you are so keen on. You insist on claiming a flawed theory is perfectly correct, then claim it is imperfect, and have no problem telling other people they are wrong. Stones, glass houses.

 

More accurately, you are embarassing yourself with such sloppy thinking. Very very sloppy thinking, and it appears you are engaging in such sloppy thinking only becase you want this theory to be true, without strong evidence for your case, and in the face of very hard to dispute evidence against it.

 

 

Paul

 

Not really sure what you're getting at, Paul. MP3 is a lossy solution. NO ONE claims it is perfect, or that there are no problematic tracks that cannot be perfectly reproduced, no matter how high you drive the bitrate. The theory is, indeed, that you simply omit sounds that are masked. The result should be very difficult to distinguish from the original - and that's exactly what MP3 at 192k or above is. So the theory is perfectly correct and it produces results that are imperfect but very good

[Paul R]

Not really sure what you're getting at, Paul. MP3 is a lossy solution. NO ONE claims it is perfect, or that there are no problematic tracks that cannot be perfectly reproduced, no matter how high you drive the bitrate. The theory is, indeed, that you simply omit sounds that are masked. The result should be very difficult to distinguish from the original - and that's exactly what MP3 at 192k or above is. So the theory is perfectly correct and it produces results that are imperfect but very good

 

HI John - to be precise, I am saying the codec reference tables are faulty. That is part of the reason why claiming what has been removed was "inaudible" in the first place is false and rather misleading. When whichever particular algorithm you use to encode the MP3 performs masking operations on one or more of the sub bands, based upon that reference table and the intensity of any particular sample, you don't get the expected results.

 

Given all that, the end result - a MP3 encoded file - sounds pretty good. But people did initially claim that MP3s were indistinguishable from 16/44.1 CD quality recordings - because the information that is masked out is inaudible anyway. Everyone knows that isn't true, but some people will try to push that old saw through yet another log. You still find people trying to claim that, and poo poohing audiophiles because no audiophile I know is believes that. Indeed, you can find that argument currently being bandied about by "experts" with nothing more than a simple google search.

 

I am probably being a bit pedantic about it, but I deplore supercilious attitudes like that.

[esldude]

HI John - to be precise, I am saying the codec reference tables are faulty. That is part of the reason why claiming what has been removed was "inaudible" in the first place is false and rather misleading. When whichever particular algorithm you use to encode the MP3 performs masking operations on one or more of the sub bands, based upon that reference table and the intensity of any particular sample, you don't get the expected results.

 

Given all that, the end result - a MP3 encoded file - sounds pretty good. But people did initially claim that MP3s were indistinguishable from 16/44.1 CD quality recordings - because the information that is masked out is inaudible anyway. Everyone knows that isn't true, but some people will try to push that old saw through yet another log. You still find people trying to claim that, and poo poohing audiophiles because no audiophile I know is believes that. Indeed, you can find that argument currently being bandied about by "experts" with nothing more than a simple google search.

 

I am probably being a bit pedantic about it, but I deplore supercilious attitudes like that.

 

Simply ridiculous Paul. Talk about supercilious attitude. There might be some people somewhere claiming anything you can imagine. The people who developed MP3 never said it was indistinguishable from the full lossless files it was derived from.

 

As for expected results, yeah, the people who developed it got the results they expected. You are the one who seems to have this need to build up this unmet expectation. So you can then claim it was unsuccessful. Considering at 128 kbps throws out 90% of the bits, it darn sure sounds lots more than 10% like the original. Which indicates it works pretty well as expected by the designers of it.

 

You are building up this artificial construct of MP3 encoding being perfect so you can then say "look, anyone can hear it isn't perfect." Sounds nice other than it is pointless and gets one nowhere. And you did all this rather than simply admit, MP3's are predicated upon a masking model instead of the filling in the holes model put forth by chg in this thread. Really pedantic and punctilious on your part.

[Zakus]

Given all that, the end result - a MP3 encoded file - sounds pretty good. But people did initially claim that MP3s were indistinguishable from 16/44.1 CD quality recordings - because the information that is masked out is inaudible anyway. Everyone knows that isn't true, but some people will try to push that old saw through yet another log. You still find people trying to claim that, and poo poohing audiophiles because no audiophile

 

You only have to look at hydrogenaudio for that claim to be presented as objective.

 

Edit spelling

 

Further edit Esldude got in before me. My comment stems from one of those long night forum debates where the other party kept throwing up hydrogenaudio abx posts as objective evidence that there is no difference when you listen to mp3 to 16bit let alone 24bit.

[chg]

MP3's are predicated upon a masking model instead of the filling in the holes model put forth by chg in this thread.

 

Whoa! I never got involved in this MP3 banter! But I can see where it may have appeared that way.

 

In addressing you much earlier I did improperly include your most recent quote (at the time), but only to reference that I was addressing you. My view that followed wasn't regarding the subject matter (MP3) specifically at the time. I should have simply started my post without quote and only included your name. My mistake.

[esldude]

Whoa! I never got involved in this MP3 banter! But I can see where it may have appeared that way.

 

In addressing you much earlier I did improperly include your most recent quote (at the time), but only to reference that I was addressing you. My view that followed wasn't regarding the subject matter (MP3) specifically at the time. I should have simply started my post without quote and only included your name. My mistake.

 

You are correct chg. The origination of it I attributed to you was actually from spdif-usb. Sorry about that.