Tom Petty Hi Res Tracks

[manicm]

I did not do blind testing.

Your answer.

Ohhhhhhhh, so that necessarily makes my hearing wrong now doesn't it? Time to move on Steve. And don't give me a argument about 'expectation bias', because this was not primarily an experiment on my part, I had to downsample out of necessity, so I WANTED it to sound equally good, there's a corollary to many things that will not prove arguments to be absolute.

 

[chebby]

That's not how expectipection bias works.

How does 'expectipection bias' work?

What is 'expectipection'?

I've heard 'expectoration' in music if this is what you mean ...

https://www.youtube.com/watch?v=BhNXJGmcqNI

 

[steve_1979]

How does 'expectipection bias' work?

*unknw*

One of these days I may learn how to proof read before clicking the 'save' buttin.

 

[fr0g]

I did not do blind testing.

Your answer.

Ohhhhhhhh, so that necessarily makes my hearing wrong now doesn't it? Time to move on Steve. And don't give me a argument about 'expectation bias', because this was not primarily an experiment on my part, I had to downsample out of necessity, so I WANTED it to sound equally good, there's a corollary to many things that will not prove arguments to be absolute.

If you do not ABX then your view can be dismissed.

And yes, you do have a problem with your hearing, the same one that all humans have, that it can be affected by your other senses.

The ONLY way to be sure you can hear a difference is to run the files through an ABX test.

And you have the tools. So why not do it?

And just to really blow your mind, if you drum up the courage to do it, I would bet you can't tell the difference between HD and CD quality, HD and 320 Kbps MP3, HD and 190 Kbps MP3 and I wouldn't be surprised if you couldn't tell the difference down to 128 Kbps MP3, although at that level, some tracks become more obvious.

Give it a whirl.

 

[matt49]

And yes, you do have a problem with your hearing, the same one that all humans have, that it can be affected by your other senses.

I'm not sure why you keep on saying this. Yes, vision can be affected by the other senses, but that's not what's going on here. Expectation bias is a cognitive bias: we're being fooled by our minds, not our other senses.

And just to really blow your mind, if you drum up the courage to do it, I would bet you can't tell the difference between HD and CD quality, HD and 320 Kbps MP3, HD and 190 Kbps MP3 and I wouldn't be surprised if you couldn't tell the difference down to 128 Kbps MP3, although at that level, some tracks become more obvious.

Give it a whirl.

And if he claimed to have passed the 320kbps test, would you believe him?

 

[fr0g]

And yes, you do have a problem with your hearing, the same one that all humans have, that it can be affected by your other senses.

I'm not sure why you keep on saying this. Yes, vision can be affected by the other senses, but that's not what's going on here. Expectation bias is a cognitive bias: we're being fooled by our minds, not our other senses.

And just to really blow your mind, if you drum up the courage to do it, I would bet you can't tell the difference between HD and CD quality, HD and 320 Kbps MP3, HD and 190 Kbps MP3 and I wouldn't be surprised if you couldn't tell the difference down to 128 Kbps MP3, although at that level, some tracks become more obvious.

Give it a whirl.

And if he claimed to have passed the 320kbps test, would you believe him?

1. Semantics. Yes, we are being affected by our mind. All senses are are information pathways into the brain. The brain and the information from any of our senses can be fooled by other information in the brain. Is that better?

2. It's not a question of whether I believe him. It's more wanting to give him access to tools that will prove it to himself. But to answer your question anyway, I would not take it as fact unless it was done scientifically. If he supplied logs and access to the files he used (so they could be checked for glaring differences not related to SQ) then I'd be inclined to trust him, but would not take that as gospel. That would require more participants and an unbiased judge. Of course that aint gonna happen, so as I said, it's more of a test to prove something to oneself.

 

[steve_1979]

Ohhhhhhhh, so that necessarily makes my hearing wrong now doesn't it? Time to move on Steve...

I never said that anything is wrong with your hearing and why such an agressive tone to your posts? There's no need to take it so personally.

...And don't give me a argument about 'expectation bias', because this was not primarily an experiment on my part, I had to downsample out of necessity, so I WANTED it to sound equally good

That's not how expectation bias works. It doesn't matter whether or not you wanted to hear a difference. That's irrelevant. If you could see which version you were listening to then expectation bias can be a factor.

EDIT - Spelling

 

[manicm]

Fr0g, ABX testing is just a method and not rooted in any hard science, so your assertion that I could not be correct is complete nonsense *smile*

I am willing to concede that maybe MediaMonkey was not the ideal software to use, but I resolutely and absolutely stand by my conclusion. Maybe other software would have yielded different results for me, I don't know. *smile*

The drums on Love Me Like You Do have a lot of reverb of them, and I felt the downsample didn't quite cut it.

 

[andyjm]

Fr0g, ABX testing is just a method and not rooted in any hard science, so your assertion that I could not be correct is complete nonsense *smile*

Manic,

I dont know what you think 'hard science' is, but ABX testing definitely has a place in the scientific approach to auditory testing.

I would suggest checking out wikipedia for 'scientific method' and 'abx test'

 

[radiorog]

Just to be a spiny Norman.... I don't think real scientists touch wikipedia with a barge pole. Anyone can write anything in them, and they may never be corrected. Scientific subjects at university tell their students never to use it. There are always original and more reliable sources.

 

[Jota180]

Just to be a spiny Norman.... I don't think real scientists touch wikipedia with a barge pole. Anyone can write anything in them, and they may never be corrected. Scientific subjects at university tell their students never to use it. There are always original and more reliable sources.

Wiki's that have references are not the same as some piece that no one has referenced. It all depends on the quality of the reference.

 

[matt49]

Just to be a spiny Norman.... I don't think real scientists touch wikipedia with a barge pole. Anyone can write anything in them, and they may never be corrected. Scientific subjects at university tell their students never to use it. There are always original and more reliable sources.

That's true, but we're not dealing with university-level precision here. For our purposes Wikipedia is pretty reliable most of the time, especially if (as in this case) it's a question of getting a basic introduction to an uncontroversial subject (or at least a subject that ought to be uncontroversial).

 

[andyjm]

Just to be a spiny Norman.... I don't think real scientists touch wikipedia with a barge pole. Anyone can write anything in them, and they may never be corrected. Scientific subjects at university tell their students never to use it. There are always original and more reliable sources.

For technical subjects, I have always found wikipedia a very good source of information. For STEM subjects, there is a pretty dedicated band of guys who keep the entries on track and updated. A good wiki entry will have a bibliography that will allow a reader to delve deeper into the subject if they wish.

I would be interested if you could point out a wikipedia entry on a technical subject that you consider to be inaccurate.

 

[manicm]

Fr0g, ABX testing is just a method and not rooted in any hard science, so your assertion that I could not be correct is complete nonsense *smile*

Manic,

I dont know what you think 'hard science' is, but ABX testing definitely has a place in the scientific approach to auditory testing.

I would suggest checking out wikipedia for 'scientific method' and 'abx test'

I've read the wiki piece, and its origins are that of a very technical method, still not rooted in any science and far from infallible. I'm not detracting from it in anyway, but it also doesn't proclaim ears as useless.

 

[chebby]

Although I basically agree with fr0g and Steve on much that they say, there comes a point where the victim goes into a coma from all the bludgeoning with science textbooks (and they might come-to with a lifelong aversion to the subject as a result!).

 

[andyjm]

I've read the wiki piece, and its origins are that of a very technical method, still not rooted in any science and far from infallible. I'm not detracting from it in anyway, but it also doesn't proclaim ears as useless.

Manic,

What do you think 'science' is?

 

[fr0g]

Fr0g, ABX testing is just a method and not rooted in any hard science, so your assertion that I could not be correct is complete nonsense *smile*

I am willing to concede that maybe MediaMonkey was not the ideal software to use, but I resolutely and absolutely stand by my conclusion. Maybe other software would have yielded different results for me, I don't know. *smile*

The drums on Love Me Like You Do have a lot of reverb of them, and I felt the downsample didn't quite cut it.

You are missing the point.

The ABX test should you dare to try it will reveal "to yourself" if there really are audible differences between tracks.

It's amazing how the mind works when it thinks there should be a difference. You take away any possibility of knowing which is which, then suddenly there is none.

But for some reason you're just blabbing on. Do the test, if only for yourself. Do it with the mediamonkey files if you like.

Also, if you care to you could upload a couple of files so others could see your test files. The original and the one you created in media monkey.

The first 30 seconds would do as to avoid copyright problems.

 

[manicm]

Ok, what if I do ABX testing, and my conclusions are then the same? You assume ABX testing is the only way to listen properly, in which case none of us would have lives. ABX is a method, it's not science. You say ABX is the only reliable way of listening and I say you're just wrong.

 

[matt49]

Ok, what if I do ABX testing, and my conclusions are then the same? You assume ABX testing is the only way to listen properly, in which case none of us would have lives. ABX is a method, it's not science. You say ABX is the only reliable way of listening and I say you're just wrong.

The point I and others have been making is that blind ABX is the best way to establish whether you can really hear a difference between two components. There are other ways of doing it, but they're less reliable. The reason why ABX is the best way is that it takes expectation bias out of the equation.

If you do an ABX and the results are the same as for sighted listening, then you've learnt something good. What's not to like?

I don't understand your distinction between 'method' and 'science'. Doesn't science use methods, e.g. experimentation?

 

[The_Lhc]

ABX is a method, it's not science.

Yes it is, that's EXACTLY what it is, it is the rigorous testing of a hypothesis to determine an accurate conclusion. That's practically the dictionary definition of the scientific method. If you don't believe that then you simply have no idea what science is.

 

[steve_1979]

ABX is a method, it's not science.

Yes it is, that's EXACTLY what it is, it is the rigorous testing of a hypothesis to determine an accurate conclusion. That's practically the dictionary definition of the scientific method. If you don't believe that then you simply have no idea what science is.

Indeed!

I give up.

 

[fr0g]

Ok, what if I do ABX testing, and my conclusions are then the same? You assume ABX testing is the only way to listen properly, in which case none of us would have lives. ABX is a method, it's not science. You say ABX is the only reliable way of listening and I say you're just wrong.

Just do them.

And yes, it is scientific. It is a method of removing prior knowledge which can and does influence your choice.

It isn't the "only reliable way of listening", rather it is the only reliable way of being sure there is a difference. If you can spot a difference, say 10 times in a row, then there is a high probability that there is a difference.

And really, it doesn't take long. I don't spend time ABX testing. I simply did it a few times to prove something to myself. And the conclusion for me is that there really is no audible difference between an MP3 and a 32/192 recording (assuming of course they are from the same source recording).

So in conclusion, ABX testing (perhaps a couple of hours of my life IN TOTAL) has given me the utmost confidence to listen to Spotify with the confidence that it really is as good as it gets (again, depending on which master they used - that's another kettle of fish).

So, just do it. It will take you 10 minutes once set up.

Then, if you find a difference, perhaps take an MP3 using EAC or something more robust. Double check the bitrate (maybe you have ripped using a 128 Kbps default??)

It's actually quite fun

 

[manicm]

Is ABX testing infallable? I would hazard not. Is ABX testing repeatable with the same results? No. Is it therefore purely scientific? NO.

And to invalidate anyone's hearing just because said testing was not done is complete rubbish.

 

[matt49]

Is ABX testing infallable? I would hazard not.

You're creating a straw man. The question isn't whether it's infallible; it's whether it's less fallible than other methods of testing, and this is demonstrably the case.

Is ABX testing repeatable with the same results? No. Is it therefore purely scientific? NO.

Properly conducted and with a statistically significant number of attempts, yes, it will be repeatable with the same results. The same as for any other properly conducted psychological experiment.

And to invalidate anyone's hearing just because said testing was not done is complete rubbish.

No-one's invalidating anyone else's hearing. If you think that's what's at stake, you've missed the point. The point is that everyone (or at least everyone who believes they're really in a position to make a correct judgement) is fallible by reason of the expectations they bring to a sighted listening test. No-one's saying "if you think you can hear a difference in a sighted listening, your hearing is rubbish". What people are saying is: "in sighted tests our minds make us think we're hearing things that aren't there, and that's not just you, it's all of us".

 

[The_Lhc]

Is ABX testing infallable?

In what sense? If done correctly, then there is no room for argument, in the scientific sense then yes, that is infallible.

I would hazard not. Is ABX testing repeatable with the same results? No.

What sort of result are you expecting to repeat? If you listen to the same two tracks ten times in a random order on two separate occasions and you can't repeat the same answers then the ABX testing has done exactly what it's supposed to do, that is, to show you that there is no discernable difference between the two tracks.

Is it therefore purely scientific? NO.

Yes it is, what exactly do you think "scientific" means?

And to invalidate anyone's hearing just because said testing was not done is complete rubbish.

The whole point is that what you think you're "hearing" is not due to anything to do with your ears, that's what the ABX test is designed to remove, the outside influences.

Of course you know that already, otherwise you wouldn't be avoiding the test.