[syn_fx]

Quote:

Originally Posted by iponderous /img/forum/go_quote.gif ^ Why do you think it would be helpful to inform the OP of your musical taste? Shouldn't a top-tier IEM be able to reproduce all genres of music very well?

1. I don't intend to inform him of my tastes, I want him to tell us what he likes. My tastes aren't the point here!

2. Since the top tier IEMs sound different (IEMs generally don't sound the same), there will most likely be a preference of what one sounds like compared to another - a preference that often changes depending on track/style of music.

For example, I prefer Nina Simone on the IE8 compared to the Phonak PFE, whereas I prefer RHCP on the PFE compared to the IE8. If I listened to a lot of RHCP, I'd rather take the PFE, and similarly the converse holds.

Generally, if you look at most reviews where top end IEMs are compared, a lot of people prefer different ones for different styles of music. This, along with how the OP likes his generally sound to be presented should dictate his choice.

 

[iponderous]

^ Oh that's what you meant! I misunderstood you there. My apologies to you.

The problem with your argument is that people hear differently, that's why recommendations regarding the sound characteristics of any IEM are inherently subjective. How the OP hears music might differ markedly from how you do.

The varying and often conflicting descriptions from Head-Fier's of the same phone's sound signature fill these forum threads. You can research the threads extensively, ask questions and read reviews, but this is no substitute for hearing a pair of phones for yourself.

I would argue that the value of undertaking extensive research in this forum, in the hope of ascertaining the right phone for your particular ears based on the hearing perceptions of other people, is questionable. It can be interesting, entertaining and informative but it is subject to countless variables, making it inexact at best and misleading at worst. It can confuse rather than clarify.

I've been down that path and found that hearing really is believing, and everything else is just hearsay.

 

[Rip N' Burn]

Quote:

Originally Posted by iponderous /img/forum/go_quote.gif I've been down that path and found that hearing really is believing, and everything else is just hearsay.

Well said. I purchased the FX500 after reading posts that it was better than the IE8. I was mightily disappointing. For my ears, it didn't come close to IE8. The only thing that it did better was the low bass extension.

 

[syn_fx]

I agree with you completely.

The problem with IEMs lies in the fact that they are difficult to audition, since most shops won't let you, for hygiene reasons. Therefore, unless you have a willing friend who happens to have a pair of IEMs that you crave, you can never really know.

Given this, I think we shouldn't dismiss the impressions and reviews from people, especially in the case where they are numerous. Some things are more objective (eg headphone x is more bassy than headphone y), and other things are much more subjective (eg a certain headphone is sibilant).

Of course, we all hear sound different, but if a lot of people claim the same thing, you can assume with a certain (hopefully relatively high) probability that you are going to hear roughly the same characteristic. So you can make some judgements by looking at impressions on the whole, which means music tastes DO matter, as you can use other peoples opinions in large volume to give yourself a better chance.

Maybe you'll be unlucky and hear completely differently (it's a risk), but either way, if you're intelligent, and take things with a pinch of salt, you can use the information to make a more informed decision - clearly better than just selecting on a random basis if you could not audition.

But yes, no reviews/impressions will ever compare to a real life test...

 

[Rip N' Burn]

Impressions and reviews are great to get a general sense of the sound signature of the iem but it can also be misleading as was the case with the FX500 for me at least. With other iems that I have listened to, I can more or less agree with the general consensus about the sound signature.

 

[iponderous]

Hi syn_fx,

"I agree with you completely.

The problem with IEMs lies in the fact that they are difficult to audition, since most shops won't let you, for hygiene reasons. Therefore, unless you have a willing friend who happens to have a pair of IEMs that you crave, you can never really know."

Agreed.

"Given this, I think we shouldn't dismiss the impressions and reviews from people, especially in the case where they are numerous. Some things are more objective (eg headphone x is more bassy than headphone y), and other things are much more subjective (eg a certain headphone is sibilant)."

I'm not suggesting that we should dismiss these impressions and reviews. I think the value of these lie in the ability to pique one's interest. I disagree that there are some aspects of a headphone's performance which, are more objective or less subjective than others. They either are or they are not. If you cannot devise a way to objectively measure these sound traits that is universally accepted, then it is nothing more than anecdotal evidence, which is subjective. Anecdotally, one could draw the conclusion that the IE8 is a bassy phone based on the number of descriptions contained in the IE8 Appreciation Thread, but I've also read descriptions from people who haven't found it bassy enough. Do I conclude that it is bassy because more people who found it to be so elected to post their impressions in a forum thread? Does this account for the number of IE8 owners who don't find it at all bassy but simply didn't post there at all?

"Of course, we all hear sound different, but if a lot of people claim the same thing, you can assume with a certain (hopefully relatively high) probability that you are going to hear roughly the same characteristic. So you can make some judgements by looking at impressions on the whole, which means music tastes DO matter, as you can use other peoples opinions in large volume to give yourself a better chance."

Again, this is just anecdotal and does not take into account the numerous variables involved i.e.; different hearing perceptions, music sources, recording bit-rates, eartips and so on and so forth. You can make your assumptions and judgements as you say, but they are simply that and they may prove to be erroneous or correct, useful or useless, to you. Musical tastes do matter in that we all have our own, and if you wish to share them and your descriptions of the music that you listen to with other people here then fine, but this is a descriptive exercise, not an empirical one. You won't really "know" how this music sounds through a set of phones until you hear it for yourself. Many descriptions in these forum threads seem more authoritative and objective than others because of how they are expressed, but they are no less subjective in nature.

"Maybe you'll be unlucky and hear completely differently (it's a risk), but either way, if you're intelligent, and take things with a pinch of salt, you can use the information to make a more informed decision - clearly better than just selecting on a random basis if you could not audition."

I remain unconvinced that the approach you are advocating is any more reliable than buying an earphone on a complete whim. I took the "intelligent" approach that you subscribe to, and it did not prepare me for how the earphones "actually" sounded once I heard them. You make what you hope is an informed decision based upon the subjective impressions and experiences of others. If your own experience accords with these impressions then your research would appear to be warranted, but if it diverges markedly then does that mean that you weren't informed enough? Or does it mean that you were just unlucky? I suspect that it is very much the latter. This pursuit is about as scientific as playing the ponies.

"But yes, no reviews/impressions will ever compare to a real life test..."

We agree on this much.

 

[ibis99]

Quote:

Originally Posted by iponderous /img/forum/go_quote.gif ^ Why do you think it would be helpful to inform the OP of your musical taste? Shouldn't a top-tier IEM be able to reproduce all genres of music very well?

You have been around long enough to know better.

 

[syn_fx]

Anecdotally, one could draw the conclusion that the IE8 is a bassy phone based on the number of descriptions contained in the IE8 Appreciation Thread, but I've also read descriptions from people who haven't found it bassy enough. Do I conclude that it is bassy because more people who found it to be so elected to post their impressions in a forum thread? Does this account for the number of IE8 owners who don't find it at all bassy but simply didn't post there at all?

I think you've missed my point completely... Saying the IE8s are bassy based on descriptions means nothing.

However, if 100 people claim that the IE8s are bassy compared to a certain headphone, (for example the ER-6i), and one person says the opposite, you can conclude the following:

"with a high degree of likelihood, you will find the IE8s bassier than the Er-6is".

This statement does not mean that it will always be the case (you might have different hearing) - it is a probabilistic statement. You probably will find them bassier. If you like bassier sound, this gives you a tool that improves your chances of picking the right IEM for you out of the two.

By your argument, statements such as "£1000 custom IEMs are better than £5 buds" only "pique one's interest". This is clearly not so!

Again, this is just anecdotal and does not take into account the numerous variables involved i.e.; different hearing perceptions, music sources, recording bit-rates, eartips and so on and so forth. You can make your assumptions and judgements as you say, but they are simply that and they may prove to be erroneous or correct, useful or useless, to you. Musical tastes do matter in that we all have our own, and if you wish to share them and your descriptions of the music that you listen to with other people here then fine, but this is a descriptive exercise, not an empirical one. You won't really "know" how this music sounds through a set of phones until you hear it for yourself. Many descriptions in these forum threads seem more authoritative and objective than others because of how they are expressed, but they are no less subjective in nature.

Yes, you won't know how something sounds like until you've heard it. However, there is nothing anecdotal about probability, and the argument I've just stated above applies. Your not making assumptions about characteristics. You're making a statement about probabilities, which is in fact very useful.

This leads me nicely onto your last points:

I remain unconvinced that the approach you are advocating is any more reliable than buying an earphone on a complete whim. I took the "intelligent" approach that you subscribe to, and it did not prepare me for how the earphones "actually" sounded once I heard them. You make what you hope is an informed decision based upon the subjective impressions and experiences of others. If your own experience accords with these impressions then your research would appear to be warranted, but if it diverges markedly then does that mean that you weren't informed enough? Or does it mean that you were just unlucky? I suspect that it is very much the latter. This pursuit is about as scientific as playing the ponies.

I agree with your suspicions, it is most likely to be the latter, ie that one is unlucky. But that's the whole point of the method - it doesn't let you pick the right headphone - it gives you a higher probability of picking the right one. And in this way, it is very scientific indeed - statistics relies very heavily on this kind of approach. The whole point of using probability is to deal with subjective factors (like you listed - sources, seals, fits, etc).

To conclude, sure, sometimes your inferences will be wrong. But on the whole, you can draw quite powerful conclusions (just look at the example of the IE8 vs Er-6i).

 

[iponderous]

syn_fx:

"I think you've missed my point completely... Saying the IE8s are bassy based on descriptions means nothing."

No, I'm afraid you've misunderstood mine and consequently, failed to refute it. I'm referring to the uncontrolled sample group from which, you are drawing your conclusions that have no empirical basis. The data just isn't solid enough to extrapolate from with any certainty.

"However, if 100 people claim that the IE8s are bassy compared to a certain headphone, (for example the ER-6i), and one person says the opposite, you can conclude the following:

"with a high degree of likelihood, you will find the IE8s bassier than the Er-6is".

This statement does not mean that it will always be the case (you might have different hearing) - it is a probabilistic statement. You probably will find them bassier. If you like bassier sound, this gives you a tool that improves your chances of picking the right IEM for you out of the two."

Yes, you can conclude that but your conclusion is only as valid as the data that it is based on, which at Head-Fi is uncontrolled and subject to numerous variables that have not been accounted for. The tool you refer to is so imprecise that you may as well discard it.

"By your argument, statements such as "£1000 custom IEMs are better than £5 buds" only "pique one's interest". This is clearly not so!"

No, that was not what I meant. Head-Fi is a great source to find out about new products that you wouldn't have known existed if you didn't frequent these forum threads. That's what drew me here.

"Yes, you won't know how something sounds like until you've heard it. However, there is nothing anecdotal about probability, and the argument I've just stated above applies. Your not making assumptions about characteristics. You're making a statement about probabilities, which is in fact very useful."

You're making assumptions based on anecdotal evidence and calling them probabilities. I call that speculation.

"I agree with your suspicions, it is most likely to be the latter, ie that one is unlucky. But that's the whole point of the method - it doesn't let you pick the right headphone - it gives you a higher probability of picking the right one. And in this way, it is very scientific indeed - statistics relies very heavily on this kind of approach. The whole point of using probability is to deal with subjective factors (like you listed - sources, seals, fits, etc)."

I'm just repeating myself here but in order to determine the likelihood or if you prefer, the degree of probability, this has to be predicated on more controlled data than Head-Fi is offering, which really is nothing more than anecdotal impressions and experiences. The notion of probability that you are describing can also be called guessing. It has no empirical basis whatsoever.

"To conclude, sure, sometimes your inferences will be wrong. But on the whole, you can draw quite powerful conclusions (just look at the example of the IE8 vs Er-6i)."

And again, your conclusions, inferences, probabilities, call then what you will, are only as valid as the data that you are drawing them from, and I think you're drawing a very long bow here syn_fx.

 

[syn_fx]

No, I'm afraid you've misunderstood mine and consequently, failed to refute it. I'm referring to the uncontrolled sample group from which, you are drawing your conclusions that have no empirical basis. The data just isn't solid enough to extrapolate from with any certainty.

Your first paragraph is nonsense, and the rest of your answer stems from it.

Non-controlled data is used very often in science, in cases where we can't just do a laboratory experiment. Statistical inference is still made on it (the methods are different), and the conclusions based on this are still valid - please do some reading up if you don't believe me!

The group is not homogeneous, true, which lowers the "certainty", but everyone who posts here has ears, and although they don't hear exactly the same, there are similarities in hearing (we can distinguish a drum from a guitar, or any other mundane example you care to think of). This is all we need for inference, some common ground. It is not a requirement that the group is homogeneous to use statistics, that's the whole point of using them.

Which hat did you pull the "data isn't solid enough to extrapolate from"?! What a way to try to refute methodology with a subjective statement! Do you have at least some evidence or way of proving this? I'd love to see it! And please, while your at it, explain how you can't use statistics on non controlled data.

The notion of probability that you are describing can also be called guessing. You still don't understand why this "guessing" (as you call it) is more useful than a blind purchase. There is absolutely nothing wrong with the data. However, like all real life empirical non-experiment generated data, it is difficult to interpret well.

 

[iponderous]

^ The uncontrolled sample group to which I was referring wasn't meant to be specific. It could be any forum thread here at Head-Fi that you care to base your probabilities on or draw your conclusions from. You took me too literally.

Yes, of course we can agree on the difference between a drum or a guitar, but can we agree that the drum lacks "snap" or that the guitar lacks "crunch"? We are attempting to use words to convey how each one of us subjectively hears sound in terms that a far more sophisticated than the two examples that you provided.

Sorry to have to tell you this syn_fx, but I'm not the one who is trying to prosecute the argument that you can take an empirical approach to ascertaining the differences in sound signatures of earphones based on other people's posts in this forum; you are. I don't have a methodology other than to acquire the earphones for myself and to trust my own ears. I found that this took the guess work out of determining how an earphone "actually" sounds rather than relying on the opinions of others here at Head-Fi, no matter how sincere and well meaning they may be.

It's not that I'm incapable of understanding your point of view, I just don't agree with it. You have been making assertions and putting them forward as statements of fact. I'm sorry but I don't regard that as a substitute for a compelling argument.

Now I'm detecting some exasperation in the tenor of your last post, so I think we should agree to disagree. I wouldn't want this debate to deteriorate into something unsavoury.

 

[syn_fx]

Yes, let's agree to disagree. I'm not particularly concerned about the debate deteriorating into something unsavoury - I personally find it interesting and quite respect your viewpoint (even though I don't agree). However, we should probably avoid hijacking this thread any more than we have done!

Everyone should choose how they pick IEMs themselves, since only they are the ones who can be happy with that choice, not us.

Therefore even if you think that I'm "making assertions and putting them forward as statements of fact", or that I'm just plain wrong, let the OP decide on how he wants to choose.

In fact, why would the OP be posting his questions here if he didn't want peoples opinions and impressions to influence his choice?

 

[iponderous]

"In fact, why would the OP be posting his questions here if he didn't want peoples opinions and impressions to influence his choice?"

The OP was looking for affirmation for his/her preferred choice of the Triple.Fi Pro 10 I believe, and I was happy to provide it.

There's nothing wrong with seeking people's opinions and impressions regarding a proposed choice of earphone, just don't place too much stock in it or you might be disappointed.

Goodnight syn_fx.

PS "Statistical inference" sounds rather rubbery to me. Still, if you can produce the statistical data that best supports the outcome that you desire, then surely that can't be a bad thing.

 

[syn_fx]

Quote:

Originally Posted by iponderous /img/forum/go_quote.gif PS "Statistical inference" sounds rather rubbery to me. Still, if you can produce the statistical data that best supports the outcome that you desire, then surely that can't be a bad thing.

"The OP was looking for affirmation for his/her preferred choice" That's not what I think. I think most people who post asking for other considerations want other considerations, not an affirmation.

Please don't try to finish an argument by slating something on which an entire science is based on! You won't get very far. The point is not to produce something to support the outcome that you desire, it's to produce something irrefutable. The fact that people hear slightly differently is irrelevant.

Taking the time to explain to you that statistics is designed for conclusively solving problems, not lying, seems pretty pointless to me. It's pretty clear for me that you're not familiar with the subject, and don't understand the notion of using empirical methods on uncontrolled data to draw valid conclusions, given the assumptions you use. Perhaps with a little reading, you could convince yourself otherwise. I'm not going to link you to all the relevant text - its available in any good text book on statistical methods. This is a good start though:

Statistical inference - Wikipedia, the free encyclopedia

I hope with some reading it will stop "sounding rubbery" to you. Or perhaps, we just "hear" differently, pun intended

 

[iponderous]

^ Clearly my attempt at humour was lost on you. I did add a winking smiley in case you didn't notice.

I thought we had agreed to disagree? Your last paragraph was condescending and irrelevant to the debate that we had been having. I note that you wrote this after stating in a previous post that you "quite respect" my viewpoint - nice one.

I didn't claim to have any knowledge of the use of statistical data or related methodologies; that would appear to be your area of interest. You chose to call upon this in order to bolster your position, which I consider to be tenuous. However, you have not employed any of the methodologies to which you refer in support of your argument. You merely cited them and then provided some unsophisticated examples of your own that simply didn't hold up under scrutiny.

Rather than provide me with a link to the at times questionable Wikipedia about the subject, I suggest that you read over our exchange of posts, where you will find that you have failed to prove your point through rational argument. I have questioned your assertions and found flaws in your reasoning, which has obviously frustrated you and caused you to become increasingly dismissive of my counter-arguments, which you misunderstood on more than one occasion.

The point that I wish to make is that asserting something doesn't make it so. My point of view, which obviously does not align with yours, is based on my own research, observations and most importantly for me within the context of this forum, my own experiences. You have adopted a position that you appear to believe in and mounted a half-baked argument in support of it that I find unconvincing. You have made unsubstantiated claims, yet you seem to expect that I should be swayed by them because you have made reference to statistical methodologies that you haven't actually utilised. Well I'm afraid that you're going to have to do a little better than that syn_fx.

I invite you to develop a statistical methodology, apply it, and post the results in a new forum thread here at Head-Fi. I'd be interested to see the methodology explained, the data gathered and the conclusions that you draw from it, as I'm sure many other Head-Fiers would. In the meantime, I suggest that you refrain from attempting to assume an air of intellectual superiority based on unsubstantiated assertions, flawed argument and dodgy examples that don't even warrant the description of pseudo-science.