Criteria for choosing headphones

[Buckle-meister]

Dear all,

I am considering buying an ipod nano, but should state right at the outset that I'd be recording songs in .wav format, i.e. uncompressed. I've a couple of questions regarding purchasing quality headphones to use in place of the (presumably) basic ear buds that come with the nano itself.

1. I've read that etymotic headphones are great for reducing outside ambient noise (though can be uncomfortable to wear due to the depth they reach within the ear), but do they also reduce sound leakage the other way? In other words, do they reduce the possibility of others hearing what you're listening to?
2. When looking at the specification for headphones, what are the important features? Frequency range, impedence etc.?

 

[Nick250]

What I have to add is to pay particular attention to the efficiency of the headphone. For example my very most favorite headphones, senns HD580s, sound bloated and bland through a portable listening device(CD player in my case) and truly wonderful played via a receiver. Koss port pros are efficient sound great (relatively speaking) on the portable CD player and are small and efficient. Grados SR80s sound OK through the CD player unless I really crank it. I have not tried the in ear headphones but I wonder if the efficiency issue may come into play with them as well.

 

[warpdrive]

Yes, The IEM (In Ear Monitors) prevent leakage both ways. They are really inserted DEEP into your ears (exactly like a set of foam earplugs work). It's likely it wouldn't even both your mate sleeping beside you. A very good seal is mandatory otherwise it will sound like doo-doo.

Secondly, I think the idea of having uncompressed songs is noble, but for portable listening it's overkill. I'm somewhat of a headphone audiophile and the difference between higher bitrate MP3 (say 320kbps MP3, 320kbps AAC) and uncompressed is not noticeable unless you have a high grade system to plug it into. Using uncompressed will eat up your battery life and space like crazy.

For headphones, keep in mind that portable MP3 players may not drive some headphones to adequate levels. Some of the higher impedance ones meant for home listening would not sound their best hooked up to an iPod. It's not just a matter of volume, it's handling the dynamics of the music. Many people resort to buying a headphone amplifier to provide adequate voltage to the headphones. For the most part, sticking to earbud or IEM earphones choices except the higher end Etymotic ER4S, you'll be ok. For full sized headphones, it varies, some work well (Grado's, Sennheiser HD555), some don't (AKG K701/601, Sennheiser HD600)

Specs are pretty much meaningless in the world of headphones, especially anything related to frequency response.

For a good start about which headphones to buy, head over to headphone.com and look at their selection guides. They've also measured many of the headphones they have and that will give you a hint about the frequency signature.

For the record, I use my Etymotic ER6i's with the iPod. They are a good middle of the road budget IEM.

 

[jonnythan]

You can use the Rockbox firmware for the iPod and compress everything using lossless FLAC. You'll get about double the number of songs.

 

[Buckle-meister]

Thank you all. I'll be sure to check out headphone.com.

I've no problem with .wav files eating up memory. I'm a rather practical sort o' guy and don't feel the need to have something that holds hundreds of albums. However, I hadn't appreciated about this type of file eating the battery (presumably due to a greater amount of information being read).

If headphones have a high impedance, does that imply that because they're harder to drive, they'll also tend to eat the battery, or is the difference between what would be regarded as normal and high impedance headphones negligible?

 

[jonnythan]

Well when filling a 2GB Nano with .wav files, you're looking at maybe 3-4 albums with wav vs 6-8 with FLAC. If that's cool with you, that's fine, just making a suggestion

I also agree about high bitrate MP3's in general being indistinguishable from wav files, particularly using earphones.

 

[Buckle-meister]

Well when filling a 2GB Nano with .wav files, you're looking at maybe 3-4 albums with wav...

Hmm. Something's amiss here. Check me please:

One Gigabyte (GB) is 1000 Megabytes (MB). I've just checked the size of Conscience, an album by The Beloved, and it measures 54.3MB for 11 songs, which I'll take as being representative of a typical album. Dividing 1000 by 54.3 should return the number of albums the 1GB ipod nano can store in .wav format, and this comes in as approximately 18.

Have I messed up somewhere?

On a separate issue, regarding FLAC, are you sure it losslessly compresses? I thought only Meridian Lossless Packaging (MLP) for the new HD-DVD and BD discs was able to achieve this.

I also agree about high bitrate MP3's in general being indistinguishable from wav files, particularly using earphones.

(Sigh) I know, but I just don't like the thought that I'm not hearing all the original material, even if I'm not hearing all the original material.

 

[deftech]

I bought a pair of akg k26p's to go with my mp3 player. They are compact on ear phones. They are comfortable, sound good, keep out ambient noise and are reasonably priced. Got mine at guitar center for $40.00

 

[jonnythan]

Hmm. Something's amiss here. Check me please:

One Gigabyte (GB) is 1000 Megabytes (MB). I've just checked the size of Conscience, an album by The Beloved, and it measures 54.3MB for 11 songs, which I'll take as being representative of a typical album. Dividing 1000 by 54.3 should return the number of albums the 1GB ipod nano can store in .wav format, and this comes in as approximately 18.

Have I messed up somewhere?

54MB of uncompressed CD audio yields just over 5 minutes of music.

54MB of 128kbps MP3 yields about 56 minutes of music. So, you're obviously measuring your album as an MP3 - and probably a 128kbps one.

A single audio CD, depending on length, is typically 400-600MB in wav format. A single audio CD in 128kbps mp3 is typically 50-65MB. 128kbps mp3 typically achieves about a 10-to-1 compression ration.

FLAC, which is indeed 100% lossless, achieves about a 2-to-1 compression ratio.

There are several lossless audio codecs out there. Windows Media Player even includes a proprietary one. FLAC even supports as many channels, bitrates, sampling frequencies, etc, as you want. Lossless audio is easy. Lossless compression of *any* data source is easy. There are hundreds of different file compression tools for every computing platform that are all lossless, and all can be perfectly applied to audio files every bit as well as a spreadsheet file.

http://flac.sourceforge.net/

 

[WmAx]

Thank you all. I'll be sure to check out headphone.com.

CAUTION

Headphone.com is a snake-oil vendor. They mix up truths with half-truths and fantasy(pure fiction). Also: you can not determine the sound signature of a headphone by the frequency response graphs in the same manner as you do for speakers. For physical reasons, you must interpret them differently. I am familiar with most details of headphone operation/technical. I will try to answer any specific inquiries you may have.

-Chris

 

[WmAx]

Secondly, I think the idea of having uncompressed songs is noble, but for portable listening it's overkill. I'm somewhat of a headphone audiophile and the difference between higher bitrate MP3 (say 320kbps MP3, 320kbps AAC) and uncompressed is not noticeable unless you have a high grade system to plug it into. Using uncompressed will eat up your battery life and space like crazy.

Actually, an IPOD [and many other portable units] is high-grade; it will surpass, by magnitudes, the technical parameters required to remain transparent to human ears when used with loads[headphones]for which they were intended.

For headphones, keep in mind that portable MP3 players may not drive some headphones to adequate levels. Some of the higher impedance ones meant for home listening would not sound their best hooked up to an iPod. It's not just a matter of volume, it's handling the dynamics of the music.

Often, this is one in the same. Most[not all] digital playback devices are designed intelligently, so that 0dBFs in the digital file does not exceed the RMS voltage limit of the built-in amplifier, thus preventing voltage-limited distortion from ever occurring. In this case, it would be a matter of insufficient volume level; clipping would not occur.

Specs are pretty much meaningless in the world of headphones, especially anything related to frequency response.

I am probably guilty of nitpicking, but: specs are as meaningful here as they are in the world of speakers: when done correctly, they will let you know what to expect. However, assuming such sufficient data was ever provided, this would also require you to be able to interpret this data, which is beyond most, as it requires familiarization with various perceptual research and other related issues. But, the real issue is that sufficient measurements are rarely provided. If they were: how would the average person make good use of them? For the sake of accuracy, I would summarize it as:

The specifications/measurements typically given for headphones are useless.

-Chris

 

[Buckle-meister]

54MB of uncompressed CD audio yields just over 5 minutes of music.

Gulp.

Lossless audio is easy. Lossless compression of *any* data source is easy. There are hundreds of different file compression tools for every computing platform that are all lossless, and all can be perfectly applied to audio files every bit as well as a spreadsheet file.

Hmm, perhaps I should just stick to engineering eh?

CAUTION Headphone.com is a snake-oil vendor. I am familiar with most details of headphone operation/technical. I will try to answer any specific inquiries you may have.

Cheers mate. I'll come back to you with a couple of possible products I fancy and see what you (and others) think.

 

[warpdrive]

Actually, an IPOD [and many other portable units] is high-grade; it will surpass, by magnitudes, the technical parameters required to remain transparent to human ears when used with loads[headphones]for which they were intended.

This is false

The iPod and various other MP3 players are known for rolling off bass output under headphone load. Their headphone amps are far from adequate in terms of being transparent. This is widely documented. This is completely audible with any set of high grade headphones. And if one decides to use EQ, the problem is even worsened, the electronic EQ on these devices introduce all sort of artifacts through the frequency spectrum well within the audibility range.

 

[warpdrive]

CAUTION

Headphone.com is a snake-oil vendor. They mix up truths with half-truths and fantasy(pure fiction). Also: you can not determine the sound signature of a headphone by the frequency response graphs in the same manner as you do for speakers.

I knew this would come out eventually

I'm sorry you feel this way, but headphone.com website there is lots of useful information there in selecting headphones. Their ranking of the headphones they sell is also quite useful. A lot of my personal experiences with the headphones they sell mirror a lot of the advice they've given. In fact, I will continue to recommend their website as one of the best resources in selecting headphones as anything I've seen.

Since this is audioholics, I'm not going to make any comment regarding so called snakeoil, except to say:

They do offer a 30 day return policy, so it's of little risk if you do happen to follow their advice, buy their product and find your listening experience to be contrary to their advice, you are only out the return shipping cost on anything they sell. I've bought many things from there and their website is a valuable resource in narrowing down the headphones I like.

 

[warpdrive]

I am probably guilty of nitpicking, but: specs are as meaningful here as they are in the world of speakers: when done correctly, they will let you know what to expect. However, assuming such sufficient data was ever provided, this would also require you to be able to interpret this data, which is beyond most, as it requires familiarization with various perceptual research and other related issues. But, the real issue is that sufficient measurements are rarely provided. If they were: how would the average person make good use of them? For the sake of accuracy, I would summarize it as:

The specifications/measurements typically given for headphones are useless.

-Chris

This is all fine and dandy, but my point is: for all intents and purposes, you will not find any meaningful frequency response specs on any set of headphones (except Etymotic). Certainly not from any of the major vendors like Sennheiser, AKG, Sony, AT, Beyer
That is also why looking at frequency response graphs is useful, another reason to go to headphone.com
They also have a blurb about how to interpret the graphs which is quite relevant to any headphone buyer.

 

[WmAx]

This is false

The iPod and various other MP3 players are known for rolling off bass output under headphone load.

I recommend that you read the reply again. Additionally, the roll-off is because the output coupling capacitors used are insufficient in size for very low impedance loads, with which the unit was not designed to be compatible. However, the degree of roll-off present is not audible, or marginally audible, on average headphones, which rate in the 32 ohm range.

Their headphone amps are far from adequate in terms of being transparent. This is widely documented.

Which documentation? Please be specific as to which parameters will be altered, and how they correlate with audibility. It is true that some may not be 'transparent', but this is the minority, and the problems are clearly measurable[such as excessive noise, etc.]. Note: I will not consider audiophile speculation as valid documentation.

This is completely audible with any set of high grade headphones

Most 'high-grade' headphones require higher voltage than most portable units can provide, thus they will usually not achieve a satisfactory volume level. However, use a headphone with sufficient sensitivity and impedance, and it will be no problem.

And if one decides to use EQ, the problem is even worsened, the electronic EQ on these devices introduce all sort of artifacts through the frequency spectrum well within the audibility range.

I can not comment on the equalization of these devices, as I have never bothered to do an analysis.

I

knew this would come out eventually

I'm sorry you feel this way, but headphone.com website there is lots of useful information there in selecting headphones.

Well, since they insist on mixing truth with fantasy, the fault is their own. The problem with them is that the average person will not be able to discern the good information from the worthless information.


Since this is audioholics, I'm not going to make any comment regarding so called snakeoil, except to say:

This is all fine and dandy, but my point is: for all intents and purposes, you will not find any meaningful frequency response specs on any set of headphones (except Etymotic). Certainly not from any of the major vendors like Sennheiser, AKG, Sony, AT, Beyer

I agree.

That is also why looking at frequency response graphs is useful, another reason to go to headphone.com
They also have a blurb about how to interpret the graphs which is quite relevant to any headphone buyer.

The graphs are not very useful, except for a very rough relative comparison with each other[for example, headphone A has slightly more treble lift then headphone B, when you overlay the two graphs], but even then, it can be difficult for most people to properly interpret many graphs. This is beside the point that one can not interpret a headphone measurement as they would a speaker measurement.

-Chris

 

[JaceTheAce]

CAUTION

Headphone.com is a snake-oil vendor. They mix up truths with half-truths and fantasy(pure fiction). Also: you can not determine the sound signature of a headphone by the frequency response graphs in the same manner as you do for speakers. For physical reasons, you must interpret them differently. I am familiar with most details of headphone operation/technical. I will try to answer any specific inquiries you may have.

-Chris

Chris, why do you say they're a snake-oil vendor? I think their headphone selection is great...

 

[Geno]

I do a lot of 4 hour flights, so I'm a regular in-ear 'bud user. I have an Archos 20 GB MP3 player and a set of Etymotic ER-4 buds (about $100). I burn my own CDs using MP3 Pro 192KBS VBR. I"d have a pretty hard time telling the difference between the original CDs and the MP3 version using my setup. Other people on this site may differ with me on the sound quality of more expensive earbuds, but IMHO, you don't need to get too awfully neurotic with headphones to get what I'd consider an outstanding listening experience.
Of course, that's just my opinion; I could be wrong...

 

[WmAx]

Chris, why do you say they're a snake-oil vendor? I think their headphone selection is great...

Because they make and sell[and make up articles/details concerning] cables, amplifiers and amplifier upgrades that full under such definition.

-Chris

 

[warpdrive]

I recommend that you read the reply again. Additionally, the roll-off is because the output coupling capacitors used are insufficient in size for very low impedance loads, with which the unit was not designed to be compatible. However, the degree of roll-off present is not audible, or marginally audible, on average headphones, which rate in the 32 ohm range.

It most definitely is audible. This rolloff was long suspected in audible tests in early iPod models before anybody took any measurements to objectify the degree of rolloff. Not to mention that quite a few popular portable headphone models have 16 or 24 ohm impedance, which makes the problem more apparent.

Just a note: your posts in this thread are full of generalizations, which is not really useful to the poster in this situation.

I will be more specific: the iPod Nano suffers from audible rolloff that I can hear with 32 ohm iBuds.

Well, since they insist on mixing truth with fantasy, the fault is their own. The problem with them is that the average person will not be able to discern the good information from the worthless information.

Certainly, it's your opinion that they are peddling fantasy. Not everybody shares your opinion on that. Let's leave it at that.

The graphs are not very useful, except for a very rough relative comparison with each other[for example, headphone A has slightly more treble lift then headphone B, when you overlay the two graphs], but even then, it can be difficult for most people to properly interpret many graphs. This is beside the point that one can not interpret a headphone measurement as they would a speaker measurement.

The graphs are provided as a resource, and Headroom's site itself provides some information about how to use them. In fact, this is the most rich set of objective measurements I've yet seen in any shape or form in any publication or site. Of course it's obvious if the user does not understand how to interpret them, they are going to interpret them incorrectly. But armed with a bit of knowledge, they become extremely useful to help narrow down the many choices on something that is inherently difficult to audition (many vendors do not have display models). Even if the user starts with the iPod buds as a baseline reference graph, he can start to understand how the headphones sold there compare to what he has now.