Speaker and Headphone Choices for listening to Classical Music

[DavidW]

What was your cost, David?

They retail for a $1000/pair, I bought them on amazon for $400/pair!   You can find a favorable review of them on stereophile.

[DavidW]

Maybe something like this is an option for me; I have a tiny apartment and if I pushed the couch and the speakers two feet out from their respective walls, I'd be able to sit on the couch and kick the speakers! That said, I don't know the first thing about these sound systems and what all parts are needed between the disc drive and the speakers themselves.

Bookshelves sound much better in small rooms like that than towers do.  I use a Denon AVR as the inbetween speakers and source.  There is something neato I saw on Crutchfield.  It is a Marantz amp that has a cd player built in and also streams audio.  A nice all in one solution that if I didn't have my AVR would have been tempted by.

[Drosera]

I love headphones, and my preferences are towards the more coloured ones. Current favourites are the Audio-Technica W3000ANV and W5000, next to the Beyerdynamic T5p and T1. This already illustrates one of the things I like so much about headphones, you can change sound signature at a whim.

I like to keep the rest of the chain fairly neutral (maybe slightly warm) and let the headphones provide the colouration.

[marvinbrown]

I prefer equipment that sounds right an never think about whetjer it sounds warm or clinical. And (at least for speakers, headphones I hate) the result for a given setuo, results depends as much on (really; mostly on) room size, acoustic setup and speaker placement (and your own hearing and muusic selections) than on the equipment itself. But I don't like an overtly clinical sound, preferring a focus on soundstage and overall perspective rather than an overt focus on details. To me that is right.   

  I agree wholeheartedly here. I too prefer a wide soundstage and overall perspective over a clinical sound. This is especially true for opera/drama drama. To hear a dramatic opera unfold sonically before you where you no longer need the visual to aprreciate what is happening on stage is a priceless experience. I believe for opera/music drama you really do need floor standing speakers a decent amp and CD player to match. Headphones can be remarkably clear but can they sonically recreate an impressive soundstage like floor standing speakers can?

marvin

[mahler10th]

I prefer equipment that sounds right an never think about whetjer it sounds warm or clinical. And (at least for speakers, headphones I hate) the result for a given setuo, results depends as much on (really; mostly on) room size, acoustic setup and speaker placement (and your own hearing and muusic selections) than on the equipment itself. But I don't like an overtly clinical sound, preferring a focus on soundstage and overall perspective rather than an overt focus on details. To me that is right.   

  I agree wholeheartedly here. I too prefer a wide soundstage and overall perspective over a clinical sound. This is especially true for opera/drama drama. To hear a dramatic opera unfold sonically before you where you no longer need the visual to aprreciate what is happening on stage is a priceless experience. I believe for opera/music drama you really do need floor standing speakers a decent amp and CD player to match. Headphones can be remarkably clear but can they sonically recreate an impressive soundstage like floor standing speakers can?

marvin

Aye.  Count me in agreement with these comments.  Recently 'stopped using' a rather splendid Sony Home Theater system because of its very clear, clinical sound.  Nothing wrong with it, was great really, but I want to hear the character and resonance in the music rather than shapeless acoustic clarity.  My Marantz / Rega setup imparts all the character and resonance I could ever want in our music.     (Although the Sony still gets used for telly, movies, games, etc...it's perfect for those things).

[andolink]

Hi all, I'm new here.  Interesting to hear about what everyone's using for their music playback.

Here's my own rather modest gear:

Audiolab 8200CD (CD player)
NAD C356BEE (integrated amp)
PSB Imagine T (floorstanding speakers)
SVS SB 12-NSD (subwoofer)

HiFiMan HE-500 (planar-magnetic headphone)
Audio-gd NFB-6 (integrated headphone amp)

Norse Audio Skuld 2 Litz UPOCC HiFiMan cable
Belden 1800F Balanced interconnects

My avatar is indeed my HE-500 headphones.

[DavidW]

Is that a tube amp you have Scott?

[jut1972]

I've had a few pairs of headphones over the years, currently I use two pairs.  A set of noise cancelling headphones for commuting (Goldring NS1000's) which I use with a mobile DAC (Fiio E18).  For home use I have Sennheiser 650s.

Assuming your headphones are being used with a number of sources I'd recommend as flat a response as possible.  Any "characterful" headphones will be hit or miss when switching sources.

[71 dB]

Hi all, I'm new here.  Interesting to hear about what everyone's using for their music playback.

Here's my own rather modest gear:

:
Audio-gd NFB-6 (integrated headphone amp)
:

Unfortunately your otherwise great $400 headphone amp seems to lack crossfeeding, which is extremely important to kill spatial distortion.

[Daverz]

Unfortunately your otherwise great $400 headphone amp seems to lack crossfeeding, which is extremely important to kill spatial distortion.

Very few headphone amps include a crossfeed circuit.   I did find this thread at Head Fi listing some options:

http://www.head-fi.org/t/625678/crossfeed-amps

If your listening from a music server, a digital implementation might be an option.

[andolink]

Unfortunately your otherwise great $400 headphone amp seems to lack crossfeeding, which is extremely important to kill spatial distortion.

I seriously investigated crossfeed when considering a headphone amp to buy but found as many arguments con as pro.  The way it's implemented is crucial and very few of even the most high end h/p amps use it.  I find it's a rare recording that  has such wide channel separation that the soundstage sounds unnatural.  When I come across such, I just use my speaker gear.  In fact many recordings (especially chamber music) sound better in imaging, detail and presence using my headphones than speakers.  The speakers I find always preferable in orchestral music, opera, oratorio, etc.

[Drosera]

Unfortunately your otherwise great $400 headphone amp seems to lack crossfeeding, which is extremely important to kill spatial distortion.

This is nonsense. Cross-feed is like training-wheels, it helps people get used to headphone listening when they're new to the experience. But after a while, most people can do without it and actually prefer to listen without cross-feed, because they've become adjusted to the way headphones present the soundstage. Moreover, cross-feed always leads to deterioration of the sound, because you feed part of the information of one channel into the other, which causes cancellation of frequencies in that are in opposite phase (and there's also the added circuitry).

[71 dB]

Very few headphone amps include a crossfeed circuit.

Yes, this is unfortunate.

I did find this thread at Head Fi listing some options:
If your listening from a music server, a digital implementation might be an option.

Of course.  For us Mac user's there's Vox player.

I seriously investigated crossfeed when considering a headphone amp to buy but found as many arguments con as pro.  The way it's implemented is crucial and very few of even the most high end h/p amps use it.  I find it's a rare recording that  has such wide channel separation that the soundstage sounds unnatural.  When I come across such, I just use my speaker gear.  In fact many recordings (especially chamber music) sound better in imaging, detail and presence using my headphones than speakers.  The speakers I find always preferable in orchestral music, opera, oratorio, etc.

Well, to my experience nearly all stereophonic recordings benefit from crossfeeding, because there is almost always some spatial distortion depending on how the recording is done. Our hearing is very sensitive to differencies (both level and timing) between left and right ear and our head interprets excessive differencies as spatial distortion. Crossfeeding removes something from the sound, but it is not musical detail, it's spatial distortion that of course changes in time according to the music and therefore might seem as musical detail. But it is not. It's like saying harmonic distortion is musical detail. Of course it's not. It's distortion!

I usually prefer stronger crossfeeding with orchestral music. That might explain your preferencies. Spatial distortion makes orchestral music VERY messy. Anyway, chamber music needs some crossfeeding too.

Crossfeeding is essential. You can always turn it of when the miracle recording is playing.

[Henk]

This is just something for technicians not for people who want to enjoy music. I'm satisfied with my mp3 player with (cheap) sennheiser ear buds.

I have a DAC, but my mp3 player is a nicer experience. I use the DAC of course to save space. I want the room to fill with music, so no headphones. If I want I can use the headphone amplifier feature on my DAC. I think no cross-feed. I think Foobar has a crossfeed feature.

One option left: cd's through headphone which I rarely do. Here again no crossfeed needed. My ears are used playing with mp3 player without crossfeed.

Crossfeed makes the sound worse, just admit it.

[71 dB]

This is nonsense. Cross-feed is like training-wheels, it helps people get used to headphone listening when they're new to the experience. But after a while, most people can do without it and actually prefer to listen without cross-feed, because they've become adjusted to the way headphones present the soundstage. Moreover, cross-feed always leads to deterioration of the sound, because you feed part of the information of one channel into the other, which causes cancellation of frequencies in that are in opposite phase (and there's also the added circuitry).

Sorry, but you have got it wrong. I listened without crossfeeding almost 2 decades because I was stupid and ignorant with it. Two years ago I found crossfeeding and it revolutionized my headphone listening. So, what you say is bs. For almost all recordings, uncrossfeeded sound is unnatural. It contain level and phase differencies much larger than can be heard with natural sounds around us.

What you say about "deterioration of the sound" happens always with loudpeakers. Why aren't you worried about that?

Opposite phase: weaker crossfeeding doesn't cancel much. Statistically recordings contain phase differencies equally of all angles. Statistically nothing is canceled. Recording containing strong opposite phases are idiotic. Do you even have those? Getting rid of spatial distortion is 1000 more important than worrying about opposite canceling.

Crossfeeding happens only at low frequencies, below 1000 Hz or so. Bass doesn't sound real unless it's nearly monophonic.

Crossfeeding actually reveals detail because more detail is revealed under spatial distortion. This is perhaps the point people don't get about crossfeeding.

[Drosera]

Sorry, but you have got it wrong. I listened without crossfeeding almost 2 decades because I was stupid and ignorant with it. Two years ago I found crossfeeding and it revolutionized my headphone listening. So, what you say is bs. For almost all recordings, uncrossfeeded sound is unnatural. It contain level and phase differencies much larger than can be heard with natural sounds around us.

Not to argue too much from authority, but you're basically saying that all engineers in the high-end headphone amp game have it wrong. Because I can't think of single true high-end headphone amplifier that has crossfeed, and yet, implementing cross-feed, even a good implementation, is hardly rocket-science. Also, it appears you simply have a strong preference for cross-feed,  because headphones weren't quite working for you before this. Yet, what I get from headphones is a solid, consistent and evenly distributed 'headstage' ( to use the jargon) and cross-feed diminishes this both spatially and tonally.

What you say about "deterioration of the sound" happens always with loudpeakers. Why aren't you worried about that?

Feeding two signals into each other is very different from to loudspeakers producing actual full-spectrum sound waves.

Opposite phase: weaker crossfeeding doesn't cancel much. Statistically recordings contain phase differencies equally of all angles. Statistically nothing is canceled. Recording containing strong opposite phases are idiotic. Do you even have those? Getting rid of spatial distortion is 1000 more important than worrying about opposite canceling.

Not much? Way too much for me. And just a thousand times?  Again you seem to be stating personal preference rather than fact.

Crossfeeding actually reveals detail because more detail is revealed under spatial distortion. This is perhaps to point people don't get about crossfeeding.

Well, that would be amazing. Magic cross-feed. Are there schematics for this somewhere? I may be very sceptical, but I will always remain a curious fellow as well (in more than one way).

[71 dB]

Not to argue too much from authority, but you're basically saying that all engineers in the high-end headphone amp game have it wrong. Because I can't think of single true high-end headphone amplifier that has crossfeed, and yet, implementing cross-feed, even a good implementation, is hardly rocket-science. Also, it appears you simply have a strong preference for cross-feed,  because headphones weren't quite working for you before this. Yet, what I get from headphones is a solid, consistent and evenly distributed 'headstage' ( to use the jargon) and cross-feed diminishes this both spatially and tonally.

Yes, headphone amps without crossfeed are kind of wrong (otherwise they can be excellent).

For example, Meier-audio's headphone amps have crossfeed.

Crossfeed makes the soundstage narrower and that's good because without crossfeeding most of the sound comes from your shoulders near your ears and that is simply wrong. Crossfeeding "bends" the sound forward, away form your ears were it belongs. If you get "evenly distributed 'headstage'" without crossfeed, your recordings are strangely headphone optimized. Mine aren't, because most of them are made for louspeakers without much consideration of headphones.

Crossfeeding doesn't diminish tonally anything. It dimishes spatial distortion in brain. You lose garbage with crossfeed and that's great. If it did diminish tonally, all sounds we hear would be dimished tonally since all sounds around us are acoustically crossfeeded to our ears.

Feeding two signals into each other is very different from to loudspeakers producing actual full-spectrum sound waves.

Yes, different. The crossfeed-effect is almost the same since loudpeaker crossfeed "uses" real HRTF, while headphone crossfeed is approximation of HRTF. Loudspeaker sound is damaged by room acoustics, headphone sound isn't. Even in acoustically good listening room loudspeaker sound is affected by the acoustics much more than headphone sound by crossfeeding.

Full-spectrum sound waves? 20-20000 Hz? Well, some loudspeakers do that. So do some headphones. Crossfeeding has nothing to do with this.

Not much? Way too much for me. And just a thousand times?  Again you seem to be stating personal preference rather than fact.

We can do the math: Lets say we use -6 dB crossfeeding. Ipsilateral sound (ref. 0 dB = 1). Contralateral out-of-phase sound (ref. -6 dB = -0.25). Sum of these signals = 1 - 0.25 = 0.75 (ref. -2.5 dB).

Now, this is an extreme example since real life recordings just don't contain equal magnitude opposite phase signals below 1000 Hz (analyse your recordings if you don't believe me).

What your room does to the bass frequencies is much worse than 2.5 dB, about 10 times worse actually and it happens all the time!

Well, that would be amazing. Magic cross-feed. Are there schematics for this somewhere? I may be very sceptical, but I will always remain a curious fellow as well (in more than one way).

Nothing magical about it really. Detail hidden by spatial distortion is revealed when spatial distortion is removed. It's logical. I have many schematics of different kind of crossfeeders because I like designing and constructing them. The smallest one is for my iPod (just a 0.56 mH coil and resitor between left and right channel - works very nicely considering the extreme simplicity), but here's the schematics for my current headphone adapter with 6 crossfeed levels:

[DavidW]

Not to argue too much from authority, but you're basically saying that all engineers in the high-end headphone amp game have it wrong. Because I can't think of single true high-end headphone amplifier that has crossfeed,

Is that the one true scotsman fallacy?  You define high-end headphone amps as the headphone amps that don't include cross feed?  Meier, Headroom and FiiO all include cross feed and they encompass the range from high end to midrange to entry level.

Feeding two signals into each other is very different from to loudspeakers producing actual full-spectrum sound waves.

And the sound produced by dozens if not hundreds of instruments in a concert hall is very different from two point sources in a small room, but that doesn't stop me from using my speakers to listen to symphonic music.

Not much? Way too much for me.

Crossfeed uses interference from waves separated by a fraction of a wavelength, the interference is mostly constructive.  What is "way too much" is just a slight decrease in amplitude.  Enough to be audible, not enough to be distractingly destructive to most people.

Well, that would be amazing. Magic cross-feed.

It's not magic, it's science.  It is very easy for sounds to be what is called masked by other sounds.  Just see our discussion in the Haydn thread about the use of harpsichord in the symphonies for an example of masking.  If a recording is mastered to be played back on speakers, you might have sounds not masking say a flute playing softly due to destructive interference with sounds produced by other instruments.  Without that destructive interference, those instruments might drown out that flute.  The recordings are mastered for speaker playback, so it will sound right on speakers but not right through headphones.

Many recordings sound just fine on headphones without cross feed.  But I have plenty that do not.  Not for the subtle effects mentioned above, but just  poor imaging.  I can't stand the Angeles Quartet performing Haydn on headphones because it sounds very asymmetric between each headphone.  And the recordings sound perfect through speakers.

Now if you don't care, you don't like cross feed that is your choice... but you should know that there is a logical, rational reason for the use of cross feed.

[71 dB]

Spatial distortion is like graffiti sprayed on a crumpled silverscreen. Crossfeeding makes the screen flat (natural spatial cues) and clean (free of spatial distortion). The sound "projected" on such screen is smooth, steady and serene. Crossfeeded music sounds like it is located in silent, intact space instead of noisy and cracked space.

Crossfeed uses interference from waves separated by a fraction of a wavelength, the interference is mostly constructive.  What is "way too much" is just a slight decrease in amplitude.  Enough to be audible, not enough to be distractingly destructive to most people.

Yes. Crossfeeded signals are delayed by 200-300 µs (7-10 cm or 3-4 inches). At 800 Hz, a typical cut off frequency of crossfeeding, the wavelength of sound is about 43 cm or 17 inches. 200-300 µs timing difference corresponds angles of about 25°-35° for arriving sound (same as with loudpeakers).

Crossfeeding doesn't really decrease amplitude, unless the recording is really perverse, in which case nobody in their right mind can listen to it, crossfeed or not. Music is a very complex and alternating signal. One has to do the analyses statistically when calculating what crossfeeding does. Statistically half of the stuff is "off phase" and the other half "on phase", so it's zero sum in practise. Well, it isn't that simple, since audio channels tend to correlate, but these issues have been "taken care off" in crossfeed designs. In my design I showed, the 1.5µF capasitor and the two 240 Ω resistors do this performing subtle treble boost effect. The benefits of crossfeeding are clearly massive compared to these "destructivity" worries.

Many recordings sound just fine on headphones without cross feed.  But I have plenty that do not.  Not for the subtle effects mentioned above, but just  poor imaging.  I can't stand the Angeles Quartet performing Haydn on headphones because it sounds very asymmetric between each headphone.  And the recordings sound perfect through speakers.

Even those recordings that sound just fine as they are usually benefit from weak crossfeeding. Graupner's Partitas for Harpsichord on Naxos (8.570459) doesn't need crossfeeding.

I don't have that Haydn disc, but it probably would sound good with strong crossfeed (louspeakers mean strong acoustic crossfeeding).

[G. String]

So, do you prefer neutral or smooth and warm???