[71 dB]

Headphone soundstage with great recordings isn't as deep as with loudspeakers, but it in front of listener, outside head nevertheless. Depth is about 1-2 meters. My head is much smaller than that!

 

[RRod]

I think the point is that certain cues are impossible in 'pure' headphone listening, and thus we should avoid a speaker-derived term such as 'soundstage' when discussing headphones. I tend to fall into the 'words have meaning if people know what you mean' camp of linguistics, so I'm fine using the term soundstage on this site, but I can understand the objections.

 

[bigshot]

Headphone soundstage with great recordings isn't as deep as with loudspeakers, but it in front of listener, outside head nevertheless. Depth is about 1-2 meters. My head is much smaller than that!

Do you have a good speaker setup? I'm curious, because I have good headphones and a good speaker system and they present the sound completely differently. Are you just referring to secondary depth cues like sound bouncing off the walls in the recording venue? Because that isn't what I'm talking about. I'm talking about having the sound physically in front of me ten or twelve feet away, not right up against my ears. The depth cues are in my room, not on the recording. Soundstage involves physical space. Angled drivers in headphones won't accomplish that.

 

[71 dB]

I think the point is that certain cues are impossible in 'pure' headphone listening, and thus we should avoid a speaker-derived term such as 'soundstage' when discussing headphones. I tend to fall into the 'words have meaning if people know what you mean' camp of linguistics, so I'm fine using the term soundstage on this site, but I can understand the objections.

Not impossible, but hard to get right. The accuracy of the cues is the problem and that's why the depth and realism is weaker than with loudspeakers which render in the room 100 % accurate real spatial cues.

I listen to my music cross-fed with headphones, because it's idiotic to suffer unnatural spatial distortion ruining the spatial cues there is. Cross-feed bends the sound more forward, so that helps a lot with having depth. In the end it's having good spatial cues for the ears.

 

[71 dB]

Do you have a good speaker setup?

Good is relative term. So, whether my setup qualifies as "good" depends on how you define good. It's a affordable "bang for the buck" system by an acoustic engineer who understands what is relevant in sound reproduction and what is not. Does is qualify as "good" for you is up to you. I think it is good. Not the best system in town, but good.

I'm curious, because I have good headphones and a good speaker system and they present the sound completely differently.

Yes, they present the sound differently. Loudspeakers have bigger and deeper soundstage because the room adds real spatial cues. That's a negative thing too, because the sound is coloured quite a bit, depending on how good acoustics you have. My claim is that headphones can easily produce sound that is outside of head. It's kind of a "miniature" soundstage, but easily large enough to be outside of head. Only sounds without meaningful spatial cues or sounds with spatial distortion are rendered inside head, because brain can't do anything else with the crazy spatial information.

Are you just referring to secondary depth cues like sound bouncing off the walls in the recording venue? Because that isn't what I'm talking about. I'm talking about having the sound physically in front of me ten or twelve feet away, not right up against my ears. The depth cues are in my room, not on the recording. Soundstage involves physical space. Angled drivers in headphones won't accomplish that.

All kind of spatial cues help and spatial hearing is quite easily fooled to have some kind of feel of soundstage. To get to the loudspeaker level is of course really difficult. Yes, with loudspeakers the sound can be 12 feet away. With headphones maybe only 4 feet. That's a 1:3 miniature of the loudspeaker soundstage. Without crossfeed the sound is at your ears because of spatial distortion, but crossfeed makes the sound "jump" from your shoulders to in front of you a few feet. So, good recordings with real acoustics + cross-feed = miniature version of loudspeaker soundstage.

My Sennheiser HD-598 has angled drivers and I think the soundstage with these headphones is convincing, so I'd say angled drivers maybe have some benefits. The angle determines how the sound is reflected from pinna before entering ear. So maybe angled drivers produce more realistic spatial cues? I don't care how the drivers are as long as the sounds is great.

 

[RRod]

Not impossible, but hard to get right. The accuracy of the cues is the problem and that's why the depth and realism is weaker than with loudspeakers which render in the room 100 % accurate real spatial cues.

I listen to my music cross-fed with headphones, because it's idiotic to suffer unnatural spatial distortion ruining the spatial cues there is. Cross-feed bends the sound more forward, so that helps a lot with having depth. In the end it's having good spatial cues for the ears.

Yes but the second you add crossfeed you have left the plane of 'pure' headphone listening, which I would wager is the plane most people run around on on this site ("Let some two-buck DSP alter the sound of the beloved flagship headphone?! Unpossible!"). That means the term 'soundstage' on the site is referring to inter-headphone differences during 'pure' listening, which can't involve anything HRTFy except frequency response. I guess I just feel this is a definitional argument, not an argument about sound virtualization...

 

[bigshot]

It really wasn't a trick question. When I said "good" I meant "not just a boom box".

I've never heard any headphones where the sound felt like it was four feet in front of me. I've heard binaural recordings where a buzz clipper moves around my head, but I couldn't control if it was in front of me or behind me. It kept popping back and forth randomly. It sounded interesting, but it didn't sound real or like it had physical space.

It's interesting that you mention that your headphones sound like a miniature soundstage, because scale is something I've worked on optimizing with my own speaker system. Imagine if you have near field monitors and you're listening to a jazz quartet recorded in a natural left right placement... On the near field monitors you're visualizing a sound stage that is two or three feet across with musicians about six inches tall. It's a bit of a trick to get your brain to reconcile the scale difference because it's so drastic. Now imagine a room with speakers about sixteen feet apart, ranging in height from near the floor to about four feet high. When you sit back about fourteen feet, the soundstage you are visualizing is in human scale. The musicians are six feet tall and the stage is over twenty feet wide. That perfectly matches a small jazz club. Now play a symphonic orchestra recording through the same system. The scale creates the same perspective you would hear sitting twelfth row center in a concert hall. Scale is VERY important to creating a realistic soundstage, but the typical speaker placement chart shows size as a sliding scale based on the perfect triangle theory. I've never found any home audio site that even addresses scale. Very weird.

The best part of speakers is that there is a *natural* coloration to a real world installation that actually *enhances* the sound of the recording. In this case coloration is a good thing. I know a lot of people think good sound is just the sound, nothing but the sound, but with humans that isn't true. We want *present* sound- meaning it sounds like the music is in the room with us. That doesn't fit the "purity" dogma of a lot of audiophiles, but it fulfills the main goal of having a sound system in the first place. Headphones can sound "clean" and they can sound "balanced", but they don't sound "natural" or "present". For that, you need space around the sound. That's why multichannel is so great. It's manipulating the space around the music to create a wide variety of sound fields. That's the next step beyond soundstage. It creates an immersive experience that is even closer to reality. Now of course you don't want auditory masking caused by response imbalances, or phase cancellation or any of those detrimental colorations. But not all colorations are bad. Some are actually good for the sound.

The way I look at it is that headphones are a straight line through your ears. Soundstage is a flat plane like the wall in front of you. A 5.1 sound field is a flat plane with you in the middle of it. And an Atmos sound field is a true three dimensional space. 3D is a pain in the butt to view on TV because you need to wear special glasses. But three dimensional sound doesn't require any special equipment to perceive it. You just have to be able to move your body and ears around relative to the sound sources.

The next logical step is to combine immersive audio with immersive video. Video projection technology has advanced terrifically. The main problem is the same main problem of multichannel audio- how to integrate it into a space you live in and entertain in as well. It's easy to create an immersive A/V experience if you have a blank room and only one chair at the primary listening/viewing position. But if you do that, you might as well slap on headphones and a VR helmet and be entirely inside your own head. I'd rather have something like the "holodeck" in Star Trek where people can inhabit A/V fields together. I don't think that's too terribly far off... and when it is a reality, no one will be pining for the purity of wearing headphones. They'll be creating and inhabiting all kinds of interesting worlds.

To put it in a poetic way, headphones are great for clarity. If you wanted to figuratively be able to count every grain of sand on a beach in Hawaii, you could do it. But if you really want to feel like you're *on* a beach in Hawaii, and feel the ocean breeze and hear the waves and feel the warmth of the sun, it takes the immersiveness of a multichannel speaker setup. To achieve that, you have to cultivate "good coloration" and suppress "bad coloration". You can't ever get to immersive reality by avoiding coloration entirely. I really think that multichannel and DSPs are the future of audiophile sound, not doubling down on purity theories. Cross feed is a good way to get to pseudo soundstage if you have to listen with headphones. But imagine the equivalent of cross feed for a sound field in a good sized room! Yowza!

 

[71 dB]

Yes but the second you add crossfeed you have left the plane of 'pure' headphone listening,

No, with cross-feed you enter the plane of 'pure' headphone listening, because almost all recordings contain spatial distortion. Some recordings (2 % or so) happen to be free of spatial distortion, but cross-feeders have off switch for those rare cases. Unless you want to limit your listening to those 2 %, you need cross-feed to have pure headphone sound and a change to have some sort of soundstage.

which I would wager is the plane most people run around on on this site ("Let some two-buck DSP alter the sound of the beloved flagship headphone?! Unpossible!").

There are many ways to do cross-feed. I use mostly passive components (resistors and capasitors) in my headphone adapter to take care of cross-feed. With proper cross-feed you get everything your flagship headphone has to offer. Without cross-feed you ruin the sound with spatial distortion. Your choice.

That means the term 'soundstage' on the site is referring to inter-headphone differences during 'pure' listening, which can't involve anything HRTFy except frequency response. I guess I just feel this is a definitional argument, not an argument about sound virtualization...

Spatial distortion doesn't make anything 'pure'. Spatial distortion is supposed to be removed by cross-feed. With loudspeakers you have "invisible" acoustic cross-feed. With headphones you don't have that, so you have to do it another way.

 

[RRod]

Yes, and I am saying that on this site most people do not do these things. I myself do: I can't listen to my PM-3 without EQ and crossfeed; can't imagine life without them. But we're discussing why a particular word, soundstage, as it is used on this site, rouses the ire of people who do lots of speaker listening rather than non-DSPed headphone listening. When people say that the HD800 have a 'wide soundstage', they mean without any extra processing of the signal.

 

[WyldeGooseChase]

The way I look at it is that headphones are a straight line through your ears. Soundstage is a flat plane like the wall in front of you. A 5.1 sound field is a flat plane with you in the middle of it. And an Atmos sound field is a true three dimensional space. 3D is a pain in the butt to view on TV because you need to wear special glasses. But three dimensional sound doesn't require any special equipment to perceive it. You just have to be able to move your body and ears around relative to the sound sources.

Sort of wondering about Atmos through Stereo headphones. I downloaded the Atmos demo on Windows Store and through my AKG K551, the demos were boomier but didn't actually seem wider than what i was used to. Also in one of the demos, vocal speaking sounded "distorted" and stretched oddly with Atmos turned on. Yet Dolby seems to think this is how the ideal headphone space should sound... Definitely just sounded like an EQ that promoted the spatial cues, but unnatural.

Also IIRC there are headphones with more than 2 drivers per ear. I dunno how Atmos would affect that but I never thought that having multiple drivers that close to your ear would be any different than typical stereo drivers.

 

[bigshot]

almost all recordings contain spatial distortion.

I'm going to differ on definitions again... Music recordings don't have spacial distortion. They only have rudimentary spacial information- secondary depth cues like phase shift effects and reverb. Stereo recordings aren't supposed to have spacial information. They just have left to right information. The spacial info is supposed to be provided by the speaker placement within the room. There's a standard for speaker placement and almost all music is mixed to suit that standard. If you listen on headphones, you aren't getting that aspect, so you add crossfeed to synthesize the effect of the standard speaker placement in a room. It isn't spacial distortion when you listen on headphones, it's lack of the intended spacial placement for the sound. Music is mixed for speakers, not headphones.

Sort of wondering about Atmos through Stereo headphones. I downloaded the Atmos demo on Windows Store and through my AKG K551, the demos were boomier but didn't actually seem wider than what i was used to.

Also IIRC there are headphones with more than 2 drivers per ear. I dunno how Atmos would affect that but I never thought that having multiple drivers that close to your ear would be any different than typical stereo drivers.

I recently got the new Kraftwerk Catalogue box set which has 7.1 mixes and something called "3D Sound" for headphones. I listened to the surround track on my 5.1 system and it's amazing, with dimensional instrument placement and sound that moves through the middle of the room. I listened to the 3D headphone sound on headphones and it sounded like a little bit of phase was added to make the sound bloom, but it was still left/right with no depth or rear content. I think that emperor has no clothes.

The reason headphones with 2 drivers are set up that way has more to do with optimizing a driver for a particular frequency range than it does directionality. The only way to get directionality is to be able to turn your head while the sound source remains fixed. That's how we perceive distance and placement. Headphones flat out can't do that.

 

[reginalb]

...3D is a pain in the butt to view on TV because you need to wear special glasses. But three dimensional sound doesn't require any special equipment to perceive it...

Not to be a pedant or anything, but 3D sound in a sound system requires more special equipment than does the 3D TV and glasses that most people have. Atmos or DTS-X receivers and speakers, as well as one of the very few movies, and no music that I know of, that have a mix available for the public to actually consume.

 

[71 dB]

I'm going to differ on definitions again... Music recordings don't have spacial distortion. They only have rudimentary spacial information- secondary depth cues like phase shift effects and reverb. Stereo recordings aren't supposed to have spacial information. They just have left to right information. The spacial info is supposed to be provided by the speaker placement within the room. There's a standard for speaker placement and almost all music is mixed to suit that standard. If you listen on headphones, you aren't getting that aspect, so you add crossfeed to synthesize the effect of the standard speaker placement in a room. It isn't spacial distortion when you listen on headphones, it's lack of the intended spacial placement for the sound. Music is mixed for speakers, not headphones.

Spatial (not spacial) distortion is unnatural "contradictory/impossible" spatial information such as large channel separation at low frequencies. Almost all recordings do contain such unnatural spatial information, hence they contain spatial distortion. As you say, music is mixed for speakers, so this spatial distortion disappears when listening to loudspeakers. The problems reveal itself with headphones. Spatial distortion is about having spatial information "outside" the expected value space. Proper cross-feed maps spatial information of the recording into the expected value space making the sound natural. How much the spatial information is outside the expected value space dictates how strongly the sound must be cross-fed to map it inside the expected value space. That's why cross-feeders should have level adjustment to cover different amounts of spatial distortion.

Extreme cases of spatial distortion are for example early "ping-pong" stereo recordings with instruments playing on left or right channel only. A mild case of spatial distortion is for example an classical music recording done with OSS or ORTF microphone set up. A Jecklin disk recording may contain zero spatila distortion!

 

[bigshot]

Ping pong stereo isn't spatial distortion on speakers. It creates a localized direction for a particular sound. There's stuff that occupies the phantom center, and stuff that doesn't. Even mono music is mixed to sound good on speakers and it's just a single channel. You can add a slight reverb and it probably would sound closer to the way they intended on headphones.

I guess I'm looking at it like headphones are missing something from the way speakers present sound, and you're looking at it that since music is mixed for speakers, it's lacking something for headphones. That's certainly true, but speakers sound more like live music, so that's why engineers mix for them. It's the optimal presentation. You don't go to a concert and plug your headphones in to listen. You listen to the concert played to you on speakers.

 

[bigshot]

no music that I know of, that have a mix available for the public to actually consume.

The Kraftwerk set I mentioned before has an Atmos mix, the Beatles Sgt Pepper had a theatrical run in Atmos, and I heard that a new multichannel remix of an INXS album will be too. Atmos music is beginning to hit the market. Unfortunately, my listening room has a high peaked roof with rafters. I don't think I can do Atmos, or I'd be seriously considering it.

I just googled and it looks like there are about a dozen titles in Atmos, from Metallica to classical music.