[pinnahertz]

Hopefully this post showed good attitude from my part. It has been so bad.

You've swapped arrogance for sarcasm. Very transparent, but not better.

 

[pinnahertz]

That's more or less what was teached to me in the university, but it's nonsense according to sound engineers who know best.

Name one "sound engineer" who says it's nonsense.
Oops, sorry, the "sarcasm light" was still on.

 

[pinnahertz]

I'm not a sound engineer so I know nothing and I am dumb,

Keep that nonsense up and somebody's going to start agreeing with you.

If that is true then binaural recordings are useless. Why see the trouble of making a binaural recording, if the resulting "headstage" is all in your head anyway?

The troll has spoken!

Jecklin Disk is not 100 % binaural, but very close.

No, I wouldn't say "very close." It's not a head, not head-shaped so diffraction is wrong, no pinna (I would notice that, now wouldn't I?) so none of the HF directional mapping is there. All it does is provide a rough approximation of HRTF ILD and ITD with none of the other directional cues.

You have not experienced binaural if that's all you've heard.

 

[castleofargh]

Nonsense. That's what you'd think happens if you don't know better.

Even without crossfeed headphones have some sort of miniature soundstage (as long as the recording has good enough spatial information), just distorted due to excessive ILD. Crossfeed scales ILD so the soundstage becomes undistorted and the sounds take a distance of the head, because large ILD means closer sounds. The miniature soundstage becomes a bit larger.Crossfeed especially move sounds from shoulders to in front of the listener, from a few inches away to 1-3 feet away from ears. With bad recordings the soundstage is bad, but who tells you to listen to bad recordings if you want good soundstage? New multichannel SACD releases of classical music contain well recorded natural spatial information and when downmixed to stereo using Lt/Rt method and crossfed properly can produce amazing soundstage for headphone, almost as good as with speakers, but that only happens with the best recordings. However, even half-decent recordings produce some sort of soundstage crossfed.

I have never experienced anything like this using Xfeed alone. it's possible if you're very lucky with the recording and the headphone's own signature corresponding pretty well with your HRTF at 30° or close enough, but I disagree that it's what people can be expected to experience in general using Xfeed. I would imagine them starting from possibly almost a line between the ears without Xfeed. depending on the signature and head, they can sometimes have depth and height's impressions based on wrong cues feeling right. then using Xfeed, in my case I never perceive instruments at a wider distance on the sides, on the contrary the width tends to reduce noticeably with Xfeed for me while things are "pushed" forward. so I would almost agree with the starting drawing in some cases, but the Xfeed one is science fiction in my experience. to start getting a little bit of such a graph I need reverb, and a much more customized frequency response. Xfeed alone even those where I could change delays attenuation and where to start filtering, wouldn't give me what you show when used alone. again, I'm not saying my experience is law, but at the very least it's one experience that does not agree with your drawing.

 

[TYATYA]

thanks for this write up, i noticed in uapp there is not timming as in neutron

when you say favourable the range you gave from 800 to 3000 is alot, so where should I set it to shift bass more to the left

can you share your exact neutron setting without ranges as the ranges you gave are wide

I set 2300hz and 8.5db

 

[71 dB]

1. Keep that nonsense up and somebody's going to start agreeing with you.

2. No, I wouldn't say "very close." It's not a head, not head-shaped so diffraction is wrong, no pinna (I would notice that, now wouldn't I?) so none of the HF directional mapping is there. All it does is provide a rough approximation of HRTF ILD and ITD with none of the other directional cues.

You have not experienced binaural if that's all you've heard.

1. Well, that would feel refreshing! The thing is I write what I think and I can't help if others find it nonsense. Do you want me to be real me and disagree or fake me and agree? Why should I be what you want me to be in the first place?
2. True, not techically that close, but perceptually surprisingly close. Just shows how spatial hearing can be fooled, if the spatial cues make sense even if the small details are wrong or missing.

 

[71 dB]

I have never experienced anything like this using Xfeed alone. it's possible if you're very lucky with the recording and the headphone's own signature corresponding pretty well with your HRTF at 30° or close enough, but I disagree that it's what people can be expected to experience in general using Xfeed. I would imagine them starting from possibly almost a line between the ears without Xfeed. depending on the signature and head, they can sometimes have depth and height's impressions based on wrong cues feeling right. then using Xfeed, in my case I never perceive instruments at a wider distance on the sides, on the contrary the width tends to reduce noticeably with Xfeed for me while things are "pushed" forward. so I would almost agree with the starting drawing in some cases, but the Xfeed one is science fiction in my experience. to start getting a little bit of such a graph I need reverb, and a much more customized frequency response. Xfeed alone even those where I could change delays attenuation and where to start filtering, wouldn't give me what you show when used alone. again, I'm not saying my experience is law, but at the very least it's one experience that does not agree with your drawing.

Because the sound moves forward, the angle doesn't increace. In fact you need wider sound to keep the horizontal angle the same. So, there's relative and absolute width. In my exprerience the angle stays the same or gets a bit smaller with crossfeed, depending on the spatiality of the recording, but because the sound also moves forward, the absolute width seems to get bigger. A skyscraper a mile away looks big, because you know it's far away. However, a small house in front of you 30 feet away can easily block it.

Anyway, I have learned that despite of the science of spatial hearing many people do like excessive ILD/ITD and want to hear it for artistic intent to be heard. I don't, but that's just me. So, I have been stupid for half a decade for trying to educate people about crossfeed. I stop doing it now that I am wiser. I learned that I have to be very careful about how to use my knowledge and understanding. It makes my knowledge much less valuable, but that's life. It's hard and the you die, whether you used crossfeed or not.

 

[pinnahertz]

1. Well, that would feel refreshing! The thing is I write what I think and I can't help if others find it nonsense. Do you want me to be real me and disagree or fake me and agree? Why should I be what you want me to be in the first place?
2. True, not techically that close, but perceptually surprisingly close. Just shows how spatial hearing can be fooled, if the spatial cues make sense even if the small details are wrong or missing.

1. You were being sarcastic, which is not authentic, it's nasty. Nobody wants that unless it's also humorous, which it wasn't.
2. Have you actually compared the two side by side? Have you ever made an actual binaural-head recording?

 

[pinnahertz]

Because the sound moves forward, the angle doesn't increace. In fact you need wider sound to keep the horizontal angle the same. So, there's relative and absolute width. In my exprerience the angle stays the same or gets a bit smaller with crossfeed, depending on the spatiality of the recording, but because the sound also moves forward, the absolute width seems to get bigger. A skyscraper a mile away looks big, because you know it's far away. However, a small house in front of you 30 feet away can easily block it.

Sounds like there are more than one who do not share your experience with cross-feed. Mine is like castleofargh, it narrows but doesn't move forward. I did manage to create a form of cross-feed that did move virtual sources forward, but it took time delay and EQ (a bit of HRTF) which isn't in the 71 dB version of cross-feed, and was still very recording-specific.

Anyway, I have learned that despite of the science of spatial hearing many people do like excessive ILD/ITD and want to hear it for artistic intent to be heard. I don't, but that's just me. So, I have been stupid for half a decade for trying to educate people about crossfeed. I stop doing it now that I am wiser. I learned that I have to be very careful about how to use my knowledge and understanding. It makes my knowledge much less valuable, but that's life. It's hard and the you die, whether you used crossfeed or not.

Well, that's a first step. The next might be to further your research and find how to what degree cross-feed is accepted/desired or not, then find out why.

 

[bigshot]

You've swapped arrogance for sarcasm. Very transparent, but not better.

I'll take sarcasm over arrogance any day. Sarcasm has a better chance at sounding funny.

Personally, I've never heard anything in a headphone that sounded like it was in front of me. The closest was a binaural recording that sounded like it was a couple of inches away, but it kept snapping from front to back. I couldn't control it. In order to locate sound in front of me, I have to be able to turn my head.

 

[71 dB]

Sarcasm has a better chance at sounding funny.

Agreed.

I have to be able to turn my head.

Do you do that also when you drive and listen to your car audio system? Other people may think you are rubbernecking traffic accidents.

 

[castleofargh]

in any case it's a matter of subjective impressions. I don't know that the same cues are perceived by me the same way they are by another guy using the same headphone. in fact what I've learned over the years is that they're unlikely to feel identical(if only because of the listener's own HRFT, and maybe also his own listening habits he got used to).
so for some cues we know it will be somehow similar. like a sound panned on the left will feel like it's on the left for pretty much everybody able to put on a headphone correctly ^_^. I don't know that we'll feel the sound coming from the same place(again, probably not for a bunch of frequencies because of the size and shape of the head), but we'll all agree it's on the left.
and for some cues, it's just a mess because we're missing way too much data on the signal, the headphone, and the user's HRTF. like vertical cues for example. many people get some even though they're not supposed to be here. and for audiophiles it's even funnier because they will often mention those impressions of vertical "soundstage" as a quality for the headphone/IEM they're using(and I did exactly that myself in some reviews in the past ). showing that fake cues can be assimilated with fidelity when they impress/please the listener. I tend to have mono cues "placed" way up in my mind with a headphone I tried to EQ to my speakers or with the signatures I tend to enjoy while listening to typical stereo albums on headphones. after fooling around a lot, it turns out I can place the mono sound back in front of me on headphones with the right frequency response like suggested by David Griesinger. turns out I have some pretty significant variations to my "neutral" between full mono and 30degree left or right(when playing non binaural music on headphones and no xfeed or anything). this is in part explained with Xfeed and how to attenuate on side, what to filter out, plus maybe some comb effect here and there. the end result is indeed expected to have a different frequency response.

I imagine we all agree at least in principle if not in experience with all I've mentioned so far.

now, "to boldly go where no man has gone before", because we're ultimately dealing with subjective impressions, there is another aspect that's a mystery: what happens when we get 1 cue right and not some others? this IMO could summarize the all Xfeed discussion.
I'm all for an objective approach to sound when it's available, and I'll always favor it. but in practice we often run into a bunch of issues related to that all "dealing with only one or 2 cues in isolation". the most famous of all in audio is probably how the defuse field compensation for headphone should objectively be neutral for the average human ear. we take flat speakers in a nice room, it sounds right and measures fine, we measure that with a dummy head supposed to be the so very average HAT that we place at an ideal listening position and boom we have a signature.
others came so very close to that response using other means (listening tests, anatomical measurements,...). some stuff like the bump around 2.5/3khz on headphone has come to be a consensus.
yet very few people feel like this sounds flat when a headphone is calibrated that way. we have the expected variations from HRTF, an extra variable compared to the average human target, but one we expect and can deal with. but then comes something like the almost universal desire to get more low end that probably comes from a desire to compensate for the lack of tactile bass on headphones. how do we quantify that objectively? how a use tries to turn body vibration into increased low frequency at the eardrum? despite a purely objective approach, the result doesn't work as is because we did not and could not account for all the variables in the system. so what we have is a simplified system and only subjectivity can turn that into something feeling neutral, or not.
@71 dB , IMO when you push Xfeed as an objective improvement, you're doing the same except your model needs to dismiss even more variables to work. so how can you claim that it is an improvement outside of your very own impressions? I have a problem with that and I'm not alone.
we all agree that most albums were made to feel a certain way on speakers in a certain room by some guy. we all agree that headphones used as is to play back such a record is wrong. and of course we all understand how the sound on the left ear should reach the right ear altered in time and FR the way our own head and torso would normally do for speaker sound. this far I believe all the active participants of this thread will agree. but that's as far as we get along. the delay once set for my head will be fine and it's a variable I accept and implement in general when using headphones. then the frequency response, headphone are what they are with nonsense responses, so we need to compensate for that and for our own HRTF. something a typical Xfeed does not do!!!!! there is a generic compensation but it's safe to say that it will not be the proper compensation for most people and typical Xfeed options do not allow for enough customization to achieve it. so we're back to something that is an approximation and how a subject may or may not "fall for it" and find it realistic. often even a partial correction of FR in the right direction is felt as beneficial, but we're already deep in subjective impressions. you can't know how convincing the partial compensation will be to a given user, just like we can't really know in advance how dominant is his eyesight over his hearing. maybe so long as the listener doesn't have at least the singer on a tv screen, he will never feel like there is a soundstage/headstage or whatever. no matter how good the other cues will be. we can also throw in head tracking, as I'm fairly sure it's an important variable for at least some people. when I concentrate on sound I close my eyes, else I can get distracted like a puppy with anything moving in my field of view and I forget anything I was doing. and my second weird thing, if I try to really grasp a detail, I tilt my head(again like a puppy^_^) to get my left ear higher than the right ear and a little forward. on headphones I still do it somehow without thinking when I concentrate on the sound(testing gears or whatever), but of course it doesn't help. on the contrary it probably warns my brain that it's all BS.
of course the elephant in the room is reverb. we're used to reverb, we expect reverb, and even if we tend to fade it out in our head to better be able to focus on the original source of sound(survival tool?), the brain dismissing reverb cues when it suits it isn't the same as not having those cues from the start. and again we're in subjective land here. how will a brain react to missing some cues? will it just focus more on the others, or will it conclude that the sound is artificial and give up on trying to create a proper mental space? the answer is probably in between and I for one do not know if there is real consistency in how one listener will react compared to another one when some cues are wrong or missing. no matter how good some others might be. even the assumption that Xfeed will get us closer to a proper experience is IMO a stretch for that reason. the user, the music, the headphone, how close the Xfeed settings will come to being right for the listener, and of course habits, all those come into play to form the final subjective impressions. and we have people who dislike Xfeed, that's a fact. the unknown is that maybe they never got to try one with settings best for them. maybe then they would revise their opinion? maybe not? maybe they just consider Xfeed to be wrong because they already have assimilated the headphone's usual sound as the reference, and any change feels wrong to them thanks to erroneous reference? it's hard to tell without some vast testing so I won't risk claiming anything aside from IDK. all I can say is that you cannot fully dissociate subjective listening and subjective impressions when discussing Xfeed. I leave the arguments about spatial distortion to those who care, to me it's always all wrong and there is little point discussing how something is less wrong than another wrong given how most stereo albums are only panned, so they don't even bother with all the cues even for the average human head. it's an artistic creation and nowadays maybe some start mixing with stuff involving object based simulation, or with some standard HRTF compensation. but it's safe to say that 99.9% of the albums I own and love were panned by making one side a little louder and that's it. do we really need to argue about the correctness of something that fake from a "spatial distortion" point of view?
I get the idea of the Realiser A8/A16 where you take one specific room and try to mimic the sound for one specific user with one specific headphone. the target is clear, the reference is clear. it's all user specific. that to me makes sense, it's not a matter of spatial distortion, it's an attempt to simulate the experience of one room for one dude. we lack tactile subs and maybe looking at the speakers, but otherwise it's pretty complete. but a generic oversimplified effect like Xfeed, I'm sorry I find the argument that it's an improvement debatable. and again I personally love Xfeed and used some for years. to me it was a step in the right direction for sure, but I wouldn't go so far as to say it's a step in the right direction for everybody or objectively a step in the right direction. too many variables, too many unknowns and way too much subjectivity that we cannot remove from the system.

wow I made a giant post, I'm one of you now \o/. @bigshot you're gonna be left behind if you don't make a huge post too (sarcasm? what is it?)

 

[71 dB]

1. Sounds like there are more than one who do not share your experience with cross-feed. Mine is like castleofargh, it narrows but doesn't move forward. I did manage to create a form of cross-feed that did move virtual sources forward, but it took time delay and EQ (a bit of HRTF) which isn't in the 71 dB version of cross-feed, and was still very recording-specific.

2. Well, that's a first step. The next might be to further your research and find how to what degree cross-feed is accepted/desired or not, then find out why.

1. There are people, who do share my experience including the creator of this thread to who I build the crossfeeder. There are also people who do not share my experience including you. If I have learned anything, it's that people can be divided into two groups: Hyper-dimensionalists and Naturalists. The former don't like crossfeed and the latter do like it.

2. I suppose, but if I am to stop educating people altogether, I might not care about researching these things and instead use my time on other things.

 

[bigshot]

wow I made a giant post, I'm one of you now \o/. @bigshot

I make up for the real estate I take up with my long posts by balancing it out with a few snappy one liners!

 

[pinnahertz]

1. There are people, who do share my experience including the creator of this thread to who I build the crossfeeder. There are also people who do not share my experience including you. If I have learned anything, it's that people can be divided into two groups: Hyper-dimensionalists and Naturalists. The former don't like crossfeed and the latter do like it.

If that's what you've learned, that's disappointing and superficial. You didn't go any deeper than that? Why must you assign labels, especially ones with bias? Which sounds better, a hyper-dimensionalist or a naturalist?

Of course, there is no such thing as a "hyper-dimensionalist", and a naturalist means something quite different from the way you've used it here. So your labels make no sense anyway.

2. I suppose, but if I am to stop educating people altogether, I might not care about researching these things and instead use my time on other things.

The problem with your method of education is that all it is teaching is theory and opinion, not application or perception and acceptance. Some may swallow your propaganda whole, but anyone looking for scientific backup won't. If your aim is to teach, learning must come first. I don't mean this to sound like I know the answers, I don't. But I've suggested some areas to study to learn more and make your teaching more balanced are scientific.

Use your time however you want, though.