Not sure what you're saying. If the ground bounce would be removed then one would get in-head localization under anechoic conditions?
No. At least I don't think so. And practically speaking it was removed from around 100 Hz up when I did it. It's just that I haven't been able to remove any freq.s *lower* than that so I can't be absolutely sure in my belief that you would NOT get in-head localization under complete anechoic conditions.
..you know, that .009% caveat. 😉
People here made statements that have a 99.991% certainty of being false so I appreciate your thoroughness 🙂
Did you make a virtual setup using your CD speakers? Is it better than the studio everyone uses (AIX)? Have you tried outdoor measurements?
I don't want to post off-topic stuff here, but I think that integrating headphones/nearfield speakers is a key path in a 'putting the science back' way forward. With computers becoming faster, they offer so many DSP benefits that you can't take advantage of when the acoustics are not controlled. When convolution is offloaded to a GPU it can be done in realtime. If 2 ch stereo is as flawed as reading this forum makes it seem, and the way forward is a choice between including headphones or else very many loudspeakers, then I think that headphones will win. Even Audyssey is making a system that combines headphones and speakers (Personal Surround, I think it's called).
A real conundrum isn't it.
(from out in left field) Might the focus to accomplish what you suggest lie not in the loudspeaker(s) alone but in the electronics as well?
A completely automatic, wireless , variable state processor .. coupled with a set of loudspeakers who's placement is flexible enough to help contour/map a room acoustically.
SEdwards,
Since the majority of headphones are going to be used in mobile applications, meaning you aren't sitting in one place, how it that going to work with the need for a loudspeaker to go with your headphones. This is the point nobody seems to want to answer, what are all you guys talking about? Where are you going to use this science. You seem to be ignoring the elephant in the house, how are you going to change an entire industry from the recording process to the final implementation of how we listen to music? Two channel stereo isn't going anywhere, this just seems to be an intellectual discussion and nothing more at this point.
Since the majority of headphones are going to be used in mobile applications, meaning you aren't sitting in one place, how it that going to work with the need for a loudspeaker to go with your headphones. This is the point nobody seems to want to answer, what are all you guys talking about? Where are you going to use this science. You seem to be ignoring the elephant in the house, how are you going to change an entire industry from the recording process to the final implementation of how we listen to music? Two channel stereo isn't going anywhere, this just seems to be an intellectual discussion and nothing more at this point.
You're of course right that two channel stereo isn't going anywhere and we can't steer the industry. For me, the science is about doing what I can as a DIYer with what they give me. If I'm going to have a sweet spot somewhere then I may as well put something on my head there that can enhance the sound more than $xxxx worth of additional drivers. I see the physics side as more limited than the DSP/psychoacoustics side. (If I were rich I may think differently). I admit that this isn't strictly in the spirit of the thread and doesn't address the elephant.
SEdwards,
I am glad to see your answer. Now to your point, even if you can do this as a DIY concept where are you going to get the source material to listen to? Are you planning to make your own recordings or do you think that you can synthesize the music from a previously recorded two channel recording. How do you implement any of this in real life?
I am glad to see your answer. Now to your point, even if you can do this as a DIY concept where are you going to get the source material to listen to? Are you planning to make your own recordings or do you think that you can synthesize the music from a previously recorded two channel recording. How do you implement any of this in real life?
Well, it doesn't have to be one extreme or the other!
I still like the concept of Ken Kantor's Magic loudspeaker. A highly directional front firing speaker gets you the direct sound with minimal room interaction. Side firing speakers with time delay give a good proportion of lateral sound but with sufficient time delay to stay clear of coloring the direct sound.
You can build it into the speaker as Ken did, build it into a multichannel playback system (ambience sysnthesis) or record it live in a good concert hall and reproduce at home...as long as you have more than 2 channels.
In the early days the term "a window on the sound" was considered the highest compliment for a mono playback system. More recently it has become a derogatory remark for concert halls of poor design (inadequate lateral energy) that you are listening to the musicians as if through a window.
David S
Object based audio combined with headphone playback and (individualized) HRTFs could be a major advance. Innovation is nowadays driven by the movie industry.
That would be a nice DIY effort. Thinking of multiple synergy horns...
Already available: Audyssey DSX.
I watch a lot of 5.1 content so that part is fairly straightforward (also for videogames), where I would mostly be deciding on crosstalk and reflection content.
For stereo, I will will need to experiment. In the first instance, I would try some kind of ambio setup. Frontal HRTF cues are most important if you're using headphones, I think. I listen to all of my music using a PC and Reaper is already in my audio chain. I have the convolution setup already, but haven't constructed many 2 in 2 out impulse responses, as yet. I can say that I was able to virtualise a center bookshelf speaker quite nicely. Not hifi yet, though.
Note that I lie down when I watch things, so my head doesn't move much. Eventually I will need to implement head tracking to get closer to the Realiser. Still, I find that even stock Dolby Headphone starts to stick to my projector screen after a short whiile. Visual cues are so strong, I think that they should be exploited more for music.
So the overall realisation is -
- audio from application
-> Reaper
-> convolve with impulse responses
-> send to headphone/speaker outputs
For stereo, I will will need to experiment. In the first instance, I would try some kind of ambio setup. Frontal HRTF cues are most important if you're using headphones, I think. I listen to all of my music using a PC and Reaper is already in my audio chain. I have the convolution setup already, but haven't constructed many 2 in 2 out impulse responses, as yet. I can say that I was able to virtualise a center bookshelf speaker quite nicely. Not hifi yet, though.
Note that I lie down when I watch things, so my head doesn't move much. Eventually I will need to implement head tracking to get closer to the Realiser. Still, I find that even stock Dolby Headphone starts to stick to my projector screen after a short whiile. Visual cues are so strong, I think that they should be exploited more for music.
So the overall realisation is -
- audio from application
-> Reaper
-> convolve with impulse responses
-> send to headphone/speaker outputs
SEdwards,
Okay so for most of what you are talking about it is a movie or game track you would be using as of now. I would think that you would need an excellent set of headphones, not a mediocre consumer version, thinking of how bad the "Beats" headphones really sound. So we are talking about a Sennheiser AKG or something of that quality to actually get the full effects of what you are talking about.
And Marcus,
With you also talking about trying to get all this realism and localization from your system I have to say anyone using the synergy horns for high fidelity is barking up the wrong tree there. That is fine for PA but I do not consider that hifi in any respect, no way is the frequency response even close to smooth enough, eq all you want you will not get there with those horns. You are trading point source coherence for poor frequency response, There is no way you are going to come even close to +/- 1db across the frequency band and you have to use a sub on the bass end unless you have a huge and I mean huge horn to have any real bass response. So as I said this is purely a theoretical exercise at this point as you are not even close to your own requirements. Go ahead and show me a real frequency response, not smoothed of that device over the full frequency range. I dare you. And not with a cheap microphone, let's see an instrumentation mic show the true output of a Synergy. I understand it for PA but not for what you are talking about as if it can come close to a perfect device in an anechoic chamber. I have written with Tom Danley and I think that at this point he would concur on the actual frequency response. Without that you are not being real here.
Okay so for most of what you are talking about it is a movie or game track you would be using as of now. I would think that you would need an excellent set of headphones, not a mediocre consumer version, thinking of how bad the "Beats" headphones really sound. So we are talking about a Sennheiser AKG or something of that quality to actually get the full effects of what you are talking about.
And Marcus,
With you also talking about trying to get all this realism and localization from your system I have to say anyone using the synergy horns for high fidelity is barking up the wrong tree there. That is fine for PA but I do not consider that hifi in any respect, no way is the frequency response even close to smooth enough, eq all you want you will not get there with those horns. You are trading point source coherence for poor frequency response, There is no way you are going to come even close to +/- 1db across the frequency band and you have to use a sub on the bass end unless you have a huge and I mean huge horn to have any real bass response. So as I said this is purely a theoretical exercise at this point as you are not even close to your own requirements. Go ahead and show me a real frequency response, not smoothed of that device over the full frequency range. I dare you. And not with a cheap microphone, let's see an instrumentation mic show the true output of a Synergy. I understand it for PA but not for what you are talking about as if it can come close to a perfect device in an anechoic chamber. I have written with Tom Danley and I think that at this point he would concur on the actual frequency response. Without that you are not being real here.
The responses on Danley's site are unsmoothed and a fair bit flatter than the majority of hifi speakers that I've measured. Have you any first hand experience with a Synergy setup?
Just talking to Tom himself and seeing the response curves he has provided. They are not smooth in any sense, that is nonsense. Point source yes, but the devices used through a Paraline and the chambered midrange will be high Q, not flat. Been designing horns for way to long to believe that for a second.
@Kindhornman, yes, you need some nice headphones with a well behaved response for this DSP business. More important than the shape of the response is that it's smooth and the decay is fast. I use Fostex T50RP drivers in my headphones (one pair is DIY open backed, one closed). Smyth like Stax, which are too pricey for me. I agree 100% on the state of popular mainstream models and their horrendous sound.
RE: horns, has anyone tried a synergy horn shape but as a wedge on the ground? I'm considering putting a 3" FR on the apex of a floor corner horn for these HRTF measurements.
I also listen from the floor a lot so maybe a floor horn would be a worthwhile experience anyway.
RE: horns, has anyone tried a synergy horn shape but as a wedge on the ground? I'm considering putting a 3" FR on the apex of a floor corner horn for these HRTF measurements.
I also listen from the floor a lot so maybe a floor horn would be a worthwhile experience anyway.
That's something I don't know and I haven't seen any unsmoothed data of a synergy horn yet. Maybe Dan can chime in?
I wouldn't want to use anything else than multiple subs. Stereo bass in acoustically small rooms is a myth. There are directional effects that are caused by modes but this is far from something that could be called stereo bass with controlled phantom bass sources lining up within a few degrees in front of the listener.
Markus,
I am not here to knock anybody's speaker of choice or implementation whether horn or direct radiator or any other method of reproduction. And in your last post you have said something revealing in that when the frequencies are in the lower range we lose the ability to determine directionality, our rooms just won't allow that. So what we are talking about are the directional cues from higher up in the frequency band. And beside a very narrow and directional device such as a directional horn we just can't truly control the room response in a typical room. I don't know about you but I do not have the room or the money to build an anechoic chamber in my house. I don't think many of us do, and it would not exactly help the resale value of our homes if we did that! So what it seems is that you are trading one type of sound implementation for another and then have to deal with the limitations of what you have created. I wish there was a perfect answer to all that we would love to have, which is the sound of a live performance in our own homes, but if you have every had a band rehearsing in your home you would soon tire of that also. SEdwards idea of trying to do this with a set of headphones is probably the most practical for the effects you are looking for if we just had perfect headphones and we actually only heard through our ears and did not have bone conduction and other effects that effect our hearing and what we have come to expect in a live music performance.
Steven
I am not here to knock anybody's speaker of choice or implementation whether horn or direct radiator or any other method of reproduction. And in your last post you have said something revealing in that when the frequencies are in the lower range we lose the ability to determine directionality, our rooms just won't allow that. So what we are talking about are the directional cues from higher up in the frequency band. And beside a very narrow and directional device such as a directional horn we just can't truly control the room response in a typical room. I don't know about you but I do not have the room or the money to build an anechoic chamber in my house. I don't think many of us do, and it would not exactly help the resale value of our homes if we did that! So what it seems is that you are trading one type of sound implementation for another and then have to deal with the limitations of what you have created. I wish there was a perfect answer to all that we would love to have, which is the sound of a live performance in our own homes, but if you have every had a band rehearsing in your home you would soon tire of that also. SEdwards idea of trying to do this with a set of headphones is probably the most practical for the effects you are looking for if we just had perfect headphones and we actually only heard through our ears and did not have bone conduction and other effects that effect our hearing and what we have come to expect in a live music performance.
Steven
You do get a marginal degree of bass "separation" from a traditional full-range loudspeaker, but it's rarely noticeable.
However, in the near-field it is usually noticeable (..recording dependent). (Ex. Listening chair with L & R subs as side tables with drivers pointing at the listener.. rather like headphones.)
Oh, also you can "synthesize" it by using multiple summed mono subs and phase manipulation (..to essentially set-up a phase-induced low pressure "divide" down the middle of the room/listening position).
Last edited:
- Status
- Not open for further replies.