Indeed. Just how much time did all that puffing take?
I would guess far less than yours
Ambiophonics helps in overcome some of the fundamental problems of stereo triangle, the consequences of interaural crosstalk and gross HRTF errors. By doing that it is already much better than stereo triangle. It's just that stereo is so bad any other method is easily better.
Another thing is ITD, and the fact that not almost any conventional speaker in a normal room reproduces ITD cues in such plausibility that stereo would not sound artificial. Typically ITD cues are conflicting with other cues, or are totally ambiguous. In nature ITD cues are considered dominant. The lack of ITD cue coherence with other cues is one of the main reasons stereo triangle in a room does not sound realistic, I think.
- Elias
I would guess far less than yours
..about 20 minutes for me.
I didn't have a problem with the content of your post, but the format (with the spliced-up quotes) read more like a p!ssing match.
Next time consider just one quote (or maybe two).
Are we talking about the same thing? Here's Dolby's take on object based audio:
http://www.dolby.com/uploadedFiles/...by-Atmos-Next-Generation-Audio-for-Cinema.pdf
Ambiophonics overcomes gross HRTF errors?? Instead of delivering everything through ±30° it delivers everything through 0°. Is that less erroneous?
Are you sure stereo itself doesn't sound artificial because of low accuracy of the wave field and inability to deliver pivoting cues? Did you try listening under anechoic conditions, e.g. outside?
Last edited:
SRS & object based audio
Hi there: Found a picture of SRS's version of"OBA". Google-up Orange County Register. They posted a photo of SRS dmonistration system, sems to use 6 computer sized front speakers and 4 pole mounted boxes about .75cf mid room. Great WAF! Looks like this is another market driven hitec gambit. ...regards, Michael
Object based audio is coming and terms like "multichannel" or "binaural" as separate and incompatible techniques won't be relevant anymore. People can chose any renderer they like, even if it's two dipole speakes
Hi there: Found a picture of SRS's version of"OBA". Google-up Orange County Register. They posted a photo of SRS dmonistration system, sems to use 6 computer sized front speakers and 4 pole mounted boxes about .75cf mid room. Great WAF! Looks like this is another market driven hitec gambit. ...regards, Michael
Originally Posted by markus76
Object based audio is coming and terms like "multichannel" or "binaural" as separate and incompatible techniques won't be relevant anymore. People can chose any renderer they like, even if it's two dipole speakes
I totally agree with you here. I will be trained on mixing for Atmos in two weeks. Kind of excited.
Dolby still has to tweak the system, and there are some auditoriums who shape will introduce some spatial error, but I know Dolby will figure all of this out. Dolby theater here in San Francisco had some problems with Atmos during their demonstration.
Last edited:
Not true. The system can be used with any soundtrack from any period. Just like Hollywood is 3D' ing everything, you can bet they will use Atmos for a "A" list vintage classic title.
..about 20 minutes for me.
I didn't have a problem with the content of your post, but the format (with the spliced-up quotes) read more like a p!ssing match.
Next time consider just one quote (or maybe two).
Sometimes a p!ssing contest can just be an honest disagreement of thoughts and facts. I guess it depends on your perspective.
Object based audio is independent of the renderer. Could be a mono speaker in front of you.
They will probably do that but honestly it does a disservice to object based audio if they don't do a complete remix or use sounds that can't be handled as objects.
The inclusion of "beds" in Atmos is also a very bad idea in my opinion. It's there for practical reasons but as much as it helps maintaining old workflows, it hinders or slows down adoption of new workflows that would make use of the new possibilities, e.g. accurate reverberation simulation with reflections coming from multiple angles.
Yep.. how funny it is to see how poor an artistic content can bear these new technologies. It's a sad shift from a medium amongst others to... a mean in itself!
As a result, I am not going to the theatre anymore, maybe five times a year total. I used to spend my days in them.
That is not actually right, the recommended speaker angle is not 0deg but something between 5-15deg.
And yes it is less erroneous. If you take propability distribution of the intended image span angle of stereo recordings, it is more likely the image is within, say, +-10deg than +-30deg. Then it can be said HRTF errors are statistically reduced.
Stereo triangle ITD errors are related to erroneous wavefield. Head movements make them more apparent, but they exist also with fixed head. The room is a dramatic ITD error contributor as well.
- Elias
And don't tell me that mono is better.
Mono is better than stereo triangle in some aspects
The wavefront of a single mono speaker is more natural than the wavefront of a stereo triangle, because the velocity vector of propagating wave is unity as in a real sound source.
I understand that. But "better than stereo triangle in some aspects" doesn't make it better on the whole. I'd rather listen to something not perfect while waiting for the next audio revolution.Mono is better than stereo triangle in some aspects
Don't know about the statistical distribution of phantom sources but is less HRTF deviation perceptually less severe?
Given the fact that a) virtually no binaural recordings exist and b) the majority of all recordings work with interchannel level differences that are greater than interaural level differences of natural sounds, is it sensible to look at stereo as a binaural reproduction technique at all?
Oh come on Elias, velocity vector of propagating wavefront is speed of sound. Sound field formation is wave superposition.
No standards in stereo: Recordings are mixed/monitored with approximate equilateral triangle, and then listened to with remote headphones using a partition, and all sounds better? Just a variation on a theme of how brain responds to two sounds correlated in time and disjointed in space. Really, for everyday listening?
5.x 7.x 9.x sound: Each speaker pair forms phantom image for correlated signals. If one poor phantom image doesn't work well, a bunch of them will?
Imaging performance works best with point source. The degree to which radiating wavefront of speaker projects back to single point for all frequencies, the better the speaker's imaging performance will be. Theory of omnidirectional speaker obviously seeks this goal. Dipole and cardioid in theory also project to point. In theory, some waveguides do, many don't.
Speaker transducers with excited break up modes project back to a distribution of sources that sum to point that is state driven, and thus bounces about in space.
Ripple in frequency response, even in anechoic chamber, with small shifts in microphone demonstrate deviation from point source behavior, and is good starting place for assessing point source character of drivers/speakers.
Clearly the IMP presented more point like source than Behringer or Orion. Test wasn't about timbrel accuracy. Test was about detail of phantom image.
IMP is crappy box, but shading of drivers, and identical drivers gives it better point source characteristics.
Baffle width of Orion, midrange size, crossover point to tweeter, and spacing to tweeter limit phantom image detail. At distances where angular aperture becomes narrow, direct/reflected ratio is poor and room sound buries low level spacial detail in recording.
Behringer has narrower baffle, but crappy box.
Surprising results? Not when the physics are understood.
No standards in stereo: Recordings are mixed/monitored with approximate equilateral triangle, and then listened to with remote headphones using a partition, and all sounds better? Just a variation on a theme of how brain responds to two sounds correlated in time and disjointed in space. Really, for everyday listening?
5.x 7.x 9.x sound: Each speaker pair forms phantom image for correlated signals. If one poor phantom image doesn't work well, a bunch of them will?
Imaging performance works best with point source. The degree to which radiating wavefront of speaker projects back to single point for all frequencies, the better the speaker's imaging performance will be. Theory of omnidirectional speaker obviously seeks this goal. Dipole and cardioid in theory also project to point. In theory, some waveguides do, many don't.
Speaker transducers with excited break up modes project back to a distribution of sources that sum to point that is state driven, and thus bounces about in space.
Ripple in frequency response, even in anechoic chamber, with small shifts in microphone demonstrate deviation from point source behavior, and is good starting place for assessing point source character of drivers/speakers.
Clearly the IMP presented more point like source than Behringer or Orion. Test wasn't about timbrel accuracy. Test was about detail of phantom image.
IMP is crappy box, but shading of drivers, and identical drivers gives it better point source characteristics.
Baffle width of Orion, midrange size, crossover point to tweeter, and spacing to tweeter limit phantom image detail. At distances where angular aperture becomes narrow, direct/reflected ratio is poor and room sound buries low level spacial detail in recording.
Behringer has narrower baffle, but crappy box.
Surprising results? Not when the physics are understood.