I think that wouldn’t be the best in all cases. For true spatial audio we need the same spatial distribution of the mirror sources as occur in the recording room. By certain circumstances the best adaptation would be graafs super speaker, which would fail completely when a dry recorded sound field must become restored. In such case a strongly directed radiation would deliver the best results. We would need a variable and steerable directive effect.
H.
gedlee said:
From technical view, it would be possible to restore the mirror sources of the first recording room reflections in its correct positions by means of a synthesis from a mono signal, yet.
H.
Helmut, don't want to make this a discussion about wave field synthesis but how would a real world recording and mixing situation for WFS look like?
An unknown number of channels each carrying a mono sound source and a data channel that carries position/acoustical parameters of the room they're in? Then the WFS signal is calculated in realtime at the end users side?
Or will the end user receive all channels pre-mixed like in traditional stereo/surround? How many channels would that be?
Best, Markus
An unknown number of channels each carrying a mono sound source and a data channel that carries position/acoustical parameters of the room they're in? Then the WFS signal is calculated in realtime at the end users side?
Or will the end user receive all channels pre-mixed like in traditional stereo/surround? How many channels would that be?
Best, Markus
No, Marcus,
mixing down would be establishing phantom source based procedures again, no chance for true spatial audio. But it isn’t a problem to carry 24 or 32 mono audio channels and according data today. Such MPEG 4 based transmitting in the computer world widely spread, more as 32 source positions nobody can separate. Real time would become possible if the procedure becomes reduced up the first reflections instead of the horizontal plain. Many recording problems, unsolvable until, wouldn’t occur for the dry recording of each source. Establishing the spatial sound field during the reproduction process in the same way as the recording room does it seems me the only possibility for true spatial audio, what should be a better place for discuss such thoughts?
Best Helmut
www.holophony.net
So today I have set up my amp and lousy bookshelf speakers in my living room - and I moved the loudspeakers and listened to music a lot🙂
I have some points here:
1. As long as there was any emotion present in the music, it did not matter how I listened to it.
2. I liked speakers at ear height firing up to to the cellar - I love that spaciousness be it HiFi or not.
3. More or less same result with the SLS - but loss of high frequencies was apparent, the sound became imbalanced. Maybe a mono front firing tweeter is not as bad idea as it might seem at the first look.
4. I got a great deal of spaciousness with standard stereo configuration, but the speaker axes crossed before me. I actually prefered this to SLS because there was no hi end loss (as expected).
5. Ceiling firing setup somehow smooths out the response of my speakers, which are a bit unpleasant while listened on axis.
6. SLS and CFS both preserved positions (albeit imaging is not pinpoint sharp) of the instruments/singer while walking around the room - IMHO the greatest advantage to standard stereo setup.
7. Ceiling firing setup is the winner for me (and my music) - I have to find ideal position and height in the room, so that my wife accepts it and I am done🙂
I tried as many different music as I could starting from BW audiophile CD sampler over Bach organ, chamber orchestra, Wagner to my favourite rock/metal recordings (Nightwish, Metallica and that kind of stuff). Some of it in FLAC, some in 192 kbps MP3 (I do not have my CDs moved yet).
All with a Sansui 2x20 W integrated amp, HP laptop as sound source.
I have some points here:
1. As long as there was any emotion present in the music, it did not matter how I listened to it.
2. I liked speakers at ear height firing up to to the cellar - I love that spaciousness be it HiFi or not.
3. More or less same result with the SLS - but loss of high frequencies was apparent, the sound became imbalanced. Maybe a mono front firing tweeter is not as bad idea as it might seem at the first look.
4. I got a great deal of spaciousness with standard stereo configuration, but the speaker axes crossed before me. I actually prefered this to SLS because there was no hi end loss (as expected).
5. Ceiling firing setup somehow smooths out the response of my speakers, which are a bit unpleasant while listened on axis.
6. SLS and CFS both preserved positions (albeit imaging is not pinpoint sharp) of the instruments/singer while walking around the room - IMHO the greatest advantage to standard stereo setup.
7. Ceiling firing setup is the winner for me (and my music) - I have to find ideal position and height in the room, so that my wife accepts it and I am done🙂
I tried as many different music as I could starting from BW audiophile CD sampler over Bach organ, chamber orchestra, Wagner to my favourite rock/metal recordings (Nightwish, Metallica and that kind of stuff). Some of it in FLAC, some in 192 kbps MP3 (I do not have my CDs moved yet).
All with a Sansui 2x20 W integrated amp, HP laptop as sound source.
Why is that? Like in surround sound you would have a dedicated number of channels in the control room and at home. Or is the number of channels (i.e. speakers) room dependend?
Best, Markus
...but your ceiling fires back on the same time, regardless the recording room was 3 or 13 meters in high. Or your wife can elevate Your living room ceiling sometimes a little?
Regards Helmut
www.holophony.net
Although I dont understand all what you are saying, but from my practical work with speaker crossover I am in serious doubt that things are as "simple" or "squared" as you tend to explain things
I very much feel that there is a very specific point when soundstage, ambience, spaciousness and other related terms becomes very much true and real and seem to be totally free from artefacts related to the listening room
Though I would like to understand the science behind it
I think it a "key-word" could be something like "truthfull recognition"...tho I know, we dont all expect the same thing
I very much feel that there is a very specific point when soundstage, ambience, spaciousness and other related terms becomes very much true and real and seem to be totally free from artefacts related to the listening room
Though I would like to understand the science behind it
I think it a "key-word" could be something like "truthfull recognition"...tho I know, we dont all expect the same thing
by the realized wfs plants by speaker rows all around the listener the conventionally records become rendered in “virtual panning spots”. That’s are virtually sound sources (no phantom sources!), aligned far beyond the real playback room walls. By that way the sweet spot covers almost the whole playback room, because changing of the listener position hardly changes level or angle regarding the virtual panning spots. But those are working in the same manner as real loudspeakers, phantom source based.
Best Helmut
www.holophony.net
I must say I do not really care about any scientific/non scientific explanations, imaging sharpness, real space feeling , neither do I know what is really going on, the ceiling reflected sound gives me music with some kind of spaciousness and the most uniform sound cover of the room without significantly degrading sound balance, the speakers just disappeared acoustically - and that was my goal - a pretty good result with a pair of lousy speakers with limited space and volume. A multiway horn stack can do it much better in a good room (I've heard that) - that is out of my reach right now mainly due to space constraints.
Hi tinitus,
thank you because you wasn’t audible until for my. 😉
It isn’t an mysterious since behind the wfs. It moves a little air. In principle you can understand the procedure roughly without reading any text from those two little animations alone:
http://www.holophony.net/Principle of wave field synthesis.htm
http://www.holophony.net/sound field transformation.htm
H.
This really isn't "loudspeaker perception" but "soundfield perception". The concepts of multichannel are well established and interesting, but really beside the point IMO.
I have heard numerous implimentations of different effects, much akin to what is being talked about here. In almost every case the playback suffered from poor loudspeaker quality. What is often forgotten is that EACH speaker in multichannel needs to be high quality or the effect is seriuosly degraded by the loudspeaker colorations. This poses a real problem for the consumer in terms of resources - both the space required as well as the cost. This will always seriously limit the market acceptance.
When I hear a 5.1 or 7.1 system set up with poor speakers, such as Bose cubes etc. I am always disappointed because of the colorations etc. I far prefer two channels of great quality to five of poor. But I can't afford the space or the cost of 7 Summas.
To me the current implimentations of 7.1, where the surround channels are not usually mixed for full spatial effect (more like an augmentation of the main channels and location effects in film) and hence their "quality" is not as critical as the front channels is a good first step towards market acceptance because it is practical to impliment. However, I also don't find 7.1 to be a quantum leap forward in effectiveness. In other words, two channels gets you 80% there, the center channel helps a lot adding perhaps another 10%, or 90% there, but the next four channles only adds maybe 5%. It should come as no surprise that there is a serious diminishing return with the number of channels, but yet, done correctly the costs scale linearly. The "value" of this proposition is dropping fast and I seriously doubt that in the long run the marketplace will accept this low value solution to a problem that they may not even recognize to begin with.
For me two channels of superb quality are going to rule for a very long time and it is this application that I have choosen to focus on.
Multi-channel will become a main feature of commercial installations where the space and costs can be absorbed by the larger audience. But this will be a very tough sell to get into the homes of the consumer.
The bottom line to me is how many "premium quality" loudspeakers can be afforded in any given siutation because lower the quality of the mains to afford more speakers is NOT a good compromise. Better to limit the number of channels than the quality of the sources.
I have heard numerous implimentations of different effects, much akin to what is being talked about here. In almost every case the playback suffered from poor loudspeaker quality. What is often forgotten is that EACH speaker in multichannel needs to be high quality or the effect is seriuosly degraded by the loudspeaker colorations. This poses a real problem for the consumer in terms of resources - both the space required as well as the cost. This will always seriously limit the market acceptance.
When I hear a 5.1 or 7.1 system set up with poor speakers, such as Bose cubes etc. I am always disappointed because of the colorations etc. I far prefer two channels of great quality to five of poor. But I can't afford the space or the cost of 7 Summas.
To me the current implimentations of 7.1, where the surround channels are not usually mixed for full spatial effect (more like an augmentation of the main channels and location effects in film) and hence their "quality" is not as critical as the front channels is a good first step towards market acceptance because it is practical to impliment. However, I also don't find 7.1 to be a quantum leap forward in effectiveness. In other words, two channels gets you 80% there, the center channel helps a lot adding perhaps another 10%, or 90% there, but the next four channles only adds maybe 5%. It should come as no surprise that there is a serious diminishing return with the number of channels, but yet, done correctly the costs scale linearly. The "value" of this proposition is dropping fast and I seriously doubt that in the long run the marketplace will accept this low value solution to a problem that they may not even recognize to begin with.
For me two channels of superb quality are going to rule for a very long time and it is this application that I have choosen to focus on.
Multi-channel will become a main feature of commercial installations where the space and costs can be absorbed by the larger audience. But this will be a very tough sell to get into the homes of the consumer.
The bottom line to me is how many "premium quality" loudspeakers can be afforded in any given siutation because lower the quality of the mains to afford more speakers is NOT a good compromise. Better to limit the number of channels than the quality of the sources.
pelanj said:
But I think that a lot of the rest of us do. To me, what you "don't care about" is really "what its all about".
Did that answer my questions? I'm not sure. Is it possible to have WFS working with a dedicated number of speakers (i.e. channels) or will the number of speakers always depend on the (size of the) room?
Best, Markus
Hello gedlee,
consider, say thousand of poor speakers aligned each by other in a plain doesn’t remain poor speakers. The huge resulting membrane is adapted beyond comparison better in its air load resistance. Very small membrane elevations remain for High sound pressure down to the wavelength of the resulting diameter. On the other hand around 4 kHz would be sufficient on the upper limit, above phantom source supplement by two very good ribbons would be possible. 4kHz today is a bearable task for poor speakers.
Markus,
I haven’t understood your question because the wfs isn’t a channel orientated procedure, but a object orientated. The number of speakers is independent from the number of sound sources.
It is only caused by the limits of the depiction range, which appointed the size of the loudspeaker row or field. For avoid aliasing effects you must the speakers spacing each other in less than two inch in principle, in other case tonal artifacts become audible.
The speaker rows from your picture (Technical University Berlin) are currently the world largest with 2.700 single speakers. I would need for the 2,7m living room speaker field behind the screen wall 1.296 pcs.
H.
syntheticwave said:
I don't know how much you know about array processing, spatial aliasing etc. but its not a trivial task. And if your saying that 4 kHz is all the higher we need to go then I totally disagree. The concepts of large arrays of small transducers have been arround for a very long time and we still haven't seen any implimentations that have been a commercial sucess. I suggest that you are far more optimistic than I am and I can see little evidence on which to base such an optimism.
Helmut, understood. So there'll never be a "real" recording of a source within an acoustic environment like a concert hall but only dry monophonic recordings played back in an electronically simulated virtual room on the end user side.
On the other hand "classical" stereophonic recordings could be played back by simulating 2 speakers within an user-defined virtual room. The same is true for multichannel recordings. Are there any systems available? I only know the Beyerdynamic Headzone that is capable of doing that for headphone playback.
Best, Markus
On the other hand "classical" stereophonic recordings could be played back by simulating 2 speakers within an user-defined virtual room. The same is true for multichannel recordings. Are there any systems available? I only know the Beyerdynamic Headzone that is capable of doing that for headphone playback.
Best, Markus
I don’t know if you know OPSI, optimized phantom source imaging approach for wfs. It was developed on the German IRT institute and shows, because the ITD evaluation becomes ambivalent doesn’t audible, if you change to some simply phantom sources for the amplitude based detection.
I’m so optimistic because I am know, by the phantom source based procedures tonal accuracy is all what obtainable, true spatial audio will never become possible.
Markus, hope i understand what you mean. My idea was to including the playback room acoustics into the synthesis by a model based approach. The playback room acoustics hardly remain audible by such way; the principle is patented in Germany.
Now you can put into those virtually recording room two or fife virtually loudspeakers. That’s the same effect as Bayer does it by HRTF related changes. But it would work in acoustically untreated rooms only in the near field of my loudspeaker wall.
H.
markus76 said:
Hello Markus!
My experience with real stereo recordings and directive speakers are quite good, in fact better than with "uncontrolled reflective" speakers. The only problem for me was that they are very sensitive to changes of position and turning the head. What I would try with such a ton-of-speakers-array is creating just two plain wavefronts, quasi two speakers in infinite distance in a non-reflecting room.
Oliver
- Status
- Not open for further replies.