JM skrev:petersteindl skrev:JM skrev:Griesinger beskriver i flera av sina presentationer närvarokänsla. Enligt honom är denna subjektiva upplevelse näst intill digital. Psykologiskt finns närvarokänslan i ljudet i en konsertlokal från första bänk och bakåt i salen för att abrupt försvinna upplevelsemässigt i en bestämd position i lokalen.
Fysikaliskt är det när kvoten på reflex-/direktljudstyrkan går över ett visst värde som avgör grovt förenklat. Även tiden har påverkan. Närvarokänslan från detta avstånd och vidare bakåt i lokalen försvinner och ljudet upplevs som långt borta.
I våra små rum finns alltid närvarokänsla enligt hans definition.
JM
Då har Griesinger knasig definition på närvarokänsla, eller du har inte förstått det han skriver.
Mvh
Peter
Skulle uppskatta om tar reda på fakta innan du sågar något. Du har uppenbarligen noll koll på Griesinger och närvarokänsla.
Att dessutom påstå att jag inte har fattat vad Griesinger skriver utan att du själv har har en aning om vad han skriver är uppseendeväckande.
Förslag att du tänker till innan du skriver och efter du skrivit läs med eftertanke samt tänk en gång till.
Fakta och referenser till Griesinger finns i massor i mina tidigare inlägg.
JM
Jag har de flesta av hans artiklar. På det sätt du beskriver närvarokänsla enligt Griesinger så beskriver du snarast närhet. Det är två skilda saker. Griesingers artiklar ses inte alltid som vederhäftiga och helt korrekta i vetenskapliga sammanhang. De är ok, men man får läsa med extra förstoringsglas och lite salt. Dessutom har han vid tillfällen ändrat och korrigerat sin uppfattning om saker under resans gång. Han var ju pappa till den digitala efterklangsenhet med varumärke Lexicon. Lexicon började i början på 70-talet. Lexicon ingår sedan 1993 i Harmankoncernen och numera i Samsung. Där fanns ett visst jäv. Idag använder de slogans som '
Immersive Processor' och '
Immersive Audio Experience' som ord i sin marknadsföring. Jag har inte sett att han använder närvarokänsla på det sätt du beskriver det. Därför gör det mig lite förvånad.
Jag skall ge ett exempel från en av hans artiklar.
David Griesinger skrev:Several tests have been developed to help determine reverberation quality.
1. Plots of reflection density versus time. My preferred way is to count the number of echos within 20 dB of the largest echo, using a sliding window size of about 20ms. This function can be very different for different algorithms.
2. Plots of energy (integrated impulse response) as a function of time.
3. Plots of the autocorrelation function of the reverb output when excited by a click. Before doing the autocorrelation, the reverberator output should be multiplied by an exponentially increasing function, so the total level remains relatively constant as the sound decays. This plot is very good at showing repetitive behavior in the reverberant decay. See K.A. Law.
4. Sonograms - plotting the frequency content as a function of time in a reverberator as the sound decays. Again, the reverberator output should be multiplied by an increasing exponential. These plots are very good at detecting frequencies which have longer reverberation times than others. This technique was developed by MR. Schroeder.
5. Plots of the reverberation time of each mode as a function of frequency. In general, different modes of a reverberator can decay at different rates. At the end of the decay only the longest-lived modes remain, and the sound can be thin and metallic.
6. Plots of the steady-state amplitude response as a function of frequency. This is helpful in determining the modal density and can point out if there are regular gaps in the reverberant modes.
7. All these tests can be tedious. The best test is to simply listen to the algorithm, using a combination of clicks, boings, and music.
HOW DOES IT SOUND?
Lets start with “Acoustic" sounds, such as room sound and concert halls. It is important to realize that the best reverberation for recorded sound may be different from the best live concert hall acoustics. Reverberation in halls and rooms has considerable spatial and directional information which is poorly reproduced by conventional stereo recordings, and this has a major impact on understanding the relation between architectural acoustics and recorded acoustics.
ACOUSTIC SPACES
A few general suggestions can be made from the impulse response of actual halls:
First, the overall complexity or density of the echogram is important, especially when comparing different types of artificial reverberation. Schroeder proposed that the echo density should be at least 1000 reflection/second if the reverberation was to sound natural. A quick look at figure 1 shows this hall easily meets this criterion, almost from the very start. In addition, the echos in figure one are probably not a precise copy of the original impulse. Since many or most reflecting surfaces are small and/or rough there will be considerable smearing of the reflections in time, corresponding to various kinds of filtering in frequency. Digital reverberation has difficulty providing enough density, and even more difficulty smearing each reflection differently in frequency.
Second, the general shape of the energy content in the first 200ms is important. Figure 2 shows the impulse response of a hall at different source-receiver distances. Note that reverberant energy comes in several distinct groups, but in general builds to a maximum 30 to 40 ms after the direct sound. The patterns formed by these fluctuations are very important to the perceived size and spaciousness of the room. In fact, simply changing the time scale of the echos in the echogram can vary the apparent size from a large concert hall to a small room. The apparent size and reverb time might be the same in different seats, but the clarity of the music might be quite different. Artificial reverbs with these echograms might be perceived as having different spatial properties while the music was playing but might be quite similar when the music stopped.
Third, the effective reverb time while the music is playing (which is most of the time) is determined by the initial rate of decay of the impulse response, say from max to about -15dB. The classical reverberation time as formulated by Sabine is usually of secondary importance.
Fourth, many of the desirable properties of halls, described by such words as spaciousness, spatial impression (SI), and envelopment, depend on directional properties of the reflections in the first 200ms, and it is not clear how these properties translate to artificial reverberation from stereo loudspeakers. However, it seems to be very important that a stereo reverberator should have uncorellated impulse responses at the two outputs.
Fifth, reverberation for recording differs from reverberation for live listening in the importance of early (10 to 50 ms) reflections. In halls early reflections allow the musicians on stage to hear each other and add to the loudness and intimacy of the sound to the audience. However, most engineers find that these reflections, usually from the ceiling, floor, and side walls of the stage enclosure, can seriously muddy recorded music. Engineers try to place sound sources and microphones to minimize these reflections, and the common practice of moving the orchestra off the stage onto the floor of a concert hall for a recording is a good example.
Sixth, the reverberant decay at time delays greater than 200ms is important too, there are also complications here. Decay in actual rooms may not be either exponential or smooth. Any rooms exhibit a longer reverb time at the end of the decay than at the beginning or may have an obvious pulsation as sound bounces from the back to the front which never really smoothes out. Depending on the algorithm used, an artificial reverberation device may have other problems as the sound decays, either insufficient diffusion, unevenness, or a metallic quality. Some electronic reverberation devices allow the user to select unnatural final decay rates, either much shorter than you would expect from the apparent decay rate in the first 15 dB of decay (the running reverb), or much longer. These adjustments (especially making the stopped decay short) are very useful and popular.
Seventh, the time density of reflections is very important to the subjective quality of the reverb, especially if sharp percussive sounds are present in the music. As discussed before, in natural acoustics a reflection is almost never really completely discrete. In good spaces walls are seldom perfectly flat or parallel, and reflections are always somewhat smeared out in time. Unfortunately, in a digital reverberation device, reflections are usually very discrete. If the device is excited by a sharp click, the resulting multiple clicks can sound very unnatural, giving a grainy, almost distorted sound to drums or woodblocks. The density of echos necessary to remove this graininess depends both on the bandwidth of the machine and the abilities of the listener, but can be much higher than Schroeder’s proposal, up to 10,000 echos/second or beyond. The higher the bandwidth of the reverberation device the greater the density must be. Techniques used to add diffusion (density) to the electronic reflections add problems of their own, such that a device set for high diffusion might sound good on one type of music, and one set for low diffusion might sound better on another type of music.
In natural spaces the time density of reflections starts at some value determined by the complexity of the wall surfaces and builds up at a rate which depends on the size and shape of the space. Even the initial reflections are smeared enough to lose the sharp grainy quality of a precise digital delay. By the time the reverberation is 15 dB down the density will most likely be so high individual reflections will be impossible to distinguish by ear. Most digital reverberation algorithms used today do not increase the time density fast enough, nor do they continue building the density as the sound decays.
Detta får representera det sätt som Griesinger skriver. Jag ser inga problem med denna text.
Det är dock väldigt stor skillnad i sätt att uttrycka saker i jämförelse med det du skrev. Därför är jag lite förvånad.
Mvh
Peter
VD Bremen Production AB + Ortho-Reality AB; Grundare av Ljudbutiken AB; Fd import av hifi; Konstruktör av LICENCE No1 D/A, Bremen No1 D/A, Forsell D/A, SMS FrameSound, Bremen 3D8 m.fl.