How to obtain good values for reverberation time and how different settings affect the result.

Q: How do I make a good simulation of reverberation time in a large room?

A: First consider if the reverberation time is the best parameter to use. E.g. DL2 could be a better parameter for a larger room. Simulating reverberation time in a point response calculation in Odeon, corresponds to measuring the reverberation time. Therefore refer to a measuring standard e.g. ISO 3382. Make sure that you use only one Omni-directional source for each calculation job. In large rooms it is important to have sufficient distance between source and receivers, receivers should not be placed in the near field of the source but in the reverberant far field characterizing the room. ISO 3382-2 includes an equation describing the recommended minimum distance between source and receiver:
Where: V - is the Volume in cubic meters; C - is the speed of sound in meter/second and ^T - is the expected reverberation time in seconds.
Furthermore it is a good idea to use a larger number of source receiver pairs in large rooms than the measurements standard demands, to cover the larger area.
If the room consists of many coupled or unregular areas with different properties, it will may not be possible to find one reverberation time that characterizes the room and we advise you to supplement the reverberation time with analyses of STI, SPL, EDT, DL2 and auralisations in different areas.

Q: When I increase the absorption coefficient in the ceiling reverberation time gets longer?

A: First a general comment on how the reverberation time is calculated from the impulse response. ODEON follows closely the ISO 3382 standard for measurements, i.e. the T30 parameter is found from the slope of a linear regression line that follows the decay curve from -5 dB to -35 dB. In any room the decay curve is a mix of 1-, 2-, and 3-dimensional modes, and generally they have different decays, i.e. the total resulting decay curve is not a straight line, but can be more or less bent. (The 1-dimensional modes have typically the longer reverberation time, and the 3-dimensional modes have the shorter reverberation time).
When the ceiling absorption is high and the walls are reflecting this can give a strong one- and two-dimensional horizontal reverberation. When making the ceiling even more absorbing it means that the one- and two-dimensional sound field gets relatively stronger, and so the reverberation time gets longer. Normally the sound pressure level will decrease. Therefore we recommend that you use auralisation and other acoustical parameters, e.g. STI, SPL and for large rooms DL2 to analyze the acoustics.

Non-diffuse room - example case

Claus Lynge Christensen, Gry Bælum Nielsen, Jens Holger Rindel, "Danish Acoustical Society Round Robin on room acoustic computer modelling". ODEON A/S, 2008. 20 pages.

Q: Which calculation method should I use for reverberation time: Quick Estimate, Global Estimate, Point/Multi point Response or Grid Response?

A: The Quick Estimate should be used in connection with a rough distribution of materials. The Global Estimate ca be used for regular rooms, to estimate the reverberation time for selected source positions to give a fast average over the entire volume.

Normally you need to simulate the reverberation time for receivers in a room like in a measurement. To simulate reverberation time for a range of receiver positions you should use the job list and calculate point responses to simulate a measurement most realistically. generate the reverberation time for certain positions with Point/Multi point response. If the room is an auditorium with an audience, the grid response receivers should be used.

Q: Reverberation time gets longer, when I increase the number of rays in simple rectangular model? (nonlinear decay curves)

A: This is probably due to low diffusion of surfaces in the room, and 1-dimentional reflection paths. The decay curve will not be linear but will have a steeper curve in the beginning and a less steep curve for the later reflections. With nonlinearity in the decay curve a larger number of reflections is necessary to simulate the reverberation time. also coupled rooms with non linear decay curves should have a large number of rays in the room setup. These are good examples of a volume, where a simple Sabine calculation will not be enough to calculate reverberation time.

It can be a help in cases with nonlinear decays to look at the Global Estimate, to analyze how many rays are necessary to define the decay curve (The Global Estimate should a least go down to – 35 dB at all frequencies).

Brief theory and example of RT in non-diffuse rooms.

Q: Does Odeon have T20 - and how does it calculate RT?

A: T20, T10 and several extra parameters will be available from version 11.
The reverberation Time T30 and EDT was selected from the beginning as these two reverberation times is enough to know the room acoustics. However, T20 and T30 are both calculated in the Global Estimate. and from version 11 it will be possible to calculate T20 and many other acoustical parameters in point responses as well.

Reverberation time, (T30) is derived from -5 to -35 dB of the backwards Integrated energy curve according to FDIS 3382-1[6]

See below how you can analyze the decay curves from a point response in the Auditorium or Combined edition. The point response shows all curves in the 8 octave bands, to investigate decays curves in a single octave band further press A. By zooming in on the backwards integrated energy curve (E, Integrated) you can analyze different steps of the decay e.g. to see how calculations of reverberation time based on different intervals of the decay curve give different results.