International Round Robins
It is now quite some time ago that the “Round Robins” on room acoustics software were carried out. In the three Round Robins, ODEON was ranked as one of the “unquestionable reliable” software. Since the last 3rd Round Robin in 2002 ODEON has continuously improved e.g. the reflection based scattering and diffraction around edges have been developed, making it easier for the user to obtain reliable results because both developments remove some of the guessing involved in the room acoustic modelling, by adding algorithms that automatically utilizes the geometry of the room model to generate scattering or diffraction.
Danish small Round Robin
Anyhow, in 2008 the Danish Acoustical Society made a small Round Robin using only ODEON as software and 8 different users. The conclusion was that guessing especially concerning extremely low or extremely high absorption-coefficients can be problematic in an acoustic model; where this actually is no problem if the acoustician uses a traditional Sabine calculation assuming a totally diffuse field. Download report in PDF
With the experience from Danish investigation we added a new button in the Material list called “limit absorption coefficients”. This button makes sure that all absorption coefficients added to the room model is automatically adjusted to be in between 0,05 and 0,95. Our advise is that you use this button if you are not sure whether the absorption coefficients are really that low or high. Especially in cases with one or two dimensional modes describing the reverberation time, this limiting of absorption coefficient can be necessary. Furthermore you should always make a quallity check of your model e.g. using this check list.
Make your own judgment
But to come back to the question we recommend that you test the reliability of ODEON by yourself like described in the video above. Make your own test of the software e.g. in the free demo version. Open the rooms that all ready contains measurements in the free demo version (Examples shown below). Look at how the simulation fits with the measured values in the multipoint response. Then go to the room setup and change the setup e.g. the number of rays used in the model and investigate what influence this has on the result. You can try different stability tests like this and please ask if you have any questions.
See our Cases here