Since 2005 I have been working on the reconstruction of ancient theatre acoustics in order to verify which effect it had on a sound, what conclusions about ancient Greek culture follow from it, and also how modern acoustics can benefit from these results. The field is vividly investigated as you can check in the results of Nico Declercq and Cindy Dekeyser "Acoustic diffraction effects at the Hellenistic amphitheater of Epidaurus: seat rows responsible for the marvelous acoustics", published recently in Journal of Acoustical Society of America or the material in Nature based on their work. Quite a few European Union projects were devoted to the study and reconstruction of the ancient theatres, which applied also to their acoustics. Check for instance ERATO project. Work of Prof. Rindel's team on the topic (Aspendos theatre modelling) was presented at 18th International Congress on Acoustics in Kyoto 2004.
The methods in my research are more complex following the general guideline from Prof. Rindel's work that the more complicated models are closer to the real measurements. My focus was also to transfer the simulation result into the signal processing alorithm that would yield an experimental way to validate the conjecture that the Greeks have something to say on acoustics and sound arrangements despite the relatively poor technical means they had at disposal. Other improtant contribution was to focus on time-frequency effects as they are known in modern psychoacoustics to have a crucial impact on the subjective impression of the music.
As the core material for analysis we certainly used the findings of the
acousticians working in the field and gathering the measurement results
but we also kept in mind that the important source of information is "De architectura. Libri Decem"
by Vitruvius, who in Book V gives quite precise
account how a theatre with good acoustics has to be built adding a
special emphasis that the way it's built results in the sound
propagating in the optimal way and providing most of the pleasure to
the auditorium.
With
Stefan Hagel
from Austrian Academy of Sciences we worked out the
written evidence of Vitruvius with archeological findings combining
this with the analysis of impact on the processed signal. It was
important to
keep in mind that the acoustics quality may have to be tuned to the
specific type of the music played and performed at the times.
|
Example: Tune played on aulos* by Stefan Hagel. * Aulos reconstructed by Stefan Hagel. |
Same tune as heard from the first row of ancient theatre. |
Set-up. The size of the ancient theatre is assumed to be that of Epidaurus. The shape of the sitting area, too. The theatre is assumed to be built from 10cm thick wood and to have a closing wall behind the stage. Other details follow description from Vitruvius' Book V. Model. To obtain the impulse response of the ancient theatre enclosure the diffusion process was assumed to accurately describe the propagation of the digital impulse. Hence, the energy attenuation and time delay of the propagated sound were calculated as functionals over Wiener process trajectories. This model is more consistent with the reverberation measurements performed on-site in Aspendos by Prof. J.H. Rindel's team than the ray-tracing models.
The software CEZAM showing the acoustical effect as if the source was positioned over the edge of the scene of the ancient theatre and
the listener was seated in the centre of the first row is available for download below.
Download:
Main file.
Library .
Batch file.
Sample WAV file.
You place all the files in the same directory and double-click the
batch file. You can also run the program from MS-DOS command line.
It is not recommended to stop the program when operating.
For an input file you can use any stereo CD quality WAV file. The
software is free, offered AS IS and operates till the end of month.
After this date you need to check this webpage for its availability.
You can also try the following
main file (2) substituting the previous one. It combines the acoustical effect of the ancient theatre
with the original sound to obtain the less diffused sound than in the previous case.
Requirements: You need a dual core processor and around 2GB RAM to process 6 minutes stereo WAV file.
Should you have any problems with downloading the above EXE files you can
Download ZIP:
Main file. Library. Batch file.
Download ZIP:
Main file (2). Library. Batch file.
Should you have any queries address me directly at
pwoj@impan.pl.