The ST measures without the standard reference level

I have uploaded results for ST within octave bands without the standard reference level (previously not presented, see the Articles page. The reference level for ST is found to vary significantly between the stages studied, making it difficult to draw comparisons between the stages regarding levels of reflected sound levels on stage.

This article also includes results for STlate within octave bands (previously not presented).

More details on measured STlate

I have uploaded results for STlate within octave bands (previously not presented, but available as part of the project data zip file), see the above post.

Project data available

The project data from the stage acoustics project covered on this site is now freely available, see the Project data page for more information.

Gunshots, noise off – and music

When acosusticians investigate the acoustic response of rooms they normally excite the room in such a way that they are able to study the room’s acoustic response to a perfectly impulsive sound.  This response is called the room’s echogram or reflectogram. Based on this echogram a set of acoustic measures can be calculated, for instance the reverberation time. Calculating the reverberation time is based on observing the long decay after a long-lasting sound source is switced off. In the past balloons were popped off, gunshots were fired off and/or loudspeakers switched off to obtain the relevant responses. With modern measuring techniques the same responses are obtained by use of loudspeaker and taylor made excitation signals like sine sweep. Most of the time the loudspeaker as well as the microphone are omnidirectional since this makes it easier to obtain reliable results among different investigations and between acousticians.

On the other hand music rarely contains perfectly impulsive sounds or long-lasting constant sounds being switched off (the ‘noise off’ method). Apart from at very high frequencies musical instruments excite the room over a certain time, leading to reflections from the room fusing together. Addtionally, the instruments are highly directional and we hear sound in stereo, not mono. Perceptual effects like masking and the cocktail party effect appear relevant for a listener inside a reflective room. Accummulated early reflections as the music is running can contribute to mask the direct sound and make it difficult to separate different sound sources spatially. This can as proposed by David Griesinger make the sound appear more muddy and the listener will not feel engaged in the sound. Since localisation is affected by the shape of our head and outer ear, the extent to which a sound appears defined or fully meddled can only be judged by an individual listening inside the actual room.

Based on this we may claim that measurability and reliability has been given priority compared to validity, when assessing room acoustic spaces. Focusing mainly on measured characteristics may have limited the scope and approaches when searching for relations between architectural design and experienced conditions. My impressions based on the results from the PhD work is that acoustic measures are relevant to see if the proposed design is ‘in the ballpark’, to ensure that the most catastrophic results can be avoided – formally or just to ensure the end users are happy with the acoustics.

One paradox within room acoustics may be that even if absolute acoustic levels are widely used within other fields of acoustics, like formal limits for sound pressure levels from traffic noise, it is not (yet) an established single measure that is always included when assessing listener and performer acoustics. The acoustic measure Strength (G) represents the absolute acoustic gain by room, but is not always measured/reported. For stage conditions, absolute levels are measured, but by use of a different reference level (the Support ST measures).

In my view, the acoustic measures studied for performance spaces should at a minimum include T, EDT, G and C80, both in the audience area and on stage. For the audience area spatial measures are also relevan (like LFC and LEV). From measured G and C80, Gearly and Glate should also be found to provide an indication of how loud the reverberation is, not only the reverberation time (also see ‘How loud is my reverberation‘ by Griesinger).

In addition to obtaining these measures it appears highly beneficial to listen a lot to rooms with own ears, and as a substitute, listen to auralisations of potential responses for planned venues. The acoustic measures have not been developed much since the ’70s and to progress further I believe experiencing real spaces  and discussing them are very useful to test and develop new hypothesis regarding links between physical acoustic conditions and experiences.

By having a common understand of basic acoustic concepts among for instance acousticians, sound engineers and musicians it will be much easier to discuss and exchange ideas and experiences based on qualified listening in real rooms (with ensemble on stage, audience present, directivity taken into account etc etc). Such qualitative discussions may be a very fruitful complement to objective acoustic data. If musicians and sound engineer also can understand the essential use and relevance of acoustic data I believe we are heading towards exciting discussions and new exciting findings within the discipline of stage acoustics.

Get the basic measures right and be aware of their limitations.  Then make sure to have enough time and interaction to listen, discuss and explore acoustic spaces with open ears and an open mind! I believe this will make it easier for us (and more enjoyable as well?) to effectively create great acoustic spaces for everyone involved.

Stage and ensemble geometry related to masking effects

I have written a new article providing some more details on stage and ensemble geometry related to masking effects. The main focus is how the size and proportions of an acoustic ensemble set requirements to the size and proportions of the stage enclosure or the room the ensemble performs in. Some background information with an overview of different stage enclosure designs in the vertical plane is also give. See the Articles page.

Christopher Blair on stage acoustics

At the time my PhD thesis was completed Christopher Blair, both an acoustician and conductor, gave his views on acoustic conditions for symphony orchestra on a blog posting on the Adaptistration web site. Several other interesting blog postings by Blair are also available on this web site. I was unaware of his posting on acoustics for orchestras when I completed my thesis, so it appears that Blair’s views on beneficial conditions for an orchestra are independent from the results in covered in my thesis.

In the blog posting referred to above Blair writes: “The art of designing good on-stage acoustics boils down to providing just enough early energy to help with coordination, but not so much as to mask audibility of the late-energy room response.” This is in very good agreement with the conclusions from my PhD thesis: the main purpose of early reflections back to the orchestra was found to an effective compensation for low direct sound level within the orchestra. Additionally, the most popular venues covered in our research project had a certain balance between early and late acoustic response on stage (C80), as well as a certain level of the late acoustic response in the main auditorium (Glate).

Blair has previously written about the importance of reflections from the main auditorium reaching the stage, in the Journal of the Conductors Guild, Vol. 19, No. 2, 1998. In this article (unfortunately not freely available) the relevance of perceptual masking effects for perceived conditions is described, along with a study of plan and section drawing for a set of existing symphony orchestra venues. For popular stages there are identified reflections from the back section of the auditorium that reach the stage with delay(s) within typically 100–300 milliseconds. Unfortunately this article was not included in the literature review of my PhD thesis, but the conclusions in this article are also similar to my conclusions.

Improving the acoustics for orchestras under a low ceiling

One of the major findings from the research project documented on this website was that low ceiling can be very problematic for a symphony orchestra. Recently a paper by Higini Arau-Puchades (Arau Acustica) has been published in Acta Acustica united with Acustica; Increasing the Acoustic Volume of Performance Spaces without Altering the Internal Dimensions. The cases covered in this paper suggest that a low ceiling is the major reason for difficult acoustic conditions experienced by the orchestras playing in their specific venues. The paper also includes a design of a grid of vertically oriented solid panels suspended above the orchestra, called an acoustic labyrinth. The picture below is taken from the mentioned paper.

The apparent effect of this construction is to block the direct reflection from the ceiling particularly above 500 Hz. Blocking the direct path results in the sound travel a longer path from the orchestra via the ceiling back down towards the orchestra, resulting in lower level and larger delay on the ceiling reflection. The grid also contributes to scatter the ceiling reflection into different directions. The ceiling of the actual venues had curved ceiling that focused the reflected sound down towards the orchestra. Blocking the direct reflection by the grid will in such cases have a very significant effect on the level of reflection sound from the ceiling.

The orchestras reported on a significantly improved acoustic conditions with the grid installed. This is very encouraging since it suggests that there can be a cost-effective alternative to improve conditions for orchestras under a low ceiling. Raising the ceiling physically will in most cases be too expensive to be feasible.

There are only given a few suggestions in the paper to why the grid improves the perceived conditions. Apparently the experienced improvement can largely be explained by the findings from my PhD thesis; a high ceiling was here found beneficial for avoiding loud instruments becoming too loud and contributing to make the string section and the acoustic response from the main auditorium audible to the whole orchestra. This conclusion was based on comparing acoustic conditions with the orchestra present on stage, related to the psycho-acoustic effects masking, precedence effect and the cocktail-party effect. The real stages studied in the PhD also supported this conclusion.

It will be interesting to hear more about the players’ experiences with Arau-Puchade’s grid in the future. To further understand the effect of the grid it would be very interesting to see measured acoustic responses in detail and results for Glate and C80, both on stage and in the main auditorium for these venues – not only Schroeder curves, T30, G and STearly on stage.