[Archived phorum post from squ1.com v2 by Andrew]

Marco,

Thanks for your kind comments.

Unfortunately, whilst I have solved most of the problems of sound propogation around and over basic geometry, working out all the likely diffraction pathways through holes and gaps between objects is very time consuming - especially when you start to introduce line and plane sources.

My guess is that people will only really want to use this facility after they have just imported a 50 thousand polygon model of the local business district and laid out a couple of hundred traffic lines - something that with the current algorythm would take about a week on a good PC to resolve.

More work needs to be done, but its certainly up there near the top of the new features list.

Regards,
Andrew

Dr. Andrew Marsh
Square One research Pty Ltd

[Archived phorum post from squ1.com v2 by Casey]

Dear Dr. Marsh,

I'm also doing some basic modelling of acoustics and urban form. I would like to experiment with different typologies, including courtyard arrangements, terraced housing, low blocks of flats, etc in order to work out good design principles in relation to external noise (roads, etc). Right now I'm going through EcoTect to try and figure out whether or not such a thing is possible with the software. I take it from your posting above that it isn't.

Is there a workaround for simple situations? I don't want to go as far as the 'fifty thousand polygon model' you mentioned. Just a simple mass of housing and a sound source outside. I understand EcoTect can't do line and plane sources (yet!), but the final result doesn't need to be absolute in its measurements. Instead, each topology option would be considered relative to the other options. Could a series of point sources serve as a substitute?

If there is no workaround, do you know of an alternative package?

Thanks very much.

-Casey

[Archived phorum post from squ1.com v2 by craines]

Although this was not the original intention of the acoustic functions in ECOTECT, it doesn't mean that people haven't given it ago...

A user at Cambridge emailed us a while back about this. They had done a really interesting model of the space between two Mayan temples to look at the behaviour of sound in that space.
I would like to post an image of it here for you but will need to ask if she would mind me doing so first...

Anyway, I will attempt to describe it for the meantime. It was like the negative of the built form, with the surface of each opposing temple and with sides and top to define an "enclosure" with highly absorptive material which simulated sound fading into the distance kinda thing.

She seemed quite happy with the outcome she got from the model. It was a test model for her research to ensure that her method was appropriate... and as far as we are aware she has continued with the work.

Will see if she minds me posting a picture of it for you...

Regards,
Caroline

Caroline Raines
Square One research Pty Ltd

[Archived phorum post from squ1.com v2 by Andrew]

Hi Casey,

A very late response I know but...

Basically the problem in most urban areas and outdoor noise situations is diffracted sound, not reflected sound.

ECOTECT offers the statistical RT calculation (which assumes a totally homogeneous sound field - perfect mixing and infinite diffraction) and the ray-tracing method (fully specular reflection with no diffraction). This way both extremes are considered and the real answer will lie somewhere in the middle.

In an outdoor environment, if we take the example of the courtyard, the only way sound can enter the courtyard is by either reflecting off another much higher surface through the top of the central opening (unlikely) or by diffracting around the roof (highly likely). Calculating all possible diffraction paths is where the long calculation times come in.

You also get diffraction around objects within a room, however the sound level that diffracts around a balcony, for example, will be relatively low compared to sounds that bounce and reflect around the baclony and hit the same point. Internal ray-tracing can consider diffracton and diffusion (as a level of randomness) as the total internal envelope is known and assumed to be fully enclosed (as the zone has a finite volume) and, as no reflected sound can 'leak' out of the space, all its energy be distributed around the space at some time.

This is not the case outside as the majority of sound will simply disappear off into the atmosphere so any small inaccuracies in the algorithm that determinations which sound paths are important will be highly significiant.

I know that doesn't help, but I hope it explains a bit why I have refrained from adding outdoor noise calcs in ECOTECT.

Regards,
Andrew

Dr. Andrew Marsh
Square One research Pty Ltd

[Archived phorum post from squ1.com v2 by William Archer]

Where can I find studies on outdoor music venues..........shape, size, trees, grass, etc...all affect the sound.....I am looking for a perfect model of the best facility.......

[Archived phorum post from squ1.com v2 by craines]

For outdoor acoustics I think the Greeks and later the Romans have done it the best. There are many examples, but I think the smaller amphitheatres are the most successful. Some more than others of course but they eventually worked out the correct geometry and proportions, which is where most of the challenge is for achieving the best acoustics.

Regards,
Caroline

Caroline Raines
Square One research Pty Ltd

[Archived phorum post from squ1.com v2 by Henrique]

Hi dear friends,

I am a MSc Student at the Federal University of Uberl?ndia, in Brazil. I pretend work with numerical simulations in the scope of acoustic outdoors. My lecture request me to looking for some general statisticals methods, then I found the Direct Monte Carlo Simulation. I dont?t know how does it might works in the concerns of acoustic. If it's possible to do, please, send me some information about this.

Kindly,

Henrique G. Moura

[Archived phorum post from squ1.com v2 by craines]

I don't think we have any details on the "Direct Monte Carlo Simulation".
I would say your best bet is to do a search in google..

Not much help but better than nothing perhaps...
Regards,
Caroline

Caroline Raines
Square One research Pty Ltd