Site Choice
Our site selection intentionally deals with different volumetric constraints in order to see how neural network will perform in the creation of acoustic forms in unusual or difficult spaces.
New York City
Chicago
211 E 26th St
New York, NY
320 E North Water St
Chicago, IL
Site Dimension
Small Site
Large Site
Site Dimension:
100 ft (L) x 49.5 ft (W) x 47.5 (H)
30.5 m (L) x 15 m (W) x 14.5 m (H)
Volume:
233,665 cubic foot
6,616 cubic meter
Site Dimension:
328 ft (L) x 147.6 ft (W) x 98.4 (H)
68.6 m (L) x 45 m (W) x 30 m (H)
Volume:
3,272,080 cubic foot
92,655 cubic meter
Input for Graph Convolutional Neural Network
You will need the following three inputs to generate the outcome concert hall. Here is how you do it.
+
+
Result Matrix
44 selected result models from the Graph CNN. 22 of which are generated from the small site, and the other half from the large site. We vary the input mesh shape, mesh counts, acoustic parameters, etc.
Small Site
Large Site
Simulation Matrix
44 acoustic simulation results from Ecotect. The performance of sound waves is visualized through simulated particles. A range of results can be seen demonstrating how sound travels within our forms.
Parameter Diagram
14 influencing factor for our Graph CNN. This diagram shows different influences and experiment results.
Section Style Transfer
Section drawings generated through another neural network: Style Transfer. Using existing concert hall section drawings to project an additional layer of information onto sections drawn from 4 selected models generated by our Graph CNN.
Acoustic Simulation
Selected examples of acoustic simulation.
S - 2
Input Values - Seats: 719 Volume: 6615 RT: 1.91 Sa: 620 H: 24 W: 15 L/W: 2.03 H/W: 1.60 V/N: 9.2 a: 0.9
Output Values - Seats: 167 Volume: 1212 RT: 1.97 Sa: 110 H: 9 W: 12.6 L/W: 2.13 H/W: 0.71 V/N: 7.25 a: 0.9
S - 12
Input Values - Seats: 416 Volume: 3740 RT: 1.9 Sa: 352 H: 6.2 W: 10.4 L/W: 0.5 H/W: 0.6 V/N: 9 a: 0.9
Output Values - Seats: 171 Volume: 1542 RT: 1.56 Sa: 177 H: 15.6 W: 13.3 L/W: 2.21 H/W: 1.17 V/N: 9 a: 0.9
L - 2
Input Values - Seats: 2667 Volume: 24000 RT: 2.08 Sa: 2066 H: 30 W: 44 L/W: 1.48 H/W: 0.68 V/N: 9 a: 0.9
Output Values - Seats: 2433 Volume: 19351 RT: 2.16 Sa: 1600 H: 41.9 W: 43.5 L/W: 1.51 H/W: 0.96 V/N: 7.95 a: 0.9
L - 21
Input Values - Seats: 650 Volume: 5500 RT: 1.97 Sa: 500 H: 24 W: 13 L/W: 2.23 H/W: 1.85 V/N: 8.46 a: 0.9
Output Values - Seats: 247 Volume: 1940 RT: 1.93 Sa: 180 H: 19.3 W: 19 L/W: 1.58 H/W: 1.02 V/N: 7.09 a: 0.9
Interior Rendering
A sense of the interior space within mesh models generated by the Graph CNN.
