Every good simulation technique must be proven using benchmarks. The PTB in Aachen, Germany put together a few such benchmarks for testing room acoustics simulation methods, with complete records of the performance of several of the participants. These benchmarks comprise an indicator of the performance of other programs which perform Geometrical Acoustics simulation, and give an overall impression of the current state of the art in room acoustics simulation. This article will deal with Pachyderm's application to one of these benchmarks, the third Round Robin.

For more information on this historic test, please see the PTB website at the following address:

PTB Round Robin III

By performing the benchmark using Pachyderm, we not only obtain valuable information which has led to its improvement, but we can also begin to understand the settings needed to obtain an accurate prediction for a given room.

The room used for the benchmark was the music studio at the PTB. This room was carefully modeled according to the instructions of the PTB. Absorption and Scattering coefficients were provided. Source and Receiver locations were specified.

 

The simulation was run with two different settings. The first was a pure ray-tracing method using 20000 rays, and a ray cut off time of 1.2 seconds (the thick red lines). The second was the same, but with 3 orders of Image-Source added (the thick orange line).  Both simulations used the air absorption calculation method from ISO 9613-1, without the edge frequency correction.

For purposes of brevity, the results of simulations from 1 source and one receiver are shown below. The rest are available in a spreadsheet for download (see the downloads section of this website):

Reverberation Time (T-30)

Source 1, Receiver 1, T-30

Source 1, Receiver 2, T-30 

Source 1, Receiver 3, T-30

 

Source 2, Receiver 1, T-30

Source 2, Receiver 2, T-30

Source 2, Receiver 3, T-30

 

Early Decay Time (EDT)

Source 1, Receiver 1, EDT

Source 1, Receiver 2, EDT

Source 1, Receiver 3, EDT

 

Source 2, Receiver 1, EDT

Source 2, Receiver 2, EDT

Source 2, Receiver 3, EDT

 

Definition (Deulichkeit/D50)

Source 1, Receiver 1 D

Source 1, Receiver 2, D

Source 1, Receiver 3, D

Source 2, Receiver 1, D

Source 2, Receiver 2, D

Source 2, Receiver 3, D

 

Clarity (C-80)

Source 1, Receiver 1, C

Source 1, Receiver 2, C

 

Source 2, Receiver 1 C

Source 2, Receiver 3, C

Center Time (Ts)

Source 1, Receiver 1, Ts

Source 1, Receiver 2, Ts

Source 1, Receiver 3, TS

Source 2, Receiver 1, Ts

Source 2, Receiver 2, Ts

Souce 2, Receiver 3, Ts

 

Strength (G)

Source 1, Receiver 1, G

Source 1, Receiver 2, G

Source 1, Receiver 3, G

Source 2, Receiver 1, G

Source 2, Receiver 2, G

Source 2, Receiver 3, G

We hope that the results show that the software performs on par with the industry standard as of the time of the PTB Round Robin shown here. It is true that simulation program results do not match the measurement exactly. At low frequencies, this is because geometrical acoustics simulations do not take into account phenomena such as room modes. At other locations in the frequency spectrum, variation is of approximately 1-2 JND (Just Noticeable Difference), which, given the level of uncertainty associated with room acoustics measurements, makes the results of these simulations plausible. A modicum of caution concerning the possibility of errors is prudent.

This page will be updated as future versions of the software, or algorithm design scenarios are benchmarked.