thomas |
Member |
|
|
Joined: 16 Mar 2011 |
Posts: 73 |
|
|
|
|
|
|
| Hi,
Quote: | the Schroeder Curve is taking wrong part of the echogram |
I don't see a mistake. For T10 the decay (-5..-15) dB and for T20 (-5..-25) dB has to be used per definition of T10 and T20.
Quote: | I seem not to be able to get a valid decay curve for the RT to be calculated on the other side of the room. |
I suppose you think of a "valid decay curve" as a smooth, even decay. But that's not what you can expect in such a room shape.
When I look at the room shape, I think,
[receiver "sees" the cource]
- there will be an early part of the RIR with a smooth decay (faces between source and receiver, reflections arriving staggered in time after the direct sound)
- two "soundclouds" travelling through the room. one clockwise and one anticlockwise - that leads to a rise and fall in the decay according to the "traveltime" of the two "soundclouds"
- both you can see in your first RIR
[receiver dosn't see the source]
- there will be no direct sound of course and a smooth attack (reflections arriving staggered in time, like in an opera house with the source in the orchestra pit and the receiver on the floor) - there will be an early part of the RIR with a smooth attack and a smooth decay
- the two soundclouds doing the same as above
- all that you can see in the second RIR (no not perfect, there is only one rise and fall - perhaps the two "soundclouds" have the same traveltime, or the second is not simulated (short simulation length))
So I would expect, this is a more or less "valid decay curve" as it describes propably the real decay of the room.
-------
edit (21.10.): Since the room isn't a round donut, a part of the soundclouds will be reflected backwards in each corner. That will yield to a smearing of the soundclouds in time and to a more or less even decay after some "rounds". But keeping in mind that one "round" takes about 500 ms (room size!), it will take (1..2) seconds or longer to the state of a smooth even decay.
-------
Quote: | Is it realistic to put a sphere speaker on one side of the room and map the RT on the whole room? |
Don't know what You mean with "realistic". Using a sphere is the right way for meausurements of T10, T20, T30.
In my opinion in this room occurs the question "how appropriate are T10,T20 to describe the properties of the soundfield in this unusual room?".
Quote: | and manually adjusting the curve myself? |
Yeah, that would be fine! But for a start try the EDT (decay range (0..-10) dB) and look for echoes (the travelling soundclouds!). |
|