What distance is the standard PIR for?
Would it be possible to calculate PIR with distance as a parameter. Increasing distance will lead to increased slope.
I did some googling, but couldn't really find anything like what I was looking for. I believe Harman's targets are based on 3m or 3.5m distance?
Off the top of my head I don't remember if PIR in its current form is described anywhere as
necessarily corresponding to a specific distance.
However, the Early Reflections curve comes from Devantier 2002, which surveyed 15 listening rooms. The average listening distance was 3.3 meters or 10' 10". That paper also includes a crack at a predicted in-room response, but does not specify how it was calculated, if it used the 12%LW/44%ER/44%SP we use nowadays.
In the Olive preference paper part 2, 13 PIRs are compared to their actual in-room response averages. These in-room responses each consisted of an average of 9 measurements taken at 3M, representing the listening window angles (0; ±10 vertical; ±10, 20, 30 horizontal). In my experience performing a spatial average of this sort does produce results closest to the PIR. The speaker was 1.2m away from its rear wall and 'slightly-off center' from the sidewalls.
So usually when I think of the PIR, I think of it as a roughly 3M measurement, though 2M generally corresponds well too. Anything below that and the curve starts to look more and more like the on-axis. But as noted by
@thewas, it's always going to have some variation.
I'd be curious to see if people reproduced Olive's methodology how close speakers would line up in a variety of different kinds of rooms.
