A question about RTA and reflections

[moedra]

I am very conscious of what the RTA is. My question has to do not with the standard use of tracking SPL response, but with the idea of tracking reflections (i.e. the ETC graph) in real time to make hunting for room treatment points on walls faster than the standard measure-move a panel-measure-again ritual that consumes a lot of time. It would be nice if there were some way to let an RTA run while walking around the room with a panel to observe impulse reflection changes on the fly in the ETC graph. Is this possible, and if not, is it a thing that can be done?

 

[dcibel]

ETC, or an impulse response for that matter cannot be determined with constant stimulus signal measured via RTA. So, you can use RTA for many aspects of real-time response analysis, for ETC, RT60 decay, etc. you need to run a sweep measurement.

 

[moedra]

ETC, or an impulse response for that matter cannot be determined with constant stimulus signal measured via RTA. So, you can use RTA for many aspects of real-time response analysis, for ETC, RT60 decay, etc. you need to run a sweep measurement.

I figured, but what if those sweeps were relatively short - like 256k - and repeating? REW's generator can already do that. It can also be told at what frequency to start and stop the sweep for the sake of what is being done. All that's needed is a method to read these sweeps in a real-time refreshing ETC window. I get that it would still be refreshing iterations and not actually "real-time", but it would save a lot of extra shuffling around and clicking and measuring and deleting if the process happened automatically. I imagine it could also be sped up if the only data processed were the ETC and Step responses. Then all you'd have to do is move slowly around in the room checking the iterative reflection spikes or channel coherence to see where the panels need to be placed. It would be incredibly fast and efficient when hunting down reflections for room treatment purposes.

 

[moedra]

I'm working with a lot of people who are currently playing reflection detective and something like this would save so much time.

 

[Keith_W]

You could do this:

15 sec should be enough time for you to run to a new location and set up your tripod.

 

[moedra]

You could do this:

View attachment 89453

15 sec should be enough time for you to run to a new location and set up your tripod.

I thought about doing this. Haven't tried it yet. I was thinking that it would be difficult, upon examining the recorded measurements, to identify which one went with which position. I suppose I'd have to keep track of it on a diagram. Each measurement would have to default to the ETC graph for me to be able to see it in "real time" before the next one started. Hmm.