Hello all,
I use REW and rephase FIR filters for my room correction. I had exported from REW a L and R vector averaged measurments for 6 ea. positions arounf my normal sitting position. All time aligned of course. In rephase there is a significant difference in phase with the L measurment showing a -720 degree shift after 1000 hz to 20000 hz.
When I bring in the single centre position measurments from L and R, the phase matches fairly closely.
I infer that the averaging process is changing the phase, making it impossible to do phase alignment. And there doesnt seem to be a way to merge the phase from the single measurements into the gain measurement of the averaged responses. If even that makes sense. Has anyone run into this? Otherwise I would seem to have to go back to single measurents if I want to adjust phase.
Cheers,
T
I use REW and rephase FIR filters for my room correction. I had exported from REW a L and R vector averaged measurments for 6 ea. positions arounf my normal sitting position. All time aligned of course. In rephase there is a significant difference in phase with the L measurment showing a -720 degree shift after 1000 hz to 20000 hz.
When I bring in the single centre position measurments from L and R, the phase matches fairly closely.
I infer that the averaging process is changing the phase, making it impossible to do phase alignment. And there doesnt seem to be a way to merge the phase from the single measurements into the gain measurement of the averaged responses. If even that makes sense. Has anyone run into this? Otherwise I would seem to have to go back to single measurents if I want to adjust phase.
Cheers,
T
Hmm... l just use REW vector average function after I align the Impulse for each then run the eq filter function and export the l and r averaged measurement and the filters and import Into rephase. It shows large differences and worse phase compared to individual location l and rmearurements. I use linear phase filters from that point on in rephase.
It kind of makes sense to a neophyte like me that if you have multiple measurements around d the listening position thar they will each have different phase not unlike the different gains in a response measurements. So the question is whether the averaging process also averages the phase properly and is it valid to use averaged phase for crossover liberalization and such?
Sure, even if they were minimum phase responses, which they won't be, you'd expect phase to change with response.
You didn't answer my question about measuring at the listening position but that's OK. What if I asked you why you'd want to average them? ..the answer reconciles the same.
You didn't answer my question about measuring at the listening position but that's OK. What if I asked you why you'd want to average them? ..the answer reconciles the same.
