Response Comparison Across Target Levels — Seeking Insight

[mitchjrj]

I turned to REW's EQ tool for a more objective reference to an MSO target level. Based on an average response of the seats I will be feeding to MSO, REQ calculated a target level of 82.5dB (range 16–140Hz). I then ran a couple more equalizations at 85.0dB and 87.5dB to see how they compare. Created measured response curves for each.

What I find interesting is the response above 120Hz. At 82.5dB and 87.5dB target levels the response drops off. But at 85dB it's driven up. This is an improved result IMO, but I'm interested in why that is; if anything I would have figured the higher target level would have shown that result.

 

[John Mulcahy]

The behaviour depends on the shape of the uncorrected response relative to the target and the amount of boost allowed in the Filter tasks settings.

 

[mitchjrj]

The behaviour depends on the shape of the uncorrected response relative to the target and the amount of boost allowed in the Filter tasks settings.

...underscoring the value, I guess, of doing these types of comparisons?

BTW, should be noted that the only thing changed from iteration to iteration was the target level. Frequency range (16–140Hz), maximum boosts (+10dB individual/total), and flatness (±1dB) remained the same.

 

[John Mulcahy]

underscoring the value, I guess, of doing these types of comparisons?

No, underscoring that looking at the output without seeing the input is of little help in understanding how it came about. Generally helpful to attach the mdat file.