EQ filter gain and Q selection question

[dima1stg]

Hit an interesting situation while auto-generating filters. Please see attached "REW filter 1" - what took my attention is that there are two very close filters, 244hz and 250hz. The result of them are shown in attached "REW eq 1". I manually changed them to a single filter as shown in "REW filter 2", the result of which is shown in "REW eq 2". The change shows practically no difference between 1st and 2nd filter set. It shows that filter calculation is biased toward lower Q/lower gain per filter even if it means increasing number of filters. It's very interesting to hear the reasoning behind this. Many thanks!

 

[John Mulcahy]

As frequency increases so does positional dependence, the narrower a filter is the more likely it is to be wrong at a slightly different measurement position. The help explains the Q handling:

Allow narrow filters below 200 Hz determines whether target match uses filters narrow enough to counter modal resonances at low frequencies. This should be selected when applying EQ to a room measurement, but is best not selected when applying EQ to a device response (headphone EQ, for example). When this option is not selected the highest Q used will be 5.0.

If Allow narrow filters below 200 Hz is selected the Vary max Q above 200 Hz option tells REW to adjust the maximum Q from 10.0 at 200 Hz to 3.0 at or above 10 kHz. If this is not selected the max Q above 200 Hz will be 5.0.

 

[dima1stg]

a filter is the more likely it is to be wrong at a slightly different measurement position

So by spreading filter frequencies a bit makes the peak flatten which would avoid this, right?

The behaviuor of Vary max Q above 200 Hz is understood. But what does limiting Q above 200hz give?

I also noticed another kind of preferential treatment of Q: (speaking of lower frequencies only) REW tries to make filters with rather lower Q, MSO prefers higher Q. What would be advantages and disadvantages of both approaches?

 

[John Mulcahy]

But what does limiting Q above 200hz give?

As above, it avoids corrections that are too sensitive to position.

REW tries to make filters with rather lower Q

No, REW tries to match the target. It's an optimiser, it does the best it can within the constraints it is given. Other people use different optimisers and different constraints, I have no idea what they are.