REW 5.31.1 Odd Problem with Filters with Negative Gain

[ric982]

Hi,
So I was generating EQ for a subwoofer and I've exported the fitler text and loaded on a MTRX Studio. I also created a filter measurement from the filters. I'm finding that the MTRX Studio's drawing of it's EQ with the loaded filter settings do not match the REW filter measurement shape or gains. I've started trying to figure out what's going on so I set up a simpler case in REW.

I've imported a flat curve. I've created two filters at 100 Hz with Q=10 with eith gain=+12 and the other with gain=-12. In my thinking they should have the same Q but opposite polarity and cancel each other. But drawn Q's are different. Am I confused or is this a problem?

 

[John Mulcahy]

Your EQ configuration uses classic Q, which does not give symmetric filters.

 

[ric982]

OK - Symmetric EQ fixed what I saw above. Loaded the new EQ on MTRX Studio and it measures closely with REW predicted result and sounding good.

I still see some descripancies between REW's Filter measurement and MTRX Studio's EQ sum display. Shapes are pretty similar, but some of MTRX Sudio peaks off by maybe up ot 3 dB including the fact the EQ went above 0 dB which I told REW not to do.

Checked REW's filter measurement against MTRX Studio EQ display. Wave Q10 EQ display, iZotope Neutron Equailzer EQ DIsplay. These are all different from each oter. I will take that as REW is accurate and other's GUI'snot so much

Thanks for the help.

 

[John Mulcahy]

Different Q definitions give different filter shapes for the same value.

 

[ric982]

Ok. Sorry I didn't check the alternative Q type.
MTRX Studio EQ is using RBJ Q filters.
REW and MTRX Studio predicted filter curves look the same when this is selected in REW.