Now it works!
I just took a testfile I had to try.
To bad that REW dont accept PCM as import so I could compare.
Maybe later I try real measurement before-after.
The only thing, is to measure with filters activated to see how your system behaves. You may have to adapt target curve for your system. qa2 is for my small qa2010. HK5 is for qa3050i.
As you can see in tutorial, very good results are easy and quick to get.
I did a measurment with the filter DRC-Filter Creation Experimentel and my own.
Im no expert just did an dirty test.
I tried to gain all measurements to the rise at 40hz.
Measurement is done center speakers about 250cm with Umik-1 with about 70db and REW.
1.Khadas Tone Dac without Brutefir after reboot.
2.One of mine EQ with REW.
3.Another of mine with REW and rePhase.
4.Preset of DRC-Filter Creation Experimentel qa2.
It do a nice job the DRC-Filter Creation Experimentel to get a straight curve.
Problem its killing the music!
Sounds flat and lifeless.
Right now I prefer Mine 2.
So hunting for a straight line is no meaning I think.
Get rid of the peaks thats big is the obvious then its up to your ears.
My point af view.
@Danrap : I have done nothing about samplerate… So I don’t understand why it was working before…as Solveit use the same board as far as I understood.
edit : I’m going to try to add a hardware detection for samplerate.
It is relatively hard to create filters mannualy with rePhase without some practice so I suggest you stick with DRC tool. Me2 curve has decent linearity and is a little heavier on bass and that is why you prefer it over blue curve (qa2).
Blue curve looks really good but it is sounding “dull” as it is ruler flat (straight line). I suggest you try to re-create filters with a different target curve which will be tilted downwards. Usually a 10 db tilt down over the entire range gives best results so you may try the target curve from this link (we.tl/t-5fedFTSJJj). Add this file in the folder “target-curves” in “brutefirfilters”, select it as a target curve and create new filters. Good luck!
Here is an example of the response most folks preferred in blind tests. I believe you can get similar results with adjusting slightly the target curve.
I would add than when we use to listen a system, we “learn it” and when a big change occurs, as with DRC, we may find it not good, or worst than before. It can be the case, of course, but sometimes it takes hours of listenning to appreciate the new “good” result. For the same reason, lot of listener add bass and treeble with a equalizer, not to get the more exact sound, but to get something more “pleasant”, but far from HIFI…