Strange Impulse response after DIRAC...

[k900733]

Hello!

I measured room response of my 2.1 system with and without DIRAC DRC. The SPL curve looks okay, but the impulse response curve does not look right to me.

The DIRAC is the most updated version running on minidsp SHD. I did not roll off the mains. The sub is xo at 50 Hz using minidsp's software. Since SHD is not capable of bass management, I just run DIRAC measurements on the entire system (i.e. DIRAC takes measurements of L+sub, R+sub at each mic position). I know I am not supposed to do so, but I did the same when I was using the DDRC-22d and got quite good results without seeing strange impulse. Maybe this is not the case anymore.

Any thoughts on this? Or it is actually not an issue? Thanks a lot!


SPL-Left

SPL-R

Impulse - DIRAC off

Impulse - DIRAC on

Impulse of Left channel are pretty much the same, so skipped here.

 

[SIY]

Your frequency response with it on looks a whole lot smoother. Looks like it's doing just what it says on the tin. Bet it sounds great.

 

[ppataki]

In REW invert the polarity of your Dirac measurement (you can find that on the IR tab in the options)
Also I would rather check the Step response curve, I am sure the difference will be huge there in favor of Dirac

 

[k900733]

Thanks for the responses!

The invert response still doesn't look "typical". The Step response is slightly better with Dirac on.

Invert impulse

Step response - Dirac off

Step response - Dirac on

 

[Lorenzo74]

Step response - Dirac on
View attachment 133796

just repeat the measurement with mic lowered a bit... few centimetres (in step of one inch). it seems you took DIRAC calibrations measurements with the initial one, the listening position, slightly lower compared with the one you are using now for this comparison. my suggestion, to have a clear DIRAC-NO DIRAC comparison is to do the center measurement at the end and do not "touch the mic". the you can do measurement in REW enabling and disabling Dirac and you will see the clear difference in step response. pls. let me know if enough clear for you. best

 

[k900733]

Thanks a lot @Lorenzo74 ! That's very clear

I did experiment a bit on Saturday, and confirmed that it's not due to discrepancy between mic position when taking DIRAC LP MEASUREMENT and REW MEASUREMENT. What I did was, I took only one DIRAC MEASUREMENT, at LP, and generated filter to export to SHD. Then I took REW MEASUREMENT. By doing so with mic position untouched, I still see the 'non-typical' impulse when DIRAC is on. I also tried using mains only without sub -- still the same. I did not see this before, so I uninstalled DIRAC v3.0.14, and tried v2 and v3.0.0, but it seems current version SHD doesn't support those older versions and it was not able to connect to DIRAC services.

Anyway, I plan to post on miniDSP forum as well, or even issue a ticket, to see if my confusion can be addressed. (BTW my room is 24' x 17' very rectangular room; shouldn't be too complicated for DIRAC.)

 

[abdo123]

Thanks a lot @Lorenzo74 ! That's very clear

I did experiment a bit on Saturday, and confirmed that it's not due to discrepancy between mic position when taking DIRAC LP MEASUREMENT and REW MEASUREMENT. What I did was, I took only one DIRAC MEASUREMENT, at LP, and generated filter to export to SHD. Then I took REW MEASUREMENT. By doing so with mic position untouched, I still see the 'non-typical' impulse when DIRAC is on. I also tried using mains only without sub -- still the same. I did not see this before, so I uninstalled DIRAC v3.0.14, and tried v2 and v3.0.0, but it seems current version SHD doesn't support those older versions and it was not able to connect to DIRAC services.

Anyway, I plan to post on miniDSP forum as well, or even issue a ticket, to see if my confusion can be addressed. (BTW my room is 24' x 17' very rectangular room; shouldn't be too complicated for DIRAC.)

you should contact Dirac’s support not someone else.

 

[hmt]

Did you time align your Sub an Mains before dirac?

 

[bennybbbx]

Have you find a solution ?. what speakers you have ?. In your FR can see a large gap and on step response see that at 2 ms come another fast peak. seem your woofer is 2.2 ms too late. look as woofer is. It is important for good sound that the delay between woofer and other is at least 1 ms good. no measure system can change crossover phase problems see my measure examples with diffrent delays. when the delay is correct there come no gap

https://www.audiosciencereview.com/...ubwoofer-comparison.20494/page-12#post-814688

but important is that it sound good. you can try drum sound examples and hear the reverb and that the base drum did not sound as it is more away. try with and without dirac on which setting you think the example sound more realistic and reverb good

here are audio drum examples to play. they are usefull for tests. scroll a little and click on audio demos

https://www.toontrack.com/product/ezdrummer-2/

 

[usersky]

It's the same reality viewed from two points of view: frequency / impulse response. They are equally strage/normal. Just the point of view differs.