Cannot get Dirac Live to work. Cannot understand why

[gilency]

I have a MAC Pro with a MiniDSP Flex balanced, a UMik1, 2 SW Rhythmic 12 (using line in and Eq and phase off), and ELX speakers with crossover 80-120 and delay 2 on RSw. this crossover is the one with the better graph.
I made sure I had the right 90 degree UMIK-1 calibration file, adjusted the speaker loudness and mic per DIRAC instructions and every time I run DL the REW graph is much worse than before DL.
I have repeated measurements with different number of measurements and mic positions and every time the DL graph looks atrocious. I am at a loss at what to do. Can anybody provide some insight on what I am doing wrong?
The top graph is PreDirac, the 2nd and one is after DL.
I have tried equalizing with REW prior to DL, it makes no difference.

 

[Hayabusa]

With what SPL level (at the listening position) where you measuring?

 

[gilency]

on DL the speakers were at 72 dB

 

[Mort]

What's your room look like?

That top graph does not look right either.

 

[gilency]

Is a rectangle about 23x23 ft about 8 ft tall with an 5 ft opening to the hall, kitchen, stairs. Speakers are equidistant 8 ft apart and I am at a triangle. Sw1 is close to the entry and Sw 2 is at the opposite corner

 

[Mort]

How do the two sound?

 

[gilency]

They sound great. I think they sound worse with DL on though

 

[Mort]

Do you have a graph of the ELX speakers alone, no crossover, subs off. I want to see the bare room.

What correction range did you use on Diraclive?

 

[Hayabusa]

on DL the speakers were at 72 dB

Do you also have an other way to verify your SPL form the U-MIK is correct? I had once dropped it and some internal dip switches made the level incorrect.

 

[Hayabusa]

Do you also have an other way to verify your SPL form the U-MIK is correct? I had once dropped it and some internal dip switches made the level incorrect.

Also what is you background noise level?

 

[gilency]

I used

Do you have a graph of the ELX speakers alone, no crossover, subs off. I want to see the bare room.

What correction range did you use on Diraclive?

 

[gilency]

74 with the iPhone
76 with the umik1
Background is -70 on the Rew input graph

 

[GXAlan]

Try a single point measurement with Dirac and force it to create the correction based upon just a single measurement (ignore the warnings). Then run REW without moving the mic.

 

[gilency]

Only 1 measurement at the main listening ear level. looks terrible too
is there a way to check if the umik1 file or the mic are deffective?

 

[Hayabusa]

74 with the iPhone
76 with the umik1
Background is -70 on the Rew input graph
View attachment 446558

background -70? I mean SPL background

 

[Keith_W]

With Dirac, the correction is only as good as the measurement. So, how are you taking your measurements? Are you using the mini tripod that came with your UMIK-1 (this is the most common mistake)? Hand-holding your mic? Did you read this thread?

And yes, do what @GXAlan suggested - one mic position, do the correction, then verify with REW. I would add: use REW to measure before and after Dirac correction, then upload the .MDAT to ASR.

 

[gilency]

SPL background is 36
I am unable to upload .MDAT files for some reason. MDAT files are greyed out

 

[Old_School_Brad]

As mentioned, the only valid way to compare Dirac and REW measurements is by using a single mic position and a single measurement -without moving the microphone at all between measurements.

Hopefully, this will help pinpoint the issue you're experiencing.

 

[Hayabusa]

SPL background is 36
I am unable to upload .MDAT files for some reason. MDAT files are greyed out

That's very good..

 

[vicenzo_del_paris]

SPL background is 36
I am unable to upload .MDAT files for some reason. MDAT files are greyed out

Zip thè files and upload thè zip archive