miniDSP Dirac Live causing imaging problems

[bmwr75]

Looking for input. I installed Magnariser stands on my Magnepan LRS speakers many months ago. Today moved them from the shop back into the main listening system. Am using a miniDSP DDRC-24 running Dirac Live to blend the two SVS SB-2000 subs with the LRS main speakers.

The problem is the center image was not good and Left/Right imaging was off. I use the Roger Waters - Amused to Death - The Ballad of Bill Hubbard to test if speakers are set up correctly. If they are, the dogs barking at the start of the song will be to the far right (nearly behind you) and the radio man talking will be to the far left (nearly behind you). Today they were both between the speakers. Wrong!!!!

I turned Dirac Live off in the miniDSP screen on my laptop and the imaging problems all went away. Had to readjust subs volume levels since no EQ was applied via Dirac Live. Has anyone else run into this issue using miniDSP/Dirac Live??

 

[MarkS]

Does Dirac Live play well with dipoles? I would think not, but I have no experience ...

 

[ppataki]

Looking for input. I installed Magnariser stands on my Magnepan LRS speakers many months ago. Today moved them from the shop back into the main listening system. Am using a miniDSP DDRC-24 running Dirac Live to blend the two SVS SB-2000 subs with the LRS main speakers.

The problem is the center image was not good and Left/Right imaging was off. I use the Roger Waters - Amused to Death - The Ballad of Bill Hubbard to test if speakers are set up correctly. If they are, the dogs barking at the start of the song will be to the far right (nearly behind you) and the radio man talking will be to the far left (nearly behind you). Today they were both between the speakers. Wrong!!!!

I turned Dirac Live off in the miniDSP screen on my laptop and the imaging problems all went away. Had to readjust subs volume levels since no EQ was applied via Dirac Live. Has anyone else run into this issue using miniDSP/Dirac Live??

What about your phantom center? So the lead vocal for example - is that coming from the center or is that also shifted to either side?
I am asking because that is what I have been experiencing with Dirac lately and actually it might be the same issue that you are experiencing too

The 'solution' I found is to take multiple measurements, export them to the filter banks and then listen to them to find the one where the soundstage is fine
This is time consuming but could not figure out a better solution

 

[Bugal1998]

Looking for input. I installed Magnariser stands on my Magnepan LRS speakers many months ago. Today moved them from the shop back into the main listening system. Am using a miniDSP DDRC-24 running Dirac Live to blend the two SVS SB-2000 subs with the LRS main speakers.

The problem is the center image was not good and Left/Right imaging was off. I use the Roger Waters - Amused to Death - The Ballad of Bill Hubbard to test if speakers are set up correctly. If they are, the dogs barking at the start of the song will be to the far right (nearly behind you) and the radio man talking will be to the far left (nearly behind you). Today they were both between the speakers. Wrong!!!!

I turned Dirac Live off in the miniDSP screen on my laptop and the imaging problems all went away. Had to readjust subs volume levels since no EQ was applied via Dirac Live. Has anyone else run into this issue using miniDSP/Dirac Live??

Yes. Multiple times. This is a known issue.

Make sure your mic isn't too close to the wall right behind the MLP; if the problem persists, contact Dirac tech support.

 

[bmwr75]

What about your phantom center? So the lead vocal for example - is that coming from the center or is that also shifted to either side?
I am asking because that is what I have been experiencing with Dirac lately and actually it might be the same issue that you are experiencing too

The 'solution' I found is to take multiple measurements, export them to the filter banks and then listen to them to find the one where the soundstage is fine
This is time consuming but could not figure out a better solution

There is no phantom center. Vocals sound like they are coming directly out of each speaker.

 

[ppataki]

There is no phantom center. Vocals sound like they are coming directly out of each speaker.

There is something fundamentally wrong then
If you switch off Dirac, will the phantom center get into the right position?

 

[PeterNL]

Looking for input. I installed Magnariser stands on my Magnepan LRS speakers many months ago. Today moved them from the shop back into the main listening system. Am using a miniDSP DDRC-24 running Dirac Live to blend the two SVS SB-2000 subs with the LRS main speakers.

The problem is the center image was not good and Left/Right imaging was off. I use the Roger Waters - Amused to Death - The Ballad of Bill Hubbard to test if speakers are set up correctly. If they are, the dogs barking at the start of the song will be to the far right (nearly behind you) and the radio man talking will be to the far left (nearly behind you). Today they were both between the speakers. Wrong!!!!

I turned Dirac Live off in the miniDSP screen on my laptop and the imaging problems all went away. Had to readjust subs volume levels since no EQ was applied via Dirac Live. Has anyone else run into this issue using miniDSP/Dirac Live??

Take a look on Dirac Impulse left and right channel, sometimes Dirac can recognize it wrong (much different delays between channels).
Like in this case:

https://www.reddit.com/r/audiophile/comments/u2b0iq

 

[ErVikingo]

Looking for input. I installed Magnariser stands on my Magnepan LRS speakers many months ago. Today moved them from the shop back into the main listening system. Am using a miniDSP DDRC-24 running Dirac Live to blend the two SVS SB-2000 subs with the LRS main speakers.

The problem is the center image was not good and Left/Right imaging was off. I use the Roger Waters - Amused to Death - The Ballad of Bill Hubbard to test if speakers are set up correctly. If they are, the dogs barking at the start of the song will be to the far right (nearly behind you) and the radio man talking will be to the far left (nearly behind you). Today they were both between the speakers. Wrong!!!!

I turned Dirac Live off in the miniDSP screen on my laptop and the imaging problems all went away. Had to readjust subs volume levels since no EQ was applied via Dirac Live. Has anyone else run into this issue using miniDSP/Dirac Live??

You lost me at R75 (I learned to ride on dad's R 75/5 toaster curry!) and at Amused to Death. One of my go to is Amused and also Pros and Cons 5:06 for imaging and bass accuracy. I tried MiniDSP with DIRAC on my IRS Beta and ended up removing it. Perhaps I didn't get a good measurement (had a lot of issues about low signal to noise) but it was unlistenable.

 

[dan_uk]

No idea about Dirac Live, but I do love Amused to Death, epic choice!

 

[PeterNL]

What about your phantom center? So the lead vocal for example - is that coming from the center or is that also shifted to either side?
I am asking because that is what I have been experiencing with Dirac lately and actually it might be the same issue that you are experiencing too

The 'solution' I found is to take multiple measurements, export them to the filter banks and then listen to them to find the one where the soundstage is fine
This is time consuming but could not figure out a better solution

Your solution is good, I just would add to it: double check impulse response every single measurement for anomaly etc.

 

[anotherhobby]

I don't like how Dirac (or Audyssey for that matter) sounds much above the room's transition frequency. Although my graphs look better with full range correction, they don't sound better. I determined my rooms transition frequency manually by looking at the fully corrected Dirac response and compared it to the uncorrected response. I tried a few several different frequencies from 250Hz to 1Khz and landed on 550 Hz in my room giving me the best listening experience.

The graph below only has 40 dB of range to exaggerate the variations. Speakers are near field Revel M105s paired with 4 sealed 8" subs that I built. The Dirac line is flatter and smoother past 550 Hz, but it does not sound better. I won't guarantee results in your situation, but you should try lowering your correction range and see how it sounds.

 

[Nutul]

Looks like "Amused to Death" is the choice of more than one. "The ballad" is my choice too when it comes to spatiality (well, more wideness...) of the soundstage.
I also pick "Ruiner" from "The Downward Spiral" by Nine Inch Nails, for depth perception, as it has a "woooosh" coming in from up-left, behind your ear, to slightly right-front at belly level; transition takes place in about 1 second, but is astonishing when you experience it (without any surround, ehm, obviously)... if I don't "see" these details being there, even not so evident, I begin immediately thinking about speaker positioning. How cool is that?

 

[RickyC34]

I've always chosen the 13-point measurement "focused imaging" when running calibrations. Give the 9-point a try. (even try only measuring 6 points) For some reason in my system choosing the 9-point measurement and keeping each point about 12 inches apart provided almost a night-a-day difference in imaging. Obviously, this heavily focuses on the MLP but in my case, my wife and others could care less or wouldn't notice. Make sure the mic is at least 8 inches away from the couch's headrest at all times.

 

[RickyC34]

If I recall (haven't had this version of Dirac in a long time) don't you need to set the delay and gain of each speaker on the minidsp before running Dirac? If you're not doing this I would assume it would help your imaging problem.

 

[vicenzo_del_paris]

I am running Dirac full band.
I just tried "The Ballad of Bill Hubbard" with Dirac on/off. No difference for barking dogs and radio man, in both case they are located behind me.
An also, on or off, I don't have changes for the phantom center.

I suggest you do your calibration again.

it happened few times in the past to loose the phantom center after a given calibration (as I often try different speakers locations, I do calibration each time).
When it happened, I was re-running the calibration and things were ok.

 

[PeterNL]

I am running Dirac full band.
I just tried "The Ballad of Bill Hubbard" with Dirac on/off. No difference for barking dogs and radio man, in both case they are located behind me.
An also, on or off, I don't have changes for the phantom center.

I suggest you do your calibration again.

it happened few times in the past to loose the phantom center after a given calibration (as I often try different speakers locations, I do calibration each time).
When it happened, I was re-running the calibration and things were ok.

Same here. I'm checking impulse response every single measurement.

 

[bmwr75]

Additional info

I have a miniDSP SHD in my shop and used it with Dirac Live 3 on these same Magnepan LRS+ speakers. Looked at my notes yesterday. Sure enough, turning Dirac on after the calibration had been performed/saved/slot loaded caused the same imaging issues as am now experiencing in my main listening system.

 

[bmwr75]

I sent miniDSP a message today explaining the situation and asked if they knew why Dirac Live 3 is causing issues with these Magnepan speakers.

 

[RickyC34]

Odd indeed.
What mic are you using? Are you using the 90-degree cal file loaded into Dirac pre-measurement?

May want to double-check your routing screen. This sounds like you may have L/R playing the same signal at the same time. An example would be having Left Sp with Dirac 1 and 2 both selected. If you have a 2.2 it should look like the below except for the "unused" will be sub 2 with Dirac 1 and 2 selected.

You can play Left / Right speaker test tracks on Tidal or Spotify to check this as well.

Have you confirmed if you need to set delay/gain of your speakers/subs pre-Dirac measurement? I'm pretty certain you need to do this at least for mains/subs blending before running Dirac.

Some of this is 101 and by all means, no disrespect. Just trying to help get you up and running.

 

[bmwr75]

Odd indeed.
What mic are you using? Are you using the 90-degree cal file loaded into Dirac pre-measurement?

May want to double-check your routing screen. This sounds like you may have L/R playing the same signal at the same time. An example would be having Left Sp with Dirac 1 and 2 both selected. If you have a 2.2 it should look like the below except for the "unused" will be sub 2 with Dirac 1 and 2 selected.
View attachment 329088

You can play Left / Right speaker test tracks on Tidal or Spotify to check this as well.

Have you confirmed if you need to set delay/gain of your speakers/subs pre-Dirac measurement? I'm pretty certain you need to do this at least for mains/subs blending before running Dirac.

Some of this is 101 and by all means, no disrespect. Just trying to help get you up and running.

I've used the measurement software many times on different sets of speakers. The Magenpans are the only speakers that have ever had any problems with imaging when Dirac is turned on. The speaker routing is set up properly. Checked this with a BINK's test signal. Am using the UMIK miniDSP sells. Can't remember which Cal file I was using, sorry, but doubt it was the 90 degree file since the mic is pointed towards the speakers.