|
|
Post by rajdude on Jul 14, 2023 11:47:34 GMT -5
Hello everyone, Has anyone got good results with Dirac on a XMC-1 feeding Magnepan speakers, specifically the 3.6r?
I have that combination and every time I try Dirac correction, my speakers sound weird. What exactly happens is this: the large soundstage (which is typical of Magnepan speakers) collapses into a small space, mostly towards the center. It sounds like Dirac is making my Maggies sound like box type speakers. I wonder if Dirac is designed for box type speakers and has trouble with measuring and correcting dipole speakers. Bipoles have so much sound bouncing off the back walls and side walls that (maybe) Dirac gets confused?
Strictly technically, the DSP correction Dirac generates is not bad at all. I used REW to verify it. It just sounds weird.
In the past, I heard that for Maggies to sound good with Dirac, we need to tell Dirac NOT to correct over approximately 1kHz. That is what I did for my third try. It still sounds crappy. I could lower that to maybe just 500 Hz and try again.
Any tips from experienced people in the same boat here?
Thanks for your time -Raj
|
|
KeithL
Administrator 
Posts: 9,961
|
Post by KeithL on Jul 14, 2023 11:55:35 GMT -5
You've got the right idea... If you start from the assumption that the source is correct... Then "box speakers" can potentially deliver an accurate reproduction of that waveform... And Dirac Live will then do its best to measure and perfect the version of this waveform that arrives at the listening position. In contrast, with dipoles, you are getting a sort of mishmash of direct and reflected waveforms... And, while this may sound very nice, or may even sound "more like live natural sound" to you, it is quite likely to confuse Dirac Live to at least some degree. So, yes, Dirac Live IS "trying to make your Magneplanars sound like what you would hear from an accurate box speaker"... (And, while some people have achieved favorable results with Dirac Live and Maggies, you may or may not be able to in your particular system.) Hello everyone, Has anyone got good results with Dirac on a XMC-1 feeding Magnepan speakers, specifically the 3.6r? I have that combination and every time I try Dirac correction, my speakers sound weird. What exactly happens is this: the large soundstage (which is typical of Magnepan speakers) collapses into a small space, mostly towards the center. It sounds like Dirac is making my Maggies sound like box type speakers. I wonder if Dirac is designed for box type speakers and has trouble with measuring and correcting dipole speakers. Bipoles have so much sound bouncing off the back walls and side walls that (maybe) Dirac gets confused?
Strictly technically, the DSP correction Dirac generates is not bad at all. I used REW to verify it. It just sounds weird.
In the past, I heard that for Maggies to sound good with Dirac, we need to tell Dirac NOT to correct over approximately 1kHz. That is what I did for my third try. It still sounds crappy. I could lower that to maybe just 500 Hz and try again. Any tips from experienced people in the same boat here? Thanks for your time -Raj
|
|
|
|
Post by marcl on Jul 14, 2023 11:58:30 GMT -5
Hello everyone, Has anyone got good results with Dirac on a XMC-1 feeding Magnepan speakers, specifically the 3.6r?
I have that combination and every time I try Dirac correction, my speakers sound weird. What exactly happens is this: the large soundstage (which is typical of Magnepan speakers) collapses into a small space, mostly towards the center. It sounds like Dirac is making my Maggies sound like box type speakers. I wonder if Dirac is designed for box type speakers and has trouble with measuring and correcting dipole speakers. Bipoles have so much sound bouncing off the back walls and side walls that (maybe) Dirac gets confused?
Strictly technically, the DSP correction Dirac generates is not bad at all. I used REW to verify it. It just sounds weird.
In the past, I heard that for Maggies to sound good with Dirac, we need to tell Dirac NOT to correct over approximately 1kHz. That is what I did for my third try. It still sounds crappy. I could lower that to maybe just 500 Hz and try again.
Any tips from experienced people in the same boat here?
Thanks for your time -Raj
I've been using Dirac with an XMC-1 and now XMC-2 for about five years. My fronts are 3.7, and I also use two DWM panels, a CCR, LRS for surrounds and MMGW for front tops. So as you might expect ... I'm very happy with Dirac. I correct full range on all the speakers with a flat target curve. But ... there are considerations and most have to do with the room. In my room I have a lot of bass traps and a lot of diffusion, but I have tracked down some reflections that significantly affected imaging and soundstage. Dirac's impulse response correction can get confused if the out of phase rear dipole wave gets to the mic with sufficient energy ... Dirac may align to the reflection and not the main signal, or it may even flip the polarity of the speakers. So generally speaking there is no reason why Dirac will not work with Maggies - and my experience is that it works really, really well - but there are conditions that could cause the issues that you observe. So if you can give a little information on the nature of your room I might be able to help. |
|
|
|
Post by rajdude on Jul 14, 2023 12:49:04 GMT -5
Thanks for chiming in here Keith! What you said makes perfect sense. It also means I am barking up the wrong tree - actually VERY WRONG tree :-) Ha! ha!
ok, so rather than using Dirac to auto-correct......which I am assuming introducing lots of filters and time-delays etc to "fix" my "preferred" crappy bouncy sounding maggies (lol), what if I did this: Use REW to generate simple filters to correct subwoofer, lows and low-mids ONLY...maybe only up to 1kHz
Is that something someone (or you) has tried?
I just do not want to spend too much time on DSP if that is not really going to work for my personal preferences, hence this question.
PS: I did look at the .emo files REW creates. I am not sure if REW introduces time alignment in those files (I could be totally wrong)
You've got the right idea... If you start from the assumption that the source is correct... Then "box speakers" can potentially deliver an accurate reproduction of that waveform... And Dirac Live will then do its best to measure and perfect the version of this waveform that arrives at the listening position. In contrast, with dipoles, you are getting a sort of mishmash of direct and reflected waveforms... And, while this may sound very nice, or may even sound "more like live natural sound" to you, it is quite likely to confuse Dirac Live to at least some degree. So, yes, Dirac Live IS "trying to make your Magneplanars sound like what you would hear from an accurate box speaker"... (And, while some people have achieved favorable results with Dirac Live and Maggies, you may or may not be able to in your particular system.) |
|
|
|
Post by rajdude on Jul 14, 2023 12:58:50 GMT -5
Hmmm, well, thanks for your offer to help. But by reading what your setup is......you are essentially trying to absorb all the back-wave of these dipoles. To me (personally), that back-wave is what is giving my Maggies that "airy" and "open" sound and I like that. So I rather not eliminate that part of the sound.
My room is a simple living room, bare walls, no bass traps nothing. It is very bouncy in terms of sound :-( Since it is a rental house, I try to minimize stuff on the walls. I do have two 15" Rythmik subs working in stereo placed right next to the fronts which have Bryston monoblocks sitting behind them. I have a CC3 center channel and simple box type speakers for surrounds (for now). I did buy MC1s for surround a couple of years ago but they have been sitting in a corner because I need to make a stand for them sheesh! (back wall cannot accommodate them on the wall, because I have a large window there).
Almost all of this setup is almost 20 years old now :-(
Hello everyone, Has anyone got good results with Dirac on a XMC-1 feeding Magnepan speakers, specifically the 3.6r?
I have that combination and every time I try Dirac correction, my speakers sound weird. What exactly happens is this: the large soundstage (which is typical of Magnepan speakers) collapses into a small space, mostly towards the center. It sounds like Dirac is making my Maggies sound like box type speakers. I wonder if Dirac is designed for box type speakers and has trouble with measuring and correcting dipole speakers. Bipoles have so much sound bouncing off the back walls and side walls that (maybe) Dirac gets confused?
Strictly technically, the DSP correction Dirac generates is not bad at all. I used REW to verify it. It just sounds weird.
In the past, I heard that for Maggies to sound good with Dirac, we need to tell Dirac NOT to correct over approximately 1kHz. That is what I did for my third try. It still sounds crappy. I could lower that to maybe just 500 Hz and try again.
Any tips from experienced people in the same boat here?
Thanks for your time -Raj
I've been using Dirac with an XMC-1 and now XMC-2 for about five years. My fronts are 3.7, and I also use two DWM panels, a CCR, LRS for surrounds and MMGW for front tops. So as you might expect ... I'm very happy with Dirac. I correct full range on all the speakers with a flat target curve. But ... there are considerations and most have to do with the room. In my room I have a lot of bass traps and a lot of diffusion, but I have tracked down some reflections that significantly affected imaging and soundstage. Dirac's impulse response correction can get confused if the out of phase rear dipole wave gets to the mic with sufficient energy ... Dirac may align to the reflection and not the main signal, or it may even flip the polarity of the speakers. So generally speaking there is no reason why Dirac will not work with Maggies - and my experience is that it works really, really well - but there are conditions that could cause the issues that you observe. So if you can give a little information on the nature of your room I might be able to help. |
|
KeithL
Administrator
Posts: 9,961
|
Post by KeithL on Jul 14, 2023 13:00:11 GMT -5
The filters REW creates are basically imported as "PEQ settings" so should only be doing EQ. I think you've got the right idea... except that midrange falls into the portion of the spectrum that is critical to how things sound. Personally, I think I would try the entirely manual route... Download some actual full range pink noise tones... Play them through your subs... And use REQ as a manual RTA (real time analyzer)... REW works really well as a spectrum analyzer / display... And it does all sorts of other cool measurements as well... And adjust the EQ bands manually... Just don't get carried away... You don't need your subs flat to a fraction of a dB... And you shouldn't need more than one or two PEQ bands... Just use one or two properly configured PEQ bands to level out any big dips or peaks... (And, of course, figure out the best place to put your subs before trying to adjust them with EQ.) Thanks for chiming in here Keith! What you said makes perfect sense. It also means I am barking up the wrong tree - actually VERY WRONG tree :-) Ha! ha!
ok, so rather than using Dirac to auto-correct......which I am assuming introducing lots of filters and time-delays etc to "fix" my "preferred" crappy bouncy sounding maggies (lol), what if I did this: Use REW to generate simple filters to correct subwoofer, lows and low-mids ONLY...maybe only up to 1kHz
Is that something someone (or you) has tried?
I just do not want to spend too much time on DSP if that is not really going to work for my personal preferences, hence this question.
PS: I did look at the .emo files REW creates. I am not sure if REW introduces time alignment in those files (I could be totally wrong)
You've got the right idea... If you start from the assumption that the source is correct... Then "box speakers" can potentially deliver an accurate reproduction of that waveform... And Dirac Live will then do its best to measure and perfect the version of this waveform that arrives at the listening position. In contrast, with dipoles, you are getting a sort of mishmash of direct and reflected waveforms... And, while this may sound very nice, or may even sound "more like live natural sound" to you, it is quite likely to confuse Dirac Live to at least some degree. So, yes, Dirac Live IS "trying to make your Magneplanars sound like what you would hear from an accurate box speaker"... (And, while some people have achieved favorable results with Dirac Live and Maggies, you may or may not be able to in your particular system.) |
|
|
|
Post by marcl on Jul 14, 2023 13:11:00 GMT -5
Hmmm, well, thanks for your offer to help. But by reading what your setup is......you are essentially trying to absorb all the back-wave of these dipoles. To me (personally), that back-wave is what is giving my Maggies that "airy" and "open" sound and I like that. So I rather not eliminate that part of the sound.
My room is a simple living room, bare walls, no bass traps nothing. It is very bouncy in terms of sound :-( Since it is a rental house, I try to minimize stuff on the walls. I do have two 15" Rythmik subs working in stereo placed right next to the fronts which have Bryston monoblocks sitting behind them. I have a CC3 center channel and simple box type speakers for surrounds (for now). I did buy MC1s for surround a couple of years ago but they have been sitting in a corner because I need to make a stand for them sheesh! (back wall cannot accommodate them on the wall, because I have a large window there).
Almost all of this setup is almost 20 years old now :-(
I've been using Dirac with an XMC-1 and now XMC-2 for about five years. My fronts are 3.7, and I also use two DWM panels, a CCR, LRS for surrounds and MMGW for front tops. So as you might expect ... I'm very happy with Dirac. I correct full range on all the speakers with a flat target curve. But ... there are considerations and most have to do with the room. In my room I have a lot of bass traps and a lot of diffusion, but I have tracked down some reflections that significantly affected imaging and soundstage. Dirac's impulse response correction can get confused if the out of phase rear dipole wave gets to the mic with sufficient energy ... Dirac may align to the reflection and not the main signal, or it may even flip the polarity of the speakers. So generally speaking there is no reason why Dirac will not work with Maggies - and my experience is that it works really, really well - but there are conditions that could cause the issues that you observe. So if you can give a little information on the nature of your room I might be able to help. Actually there's very little broadband absorption in the room. It's all diffusion behind the Maggies so as NOT to absorb the back wave ... just diffuse it. |
|
|
|
Post by rajdude on Jul 14, 2023 13:35:02 GMT -5
oh, I understand better now. thanks.
Just out of curiosity: Do you remember the change in your Maggie's sound - before and after the diffusers?
I am just wondering if I should add some diffusion at the backside of my maggies, something which simply stands on the floor and not gets nailed/glued to the walls. But I am not sure what could be like that. Maybe a room divider type thingy filled with lots of different shaped wooden blocks?
What are you using as a diffuser?
Hmmm, well, thanks for your offer to help. But by reading what your setup is......you are essentially trying to absorb all the back-wave of these dipoles. To me (personally), that back-wave is what is giving my Maggies that "airy" and "open" sound and I like that. So I rather not eliminate that part of the sound.
My room is a simple living room, bare walls, no bass traps nothing. It is very bouncy in terms of sound :-( Since it is a rental house, I try to minimize stuff on the walls. I do have two 15" Rythmik subs working in stereo placed right next to the fronts which have Bryston monoblocks sitting behind them. I have a CC3 center channel and simple box type speakers for surrounds (for now). I did buy MC1s for surround a couple of years ago but they have been sitting in a corner because I need to make a stand for them sheesh! (back wall cannot accommodate them on the wall, because I have a large window there).
Almost all of this setup is almost 20 years old now :-(
Actually there's very little broadband absorption in the room. It's all diffusion behind the Maggies so as NOT to absorb the back wave ... just diffuse it. |
|
|
|
Post by rajdude on Jul 14, 2023 13:40:53 GMT -5
Thanks for your tips, Keith! I will try that route this weekend. I think REW can generate pink noise also but I could be wrong.
PS: I now am knocking myself for buying the XMC-1 now :-( I bought it specifically for Dirac DSP in it, and then later I paid that 100 bucks to Dirac. All for nothing. I wish I knew what I know now, earlier. Although, it has been a couple of years, and it does sound very, very good! My previous one was an Anthem AVM20. Maybe I could have got something with simple analog PEQs instead, if there is such a thing, lol
The filters REW creates are basically imported as "PEQ settings" so should only be doing EQ. I think you've got the right idea... except that midrange falls into the portion of the spectrum that is critical to how things sound. Personally, I think I would try the entirely manual route... Download some actual full range pink noise tones... Play them through your subs... And use REQ as a manual RTA (real time analyzer)... REW works really well as a spectrum analyzer / display... And it does all sorts of other cool measurements as well... And adjust the EQ bands manually... Just don't get carried away... You don't need your subs flat to a fraction of a dB... And you shouldn't need more than one or two PEQ bands... Just use one or two properly configured PEQ bands to level out any big dips or peaks... (And, of course, figure out the best place to put your subs before trying to adjust them with EQ.) Thanks for chiming in here Keith! What you said makes perfect sense. It also means I am barking up the wrong tree - actually VERY WRONG tree :-) Ha! ha!
ok, so rather than using Dirac to auto-correct......which I am assuming introducing lots of filters and time-delays etc to "fix" my "preferred" crappy bouncy sounding maggies (lol), what if I did this: Use REW to generate simple filters to correct subwoofer, lows and low-mids ONLY...maybe only up to 1kHz
Is that something someone (or you) has tried?
I just do not want to spend too much time on DSP if that is not really going to work for my personal preferences, hence this question.
PS: I did look at the .emo files REW creates. I am not sure if REW introduces time alignment in those files (I could be totally wrong)
|
|
|
|
Post by marcl on Jul 14, 2023 15:32:04 GMT -5
oh, I understand better now. thanks.
Just out of curiosity: Do you remember the change in your Maggie's sound - before and after the diffusers?
I am just wondering if I should add some diffusion at the backside of my maggies, something which simply stands on the floor and not gets nailed/glued to the walls. But I am not sure what could be like that. Maybe a room divider type thingy filled with lots of different shaped wooden blocks?
What are you using as a diffuser?
Actually there's very little broadband absorption in the room. It's all diffusion behind the Maggies so as NOT to absorb the back wave ... just diffuse it. This has been a journey. I started many years ago with the Tympani 1D and a lot of Sonex absorption behind and to the sides of the speakers ... even on the ceiling. 11 years ago when I replaced the 1D with 3.7 I got back to reading about placement of Maggies and room treatment. And I got REW. When I got the 3.7s in the room I literally broke the piece of foam that was in the packing material in half, and put a piece behind each speaker. At that point there were no bass traps or other treatment in the room. The soundstage immediately became deeper! So I looked for diffusers and found some 1D diffusers about 3" thick made of polystyrene. I put a dozen of them behind the speakers over the Sonex and the imaging and soundstage improved significantly. But I learned those diffusers only work down to maybe 1KHz. So along the way I focused on bass traps because - now having REW - I learned my room had HUGE resonance issues. I added bass traps which made a big difference but there was still a 15db peak at 40Hz followed by a significant cancellation at 70Hz. I found a device called the Antimode 8033 and used it to EQ my sub, and crossed the 3.7s up around 50Hz. I also started using REW to create parametric EQ for the 3.7s using JRiver Media Center filters. This was all before the XMC-1, so I had to do the PEQ outside my older Outlaw processor. These filters are basic filters, not anything like the sophisticated mixed phase proprietary algorithms that Dirac uses. And there was no way to do impulse response correction. I learned of Dirac, and got the XMC-1 in 2018. Around that time I also found the Vicoustic Multifuser DC2. This is a 2D "skyline" type diffuser about 6" thick. I covered my front wall behind the speakers with these and another quantum leap forward in imaging and soundstage resulted. As my wife put it, a warmer sound too. I also had removed all absorption from the ceiling except for the first reflection point half way between the speakers and the couch, and on the side walls also half way. This was critical, because without absorption there the image collapsed into the background. Sorry for the long story, but here's the bottom line. Yes, diffusers behind and to the sides (behind, not in front) of Maggies make a huge difference in soundstage and imaging without absorbing and making the room too dead. The back wave is scattered but the energy stays in the room and enhances the ambient sound around the speakers while also resulting in better image definition. Wendell Diller from Magnepan - on his tour of every dealer in North America - said the best sounding rooms had diffusion like this. I heard his demo of the "30.7 for Condos" in one of those rooms. You can use 1D diffusers that work only left/right, or 2D diffusers that work in all directions. But they have to be at least 6" deep to work to low enough frequencies. And yes you can mount them on some sort of stands so you don't have to attach them to the walls. Absorption at the first reflection side walls and overhead if you can will also improve imaging. I'll just say one more thing about Dirac. As I said, I used simple PEQ with JRiver, then Dirac V1 with XMC-1, then for a few months back to PEQ with the XMC-2 before Dirac V3 was available, and now three years with Dirac in the XMC-2. Whenever there has been an imaging issue it has always been a room or speaker placement issue ... never with Dirac. One thing I would suggest though is try a wider or narrower measurement pattern than whatever you have tried. 9 points, but spaced over a wider or narrower area of your listening position. See if that changes anything. Very often strange results from Dirac are tracked down to too few measurements, or too narrow a pattern around one seat. |
|
KeithL
Administrator
Posts: 9,961
|
Post by KeithL on Jul 14, 2023 17:02:21 GMT -5
REW can do almost anything... but sometimes it can be tricky to get it to output what you want from the computer to your audio system. (Sometimes it's easier to just make up a nice long pink noise or white noise file and play it on a player program.) Thanks for your tips, Keith! I will try that route this weekend. I think REW can generate pink noise also but I could be wrong. PS: I now am knocking myself for buying the XMC-1 now :-( I bought it specifically for Dirac DSP in it, and then later I paid that 100 bucks to Dirac. All for nothing. I wish I knew what I know now, earlier. Although, it has been a couple of years, and it does sound very, very good! My previous one was an Anthem AVM20. Maybe I could have got something with simple analog PEQs instead, if there is such a thing, lol
The filters REW creates are basically imported as "PEQ settings" so should only be doing EQ. I think you've got the right idea... except that midrange falls into the portion of the spectrum that is critical to how things sound. Personally, I think I would try the entirely manual route... Download some actual full range pink noise tones... Play them through your subs... And use REQ as a manual RTA (real time analyzer)... REW works really well as a spectrum analyzer / display... And it does all sorts of other cool measurements as well... And adjust the EQ bands manually... Just don't get carried away... You don't need your subs flat to a fraction of a dB... And you shouldn't need more than one or two PEQ bands... Just use one or two properly configured PEQ bands to level out any big dips or peaks... (And, of course, figure out the best place to put your subs before trying to adjust them with EQ.) |
|
|
|
Post by audiobill on Jul 14, 2023 20:12:13 GMT -5
IMO and experience, skip Dirac, study REW and use Roon's Convolution feature.
I think Dirac is a dumbed down solution, marketed as a "feature".
|
|
|
|
Post by fbczar on Jul 16, 2023 14:56:40 GMT -5
IMO and experience, skip Dirac, study REW and use Roon's Convolution feature. I think Dirac is a dumbed down solution, marketed as a "feature". Have you considered impulse response correction? I suppose you could use REW to measure for impulse response and move your speakers to correct a problem, but REW cannot address impulse response itself, like Dirac can. |
|
|
|
Post by audiobill on Jul 16, 2023 15:10:52 GMT -5
I suppose fir filters are helpful for higher frequencies, but most folks seem to use Dirac for bass.
Best,
|
|
|
|
Post by fbczar on Jul 16, 2023 15:34:50 GMT -5
I suppose fir filters are helpful for higher frequencies, but most folks seem to use Dirac for bass. Best, Dirac is fully capable of addressing the full frequency spectrum. I have Mangnepan 3.7i's in my Music / Home Theater system and use Dirac to provide a flat Custom Curve for all frequencies. Works great in my situation. It also works wonders for Atmos. Of course, the way you conduct measurements is key. |
|
