Is room correction cheating?

Sonnie

Senior Admin
Staff member
Joined
Apr 2, 2017
Messages
5,055
Location
Alabama
More  
Preamp, Processor or Receiver
StormAudio ISP Elite 24 MK3 Processor
Main Amp
McIntosh MC1.25KW Monoblock Amps
Additional Amp
StormAudio PA 16 MK3
Computer Audio
Intel NUC w/ Roon ROCK
Universal / Blu-ray / CD Player
Panasonic UB9000 4K UHD Player (for media discs)
Front Speakers
RTJ 410
Center Channel Speaker
MartinLogan Focus C-18
Front Wide Speakers
JTR Neosis 110HT
Surround Speakers
JTR Neosis 210RT
Surround Back Speakers
JTR Neosis 210RT
Front Height Speakers
JTR Neosis 110HT-SL
Rear Height Speakers
JTR Neosis 110HT-SL
Subwoofers
JTR Captivator 2400 x6
Other Speakers or Equipment
VTI Amp Stands for the Monoblocks
Video Display Device
Sony 98X90L
Remote Control
Universal MX-890
Streaming Equipment
FireCube for movies and Lenova Carbon X1 for Roon
Streaming Subscriptions
Lifetime Roon Subscription
Tidal
qobuz
Netflix
Amazon Prime
Satellite System
Dish Joey 4K
Other Equipment
Zero Surge 8R15W-1 | Salamander Synergy Equipment Stand
I am bad to start listening and just won't get up if it sounds really good. I may be listening for 2 or 3 hours at a time. I don't know, that may not be all that long, but we are talking critical listening, sitting in one position, hardly moving, other than to drink a sip of coffee or cola all along. It really has to be right for me to listen. I've had speakers in here previously that were just too harsh to listen to for even one song... a bit too forward in the upper mid to highs. During our last evaluation we had a speaker that had a bit of an edge on it and DL got rid of it. I have not studied the response to see if I could see what DL was doing, but it was noticeable, especially switching back and forth between the two. It's one of the few times I have noticed any significant difference in the mid to higher end of the response. Most of the time it's the bass and mid-bass that bothers me the worst.
 

tesseract

Senior Admin
Joined
Nov 12, 2016
Messages
1,266
Location
Lincoln, NE
More  
Preamp, Processor or Receiver
Emotiva XMC-1
Main Amp
Emotiva XPA-2 Gen 2
Additional Amp
Emotiva XPA-3 Gen 2
Other Amp
Dayton SA1000
Universal / Blu-ray / CD Player
Sony BDP S590 & Pioneer DV-610AV
Front Speakers
JTR NOESIS 210 RT - L/R mains
Center Channel Speaker
Chase SHO-10 - Center
Surround Speakers
Chase PRO-10 - Surrounds
Subwoofers
Chase VS-18.1 x 2 - Subwoofers
Video Display Device
Vizio E550VL
Streaming Subscriptions
h/k TC35C/Ortofon Super OM10/Pro-Ject Phono Box S
That’s certainly true of Dirac and Audyssey, that isn't true of YPAO, MCACC, ARC, EmoQ, etc. These all use various forms or IIR filters. The Yamaha and MCACC systems offer phase correction but it's always been unclear to me how it works, what it does, and I've never measured anything to help me understand.

My point about this earlier was you can apply IIr filters in the stochastic reason but only if you measure free space. Of course you can use Dirac in room and it's all a non-issue.

Yes, there is still a long ways to go. Dirac, DEQX and DSPeaker Anti-Mode 2.0 Dual Core are the only systems I would use at the moment.
 

tesseract

Senior Admin
Joined
Nov 12, 2016
Messages
1,266
Location
Lincoln, NE
More  
Preamp, Processor or Receiver
Emotiva XMC-1
Main Amp
Emotiva XPA-2 Gen 2
Additional Amp
Emotiva XPA-3 Gen 2
Other Amp
Dayton SA1000
Universal / Blu-ray / CD Player
Sony BDP S590 & Pioneer DV-610AV
Front Speakers
JTR NOESIS 210 RT - L/R mains
Center Channel Speaker
Chase SHO-10 - Center
Surround Speakers
Chase PRO-10 - Surrounds
Subwoofers
Chase VS-18.1 x 2 - Subwoofers
Video Display Device
Vizio E550VL
Streaming Subscriptions
h/k TC35C/Ortofon Super OM10/Pro-Ject Phono Box S
DIRAC actually uses its own measurement system and not REW. The love of DIRAC around here is more a coincidence than anything to do with REW.

Is Dirac resolution equal to REW? I think REW shows more.

there are in fact correction suites focused on linearizing the speaker bases on it's free space measurements and then correcting the room seperately. I actually like the concept or this approach but it's not end user friendly.

DEQX!
 

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
Is Dirac resolution equal to REW? I think REW shows more.



DEQX!

I believe that Dirac records the impulse just as raw as REW. What it shows you may be more smoothed.

I could be wrong, I have only used their computer software and it was some time ago now, but I swear I thought Dirac couldn't import an impulse response. It had to record its own within it's own software.

And yes DEQX can certainly do that. You can also do it with a MiniDsp that can handle FIR filters and one of the various FIR filter generators. I think one of them actually automates that process.
 
Joined
Oct 28, 2017
Messages
33
More  
Preamp, Processor or Receiver
Sherbourn preamp
Main Amp
Enlightened Audio Designs Powermaster 1000
Additional Amp
Emotiva UPA-700
Universal / Blu-ray / CD Player
Oppo UDP-203
Front Speakers
RCF monitor 6
Subwoofers
Bag End ELF
Streaming Subscriptions
Yamaha Motif XS
What I am trying to do is cop the performance of the DEQX system I was exposed to a few years back. It was developed by an Australian company whose roots included the Fairlight synthesizer and subsequent workstation, a very well conceived system who's cost ultimately doomed it. However they developed an algorithm that is patented, the concept of which it has the equivalent of a 20,0000 band equalizer (one band at every 1hz frequency between 20 and 20k! Each band can vary in both amplitude and phase (time) to a wide degree with no phase shift of it's own. Their concept is 1. Speaker correction 2, Room correction 3. preference EQ curves. Some speaker manufacturers have used this system as a building block for their systems, and at the Rocky Mountain Hi Fi show I was at, had the best sound at the show (in my opinion). I heard an account of a witness to a 3 day demo they did at Westlake Audio in LA. Westlake is a famous recording studio and make a highly regarded line of studio monitors which carry their name. In other words they are real audio dogs. Anyway they worked 3 days straight trying out the system with their own monitors, and ended up purchasing the system for installation in their studio. The same thing happened at Abbey Road in London. The REW system is used by MiniDSP to load correction files into their DIRAC enabled systems. What I am interested in is putting together something on a beer budget that tastes like fine wine. I intend to research the REW-MiniDSP capability, and of course I will do the crossovers in their DSP experimenting with FIR filters for the crossovers. In the 80's I hand built about 20 sets of studio monitors using Focal drivers and custom designed crossovers. I used to take them to Georgetown Masters in Nashville and with the great ears of Denny Purcell I proofed my designs against his $100,000.00 Duntech monitors so I do have some practical experience in the field. My this is a brave new world! I do know that the loudspeaker is the highest distortion element in an audio system, so all my efforts will be to minimize the distortion and bring the listening experience closer to the real thing.
 
Joined
Oct 28, 2017
Messages
33
More  
Preamp, Processor or Receiver
Sherbourn preamp
Main Amp
Enlightened Audio Designs Powermaster 1000
Additional Amp
Emotiva UPA-700
Universal / Blu-ray / CD Player
Oppo UDP-203
Front Speakers
RCF monitor 6
Subwoofers
Bag End ELF
Streaming Subscriptions
Yamaha Motif XS
Oh rats, the DIRAC part of the Mini DSP uses the DIRAC part of the unit's software. REW can program in plug ins however, and at the price of this stuff I have to try that out to the max. Alas the DEQX model I would want costs $5,495.00 a little beyond my toy range right now.....
 

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
Oh rats, the DIRAC part of the Mini DSP uses the DIRAC part of the unit's software. REW can program in plug ins however, and at the price of this stuff I have to try that out to the max. Alas the DEQX model I would want costs $5,495.00 a little beyond my toy range right now.....

You and me both!

There is no mistaking the Minidsp for the Deqx but some of their models have really great sophistication and could be used to accomplish similar goals.
 
Joined
Oct 28, 2017
Messages
33
More  
Preamp, Processor or Receiver
Sherbourn preamp
Main Amp
Enlightened Audio Designs Powermaster 1000
Additional Amp
Emotiva UPA-700
Universal / Blu-ray / CD Player
Oppo UDP-203
Front Speakers
RCF monitor 6
Subwoofers
Bag End ELF
Streaming Subscriptions
Yamaha Motif XS
You and me both!

There is no mistaking the Minidsp for the Deqx but some of their models have really great sophistication and could be used to accomplish similar goals.
Oh yeah, $400.00 for their Dirac enabled box is an unbelievable bargain! It is worth it just for the eq's and crossovers!
 

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
Oh yeah, $400.00 for their Dirac enabled box is an unbelievable bargain! It is worth it just for the eq's and crossovers!

Yes for sure. I actually was thinking about a different box, but that one does seem to be a good deal. The one Sonnie has is also hugely capable and more similar to the DEQX. The main differences I see might be DAC quality and analogue output stage quality (I could be all wet on those differences too).
 

tesseract

Senior Admin
Joined
Nov 12, 2016
Messages
1,266
Location
Lincoln, NE
More  
Preamp, Processor or Receiver
Emotiva XMC-1
Main Amp
Emotiva XPA-2 Gen 2
Additional Amp
Emotiva XPA-3 Gen 2
Other Amp
Dayton SA1000
Universal / Blu-ray / CD Player
Sony BDP S590 & Pioneer DV-610AV
Front Speakers
JTR NOESIS 210 RT - L/R mains
Center Channel Speaker
Chase SHO-10 - Center
Surround Speakers
Chase PRO-10 - Surrounds
Subwoofers
Chase VS-18.1 x 2 - Subwoofers
Video Display Device
Vizio E550VL
Streaming Subscriptions
h/k TC35C/Ortofon Super OM10/Pro-Ject Phono Box S
I believe that Dirac records the impulse just as raw as REW. What it shows you may be more smoothed.

Yes, Dirac filters are many, what it shows seems smoothed considerably.
 

Flak

Member
VIP Supporter
Joined
Jul 4, 2017
Messages
160
I prefer not commenting DEQX or REW when possible but outline instead that Dirac's solution is a unique patented one and attach links that discuss many aspects of its implementation even if some aspects of the algorithm are not disclosed for obvious reasons.
I suggest drawing the attention to the portions that discuss how not only multiple measurements are necessary but the way they are used is even more important.

If you have the time and inclination here an interview with Mathias Johansson:
https://www.soundandvision.com/content/ridding-world-bad-sound
while here a link to a recent document on the subject:
http://diracdocs.com/ISEAT15_Brannmark_Sternad.pdf

In my admittedly biased opinion with Dirac Live we are on the cutting edge of this technology and that's the reason why it's being adopted by many knowledgeable manufacturers even if solutions at no cost are available.

May be we went even too far with the next possible step... Dirac Unison,
that technology is ready now, and more powerful than anything seen before, but as of now is very difficult to manage if not by a trained skilled engineer:
https://www.dirac.com/dirac-blog/story-dirac-filterlab
We are anyhow working hard at automating the process and we should see some kind of Unison embodiment in 2018.

:) Flavio
 
Last edited:

Flak

Member
VIP Supporter
Joined
Jul 4, 2017
Messages
160
Actually there is another very real reason to not use something like DIRAC. Your speakers have ribbon super tweeters and Dirac requires specialized hardware to run above 48khz sampling. The readily available hardware that allows you to avoid redundant ad/da conversion and a computer in the signal path is stuck at 48khz due to hardware limitations. It would defeat the whole purpose of that supertweeter.

Knowing of all the products in the market is difficult... actually this miniDSP unit has a maximum input sample rate of 216 kHz and operates at 96 kHz, it's all digital as well so no AD/DA conversions
https://www.minidsp.com/dirac-series/ddrc-22d

Flavio
 

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
I prefer not commenting DEQX or REW when possible but outline instead that Dirac's solution is a unique patented one and attach links that discuss many aspects of its implementation even if some aspects of the algorithm are not disclosed for obvious reasons.
I suggest drawing the attention to the portions that discuss how not only multiple measurements are necessary but the way they are used is even more important.

If you have the time and inclination here an interview with Mathias Johansson:
https://www.soundandvision.com/content/ridding-world-bad-sound
while here a link to a recent document on the subject:
http://diracdocs.com/ISEAT15_Brannmark_Sternad.pdf

In my admittedly biased opinion with Dirac Live we are on the cutting edge of this technology and that's the reason why it's being adopted by many knowledgeable manufacturers even if solutions at no cost are available.

May be we went even too far with the next possible step... Dirac Unison,
that technology is ready now, and more powerful than anything seen before, but as of now is very difficult to manage if not by a trained skilled engineer:
https://www.dirac.com/dirac-blog/story-dirac-filterlab
We are anyhow working hard at automating the process and we should see some kind of Unison embodiment in 2018.

:) Flavio

Very cool Flavio.

I didn't realize that you worked for Dirac that is great to have you hear.

I knew about the all digital Dirac but did not realize it had the ability to operate at 96khz. That's good to know. When I beta tested for you guys I believe the software implementation could also do that, but then I noticed most hardware implementations were limited to 48khz.

Your technology truly is great. I'm a huge believer.

I can understand Skips view too. If you show someone a speaker with the intent on selling them on that speaker, then use a product like dirac which can achieve near perfection in the response, a customer may be disappointed when they get home and hear something different. You would have to embed Dirac into the system in such a scenario. While I'm fine with that notion, not everybody is.
 

Flak

Member
VIP Supporter
Joined
Jul 4, 2017
Messages
160
Yes, Dirac filters are many, what it shows seems smoothed considerably.

Hi Tesseract :)

in order to compare REW's with Dirac's graphs conditions should as similar as possible, we have two options...
1- take a measurement in Dirac in one point only (for testing purposes only), that can be done by skipping the successive eight measurements by clicking Proceed
2- measure in nine positions in REW as well and average them (there is a dedicated button in REW)
You'll find out that (even if predicted in Dirac's case) the graphs are quite consistent and accurate no matter if the smoothing is slightly different (1/8 octave I believe)

Flavio
 

Flak

Member
VIP Supporter
Joined
Jul 4, 2017
Messages
160
Very cool Flavio.

I didn't realize that you worked for Dirac that is great to have you hear.

Great reading your comments as well... I try to make my affiliation as transparent as possible, that's the reason behind the Warning in my signature ,
as always "read the fine print" :)

P.S. I just realized that signatures are not shown if one has'nt logged in...
 
Last edited:

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
Great reading your comments as well... I try to make my affiliation as transparent as possible, that's the reason behind the Warning in my signature ,
as always "read the fine print" :)

P.S. I just realized that signatures are not shown if one has'nt logged in...

They don't show up on the iPhone Mobile screen either. I am logged in but can't see anyone's signature. I guess I've not read your signature when using my laptop to have ntoiced.
 

Flak

Member
VIP Supporter
Joined
Jul 4, 2017
Messages
160
They don't show up on the iPhone Mobile screen either. I am logged in but can't see anyone's signature. I guess I've not read your signature when using my laptop to have ntoiced.

Sorry, not much that I can do... here it is anyhow:
"Warning: My posts might be biased even if in good faith, I work for Dirac Research :)"
 

AudiocRaver

Senior Reviewer
Staff member
Joined
Nov 21, 2016
Messages
974
Location
North Carolina, USA
More  
Preamp, Processor or Receiver
Onkyo TX-SR705 Receiver
Main Amp
Crown XLS 1502 DriveCore-2 (x2 as monoblock)
Additional Amp
Behringer A500 Reference Power Amplifier
Front Speakers
MartinLogan Electromotion ESL Electrostatic (x2)
Center Channel Speaker
Phantom Center
Surround Speakers
NSM Audio Model 5 2-Way (x2)
Subwoofers
JBL ES150P Powered Subwoofer (x2)
First, I have zero experience with DEQX, so these comments will be in the theoretical realm.

Agree withe the OP that the concept of EQ the speaker, then EQ the room might sound like the ideal way to correct, but it is also conceivable that Dirac LIve's analysis/correction algorithm might very well accomplish the equivalent. As we discovered in our comparisons between Dirac LIve and Audyssey XT32, bending over backwards to accomplish an apples-to-apples comparison ended up dumbing down Dirac LIve and really held it back. Unleashed, Dirac Live was the clear winner.

Comparing best performance of DEQX vs. best performance of Dirac Live is not a simple matter, would require listening tests for a fair comparison, and who knows which would come out on top?

The bang-for-your-buck prize would almost HAVE to go to Dirac Live. An all-in-one approach that handles individual drivers and crossovers seamlessly calls for features not yet in Dirac (can be done with multiple miniDSP units, but that is not what I think of as seamless).

I would love to get my hands on DEEQX for comparison testing.
 
Last edited:

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
Dirac Unison is actually far closer to DEQX than is Dirac Live. DEQX must be setup by someone with expertise, it is not for novices (which is not to say novices haven't used it and enjoyed it).

Legacy Audio has also developed a proprietary correction scheme similar to what I had suggested as ideal. Unison could do the same and then some conceivably. When you move to the level of optimizing the speaker system itself and not the speaker room interaction (as Live does) you do have to begin to account for the speakers actual linear capabilities and dispersion pattern, which increases the complexity.

I do agree that Dirac live is the most capable automated room correction system on the market that I have experience with. I've experienced nearly everyone one but Trinnov and its been as good or better than all the others.

AudiocRaver, I've been meaning to ask you, when you did the single point measurements with Dirac, what caused the upward tilt of the response past about 15khz? Nearly every correction measurement you shared showed the same thing. I had never tried Dirac with a single measurement before so thought it was an artifact of that approach, but when I tried the same thing recently, I didn't get that artifact. Is it in your target curve?
 

AudiocRaver

Senior Reviewer
Staff member
Joined
Nov 21, 2016
Messages
974
Location
North Carolina, USA
More  
Preamp, Processor or Receiver
Onkyo TX-SR705 Receiver
Main Amp
Crown XLS 1502 DriveCore-2 (x2 as monoblock)
Additional Amp
Behringer A500 Reference Power Amplifier
Front Speakers
MartinLogan Electromotion ESL Electrostatic (x2)
Center Channel Speaker
Phantom Center
Surround Speakers
NSM Audio Model 5 2-Way (x2)
Subwoofers
JBL ES150P Powered Subwoofer (x2)
I noticed. It was not in the target curve, which extended to about 15 kHz, and then dropped off sharply. My first thought was to check the mic calibration file in REW and Dirac Live. That was correct, and it was the 90-degree version, as we needed, since the mic was always pointed up. We took REW measurements, then ran Dirac Live calibration, then measured with REW, all without moving the mic, so it is not a mic position issue. We did not have time to chase it down, and knew that little HF lift would make a minimal audible difference, if it was truly a difference and not just a measurement artifact.

It possible there is a slight HF sensitivity difference between Dirac and REW, or that it resulted somehow from our not taking multiple measurements for Dirac calibration.
 

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
I noticed. It was not in the target curve, which extended to about 15 kHz, and then dropped off sharply. My first thought was to check the mic calibration file in REW and Dirac Live. That was correct, and it was the 90-degree version, as we needed, since the mic was always pointed up. We took REW measurements, then ran Dirac Live calibration, then measured with REW, all without moving the mic, so it is not a mic position issue. We did not have time to chase it down, and knew that little HF lift would make a minimal audible difference, if it was truly a difference and not just a measurement artifact.

It possible there is a slight HF sensitivity difference between Dirac and REW, or that it resulted somehow from our not taking multiple measurements for Dirac calibration.

I still think its the single measurement but when I didn't see the same results with my measurement I wasn't so sure. I'll try again and let you know if I see the same behavior. Depending on how DIRAC applies that correction, there are actually some tweeters which overheat when an eq curve like that is applied. It has to extend past 20khz. The B&C DE250 CD is one such driver (which is what I use) if you boost the treble past 20khz the driver has been know to fry the voice coil. It would be good to know if certain ways of using DIRAC can inadvertently cause such a correction.
 

AudiocRaver

Senior Reviewer
Staff member
Joined
Nov 21, 2016
Messages
974
Location
North Carolina, USA
More  
Preamp, Processor or Receiver
Onkyo TX-SR705 Receiver
Main Amp
Crown XLS 1502 DriveCore-2 (x2 as monoblock)
Additional Amp
Behringer A500 Reference Power Amplifier
Front Speakers
MartinLogan Electromotion ESL Electrostatic (x2)
Center Channel Speaker
Phantom Center
Surround Speakers
NSM Audio Model 5 2-Way (x2)
Subwoofers
JBL ES150P Powered Subwoofer (x2)
I will watch it closer in the future, and do a little test to see if it is reproducible on my end.
 

Matthew J Poes

AV Addict
Thread Starter
Joined
Oct 18, 2017
Messages
1,903
Sounds good.

By the way I got Dirac working on my new laptop last night and was listening to a few recordings. I did a 3 measurement correction to be quick about it. I really need to work on my PEQ filters for the bass before commenting extensively, but I will just say that I was floored at the difference once again. Part of the problem is that I'm applying a house curve I haven't applied without Dirac, so the tonal balance is shifting. None the less, I found not only improvements in SS&I, but also a removal of a slight cupped hand sound.

There are some tweaks you can do to a stereo that make it sound better, but they don't necessarily engage you any more in the music. There are some tweaks, like Dirac, that make you want to listen to your favorite music all over again. I didn't want to go to bed last night I was having such a good time.
 

AudiocRaver

Senior Reviewer
Staff member
Joined
Nov 21, 2016
Messages
974
Location
North Carolina, USA
More  
Preamp, Processor or Receiver
Onkyo TX-SR705 Receiver
Main Amp
Crown XLS 1502 DriveCore-2 (x2 as monoblock)
Additional Amp
Behringer A500 Reference Power Amplifier
Front Speakers
MartinLogan Electromotion ESL Electrostatic (x2)
Center Channel Speaker
Phantom Center
Surround Speakers
NSM Audio Model 5 2-Way (x2)
Subwoofers
JBL ES150P Powered Subwoofer (x2)
Ah, yes, listening instead of bedtime, I have done that many times.

Dirac live might just be the ultimate anti sleep aid for audiophiles.
 
Top Bottom