Would Dirac really help me cure my room's acoustic problems?

[flak]

I wish Dirac would make an external unit since I have multiple sources (i.e. Oppo, turn table, etc.) and that it worked with Windows XP. Hint, hint! Wink, wink!

 

Welcome Weverb,

your first post raises an interesting issue... additional sources to be corrected.

 

Processing an external source is an additional feature that is not fully supported by Dirac Live but it is often (but not always) possible if an audio card is used:

 

Windows 7 (stereo only):

1. Connect the external source output to a sound card input

2. Open the Sound control panel as described in Making sure DAP is used, above.

3. In the Sound control panel, select the Recording tab

4. Select the sound card input that the external source was connected to

5. Click the Properties button

6. In the Properties window, select the Listen tab

7. Check the “Listen to this device” checkbox

8. Press OK or Apply

 

Mac:

1. Connect the external source output to a sound card input

2. Use the free application LineIn by Rogue Amoeba or similar to feed the sound

card input into the Dirac Audio Processor™. LineIn can be downloaded at:

Rogue Amoeba | Freebies: Free Software for Mac OS X

 

Ciao, Flavio

 

b.t.w. Even Windows XP is unsupported but it may work...

[weverb]

Thanks flak. I would still prefer an external unit to replace the parametric eq I currently have. Like the OP, my issues are with reflection points, not being able to use acoustical treatments, and less than ideal speaker placement.

[Johan]

Did you already try it on a Windows XP machine? I am running Dirac Live from an XP laptop without any issues.

[weverb]

Did you already try it on a Windows XP machine? I am running Dirac Live from an XP laptop without any issues.

 

Cool! Now I need to check what type of file came with my calibrated mic.

[nununo]

Hi Nuno,

b.t.w. there are less expensive measurement mics: Dirac Online Store. UMIK-1 USB Measurement Microphone

Flavio

 

Hi Flavio, I will buy that one instead. Even if I pay for import taxes (which I hope not to) it will still be cheaper, I guess. But I wonder: it only has a 3m USB cable. That means I'll have to move my MacMini around in order to be able to do the measurements, right?

[flak]

Hi Nuno,

 

yes, 3 meters could be too short... you may buy a longer USB cable but I suggest that it is not longer than the USB 2.0 specification of 5 meters unless you wanna use an active one which can be rather long but will also cost more.

 

Ciao, Flavio

[nununo]

Hi Flavio, most probably before buying a bigger one I'll try to move my system to the center and see if 3m is enough like that. Looking forward to do the trial.

 

Thanks,

Nuno

[nununo]

Hello Flavio,

 

I finally got my microphone and yesterday did the first measurements out of which created my first filter. Actually the measurements were done all wrong because I didn't look into all the 3 different seat perspectives and thus didn't properly place the microphone in several of the measurement spots. So my results are still not significant.

 

Even so, without much listening (since it was late and I felt sorry for the neighbors) I think the sound got more focused. But these are premature conclusions. Once I do a proper measurement I'll share them here and ask for advice.

 

But I write this post to ask you one question: after putting the filter in place and playing through it in A+, the sound volume got much much much lower. Which means now I must set the preamp volume at levels never before tried in order to get the same volume as before. Is this normal? What is the reason for such big difference? I would understand a small overall reduction but it is quite substantial. At the point that now I'm afraid of forgetting the volume knob at that position and fear for my speakers if I inadvertently play music without the filter.

 

Thanks in advance,

Nuno

[flak]

Good morning Nuno,

 

yes, it's normal to accomodate for the boost at some frequencies, which is variable as it depends on the required correction... an alternative to play with will be available soon when the 1.0.4 update will be released as it will be possible to adjust the DSP gain.

If you send me a PM indicating if you are using a Pc or a Mac, a stereo or a full version, a trial or a paid license, I'll let you have a beta 1.0.4 and a revised manual with the relative instructions.

 

Good listening, Flavio

 

P.S. I've seen I have those data so I only need a PM with your email :-)

[flak]

Hi Nuno,

 

in case the sequence of microphone positions is not clear I'm posting the following two images,

please note that anyhow a minimum distance of 50 cm. from the head (the first measurement) is strongly suggested to avoid to eventually incur into over correction (so the parallepipeds should be bigger)

 

 

 

Ciao, Flavio

 

p.s. "Alto" in italian means "high" and "Basso" means "low"

[nununo]

Thanks for the precious pictures and information.

 

I realized that the maximum output of Dirac Live is 96kHz. But the maximum output of Amarra iRC is 384kHz. Since they're using Dirac's technology I was expecting the specs to be similar. What is the reason for this 96kHz limitation? Is it temporary? Will it be improved in a future update? This can be limiting if I want to listen to 196kHz or DSD stuff.

 

About the target. I'm clueless about how to design mine. The program suggest a target by default. This target seems to stay horizontal in the high frequencies beyond what I'd expect. It doesn't follow the average response frequency of most speakers which tend to attenuate at some point. Is this normal? Put in another way: when I will design my target, it's first version shouldn't deviate too much from the average response curve of my particular speakers, right? Hopefully removing bumps and holes, of course. And then I should gradually improve it in order to better suit my listening environment. Is that it?

 

If the above question is answered in the manual please disregard it. I still didn't get to read it but I will definitely do it before trying another measurement

 

Thanks,

Nuno

[Gadgety]

Hi Nuno,

 

Having used Tact's, rather than Dirac's, room correction in all its variations since 1999 this is my learning:

 

1. Room correction improves timbre, imaging, timing and rhythm etc. Sounds with room correction become "calmer" and it's just a better quality. Sound without room correction seems bloated, and is less relaxing. The sound improvement is order of magnitudes better than hardware improvements, once you are at a reasonable level.

2. Room correction is good at slicing off peaks in excited room modes, but if there are troughs it can't lift them.

3. Reflections are reflections, they don't go away because you have room correction. They can be dealt with in many ways for example through absorption, or where you place the listening seat.

4. Room dimensions, and where you sit in the room, will influence what modes are excited at the point of measurement.

5. With a microphone and measurement software you can do a lot to the actual room, and your placement of the listening seat. The more you do to the room, the less the room correction has to do. The path to the best possible sound is to adjust the room, the speakers placement, and the listening seat placement. Measure and reameasure. Then when sound is as good as you can get it, add the digital room correction.

6. I haven't tried Diracs stuff, but have friend who's tried Tact, Lyngdorf and Diracs solutions, and measured them all with a second measurement units had established that the Tact solution offers the best performance and tweakability. Too bad they're gone.

[tympathique]

Hi all

Hi Nuno,

 

Having used Tact's, rather than Dirac's, room correction in all its variations since 1999 this is my learning:

 

1. Room correction improves timbre, imaging, timing and rhythm etc. Sounds with room correction become "calmer" and it's just a better quality. Sound without room correction seems bloated, and is less relaxing. The sound improvement is order of magnitudes better than hardware improvements, once you are at a reasonable level.

2. Room correction is good at slicing off peaks in excited room modes, but if there are troughs it can't lift them.

3. Reflections are reflections, they don't go away because you have room correction. They can be dealt with in many ways for example through absorption, or where you place the listening seat.

4. Room dimensions, and where you sit in the room, will influence what modes are excited at the point of measurement.

5. With a microphone and measurement software you can do a lot to the actual room, and your placement of the listening seat. The more you do to the room, the less the room correction has to do. The path to the best possible sound is to adjust the room, the speakers placement, and the listening seat placement. Measure and reameasure. Then when sound is as good as you can get it, add the digital room correction.

6. I haven't tried Diracs stuff, but have friend who's tried Tact, Lyngdorf and Diracs solutions, and measured them all with a second measurement units had established that the Tact solution offers the best performance and tweakability. Too bad they're gone.

I agree once again, as many others have said.

I join my humble opinion to the list of needed to be convinced of DRC to be a standalone solution.

 

For me too it’s a question of mixed solution to the problem. Something good for mixed DRC filter solution (DIRAC) may be good for the entire system, software-hardware-room.

As many pointed, and IMHO, acoustic treatment is the first thing to do. When it’s done, let’s go for the icing with DRC.

Points of logic, as I see it;

- Acoustically untreated room will produced summation-cancellation of +- amplitudes. Greater the amplitude to correct, greater the amount of DRC correction to be applied.

- We pointed out that STRONG DRC corrections will generate dynamic range deterioration or clipping when ultimately the signal comes to the analog driver or amplifier. Please keep in mind the word “STRONG”

- Reducing output level to avoid clipping from DRC treatment at a specific frequency that has not been acoustically reduced will not eliminate the problem at that frequency.

- DRC alone correction passes all the stress to the analog components (poweramp-drivers). While a mixed acoustic-DRC treatments seems to me as a logic solution to reduce that stress.

In conclusion, I can’t get the idea of DRC correction as a unique solution for HR reproduction. Fine for a car but not for a dedicated room. A living room with all compromises may need mor DRC than a dedicated room, I don’t know if the word WAF is politically correct but it is a inevitable factor (could be the male version, MAF), that push people to a shorter and a compromise solution. So, there is place for DRC but, I agree that we should be wiser than computers or at least be awared of what it's doing.

 

regards

jacques

[tympathique]

oups!

[flak]

Hi Nuno,

 

yes, the limit of 96 KHz is temporary as it will soon be lifted in an upcoming free update.

Audirvana and Amarra anyhow downsample so that there is no problem in playing 192 KHz files.

Also I understand that some players can convert DSD files on the fly.

 

Your approach of preserving in the target curve the rolloff at high frequencies that is shown in your before correction curve is OK... you can easily create a second target curve and compare them by instantly switching the two filter sets in the DAP.

 

Good listening, Flavio

[nununo]

Hi Flavio,

 

You said in another thread that the "Sofa" is your favorite option. Also, in a review of your product I read that the author had tried the "Chair" option and was about to quit due to loss of some sound qualities but then he got a feedback from Dirac telling him to try the "Sofa" option and that was enough to solve his problems (here: Dirac Live Room Correction Suite | AudioStream).

 

I chose "chair" because common sense told me I would get better results for a smaller target. But given you favoring "Sofa" and Dirac also having suggested it, I wonder: why do you have the "chair" option at all?

 

So, I guess in my next measurements I should go for the "sofa" option, right?

 

Thanks,

Nuno

[nununo]

Gadgety and jacques, thanks for your precious inputs.

 

My room has serious limitations: sliding wooden door on the left of one speaker, huge floor to ceiling window on the right of the other. And I don't want to bloat my living room with a large ceiling diffuser (neither does my gf). I do consider, though, adding diffusion to the wall behind me (which she finds acceptable and ultimately can even be made beautiful), which is still about 2.5m away. But I'd like to try Dirac before engaging in these more expensive solutions.

 

Cheers,

Nuno

[flak]

Hi Nuno,

 

a warning about too small a volume has now been added so that a minimum distance of 50 cm. from the measurement point is going to be used in the chair option (a sort of a large imaginary 1/1,2 meter wide chair)

In that case the chair position is OK and you can successfull use it.

The sofa option combines a good performance with more flexibility in listening position and that's the reason I like it... won't your girlfriend listen with you ? :-) :-) :-)

 

Good listening, Flavio

[nununo]

Great Flavio, thanks. Understood.

[dallasjustice]

Flavio,

I think it's interesting that DIRAC uses a rectangular formation for the multi-mic approach. I know that JJ Johnston (post#21) has done some significant work to overcome omnimic limitations with regard to measurement for DSP. I think his conclusion was that a tetrahedron shape works best. Do you have any thoughts on this? DIRAC's approach is unique, AFAIK, and probably contributes greatly to DIRAC's success in comparison to other full-range DSP.

 

 

Michael.

[nununo]

What about the advance forward and up and down of the microphone? Is it also at least 50cm? Or does that minimum only apply to the sides?

 

Thanks,

Nuno

[nununo]

Oh, and yet one more question: In this new scenario, where should the mic point at? Vertically pointing to the ceiling or horizontally straight forward pointing at the wall behind the speakers?

 

Thanks!

Nuno

[tympathique]

Gadgety and jacques, thanks for your precious inputs.

 

My room has serious limitations: sliding wooden door on the left of one speaker, huge floor to ceiling window on the right of the other. And I don't want to bloat my living room with a large ceiling diffuser (neither does my gf). I do consider, though, adding diffusion to the wall behind me (which she finds acceptable and ultimately can even be made beautiful), which is still about 2.5m away. But I'd like to try Dirac before engaging in these more expensive solutions.

 

Cheers,

Nuno

Hey nununo

You seems stuck on each sides yes. What about behind the speakers? You know that you need at least 1,2m (nose estimation) between spks and wall for scene image and less boomy LF. Bass traps? A nice schroeder 2d between spks well made is beautifull too, looks like a modern sculpture.

This week, I made a hudge progress with 6 in (15cm) thick mineral wool traps on the floor behind my open baffle's woofers. It fantastic how absorbing thoses floor waves has cleaned the audio signal in a way that back spk wall rebounds are sharper than ever..

 

I finished the work with DRC where large corrections were to be reduced between 100hz to 300hz, a very good thing.

 

Have fun

 

jacques

[dallasjustice]

The manual says to point the mic up. I always do that anyway. That's the way Nyal taught me to do it. :-)

 

Oh, and yet one more question: In this new scenario, where should the mic point at? Vertically pointing to the ceiling or horizontally straight forward pointing at the wall behind the speakers?

 

Thanks!

Nuno

[flak]

Flavio,

I think it's interesting that DIRAC uses a rectangular formation for the multi-mic approach. I know that JJ Johnston (post#21) has done some significant work to overcome omnimic limitations with regard to measurement for DSP. I think his conclusion was that a tetrahedron shape works best. Do you have any thoughts on this? DIRAC's approach is unique, AFAIK, and probably contributes greatly to DIRAC's success in comparison to other full-range DSP.

 

 

Michael.

 

Yes Michael, you are right as the nine measurements are a key factor in implementing the Dirac Live algorithm.

As far as the mic positioning is concerned I also agree with you and I personally prefer the vertical one but in the case of Nuno he should take into account that the mic position should conform to the mic calibration file.

 

Nuno's UMIK-1 features a 0° calibration file so I think that it should be used horizontally.

(the same mic can be purchased with a 90° calibration but at an additional cost i.e. from Cross Spectrum Labs)

 

Flavio