Dirac / XMC-1 F.A.Q. (work in progress)

[KeithL]

Measurements taken using different methods will produce different results.

In specific, line arrays couple to the room rather differently than normal "point source" speakers.

1) The actual SPL will drop off at a different rate for a line source than for a "point source" speaker. (Your mains may be less efficient, but the sound level should drop off more gradually than with regular speakers as you get further away - it should be closer to an inverse square whereas point source speakers follow more or less of an inverse cube... and this will depend on where they are positioned relative to the walls.)
2) There will be less interaction with the ceiling and floor - because vertical line sources have rather narrow vertical dispersion.
3) How extreme this difference is will depend on your speakers and your room.

All of these factors would contribute to make it likely that you would see a bigger difference between different types of measurements than you would see with "normal" speakers. Since Dirac takes extra care to include room interactions into account, I wouldn't be especially surprised to see major differences between how it treats line sources and "regular speakers".

I would suggest you see how it SOUNDS and pick whatever way sounds best to you.

Jan 14, 2015 14:45:18 GMT -5 hesster said:

Anyone measure the individual Speaker Db levels with a SPL meter after running DIRAC, by using the speaker level test tones? 

Should these all be "close" to one another within a certain range, and if not, adjusted for balance or just rely on DIRAC? 

I am noticing a big difference (my mains are huge line arrays and not as "efficient" as the large center, smaller Klipsh Sides, and Bose Rear surrounds). 

[hesster]

So after a little research, more info on what exactly AGC is:

"Automatic gain control (AGC) is a closed-loop regulating circuit, the purpose of which is to provide a controlled signal amplitude at its output, despite variation of the amplitude in the input signal. The average or peak output signal level is used to dynamically adjust the input-to-output gain to a suitable value, enabling the circuit to work satisfactorily with a greater range of input signal levels."

"A potential disadvantage of AGC is that when recording something like music with quiet and loud passages such as classical music, the AGC will tend to make the quiet passages louder and the loud passages quieter, compressing the dynamic range; the result can be a reduced musical quality if the signal is not re-expanded when playing."

So - we do not want this on as it is modifying the +/- gain when running DIRAC. 

As I said earlier, my first experience turning it off was getting lots MORE dynamics out of my speakers when running DIRAC, and clipping/failure at certain Mic positions at settings I had used before even when adjusting to -24 db background noise and -12db average gain from all speakers.  The results were not good (mains way louder than other speakers, center and sub weak, surround field too low in 5.1/7.1 programs), so I am going to rerun a number of times today at lower test levels and see what I get.

[hesster]

So I was perplexed why my results were not good, and humbly admit that the MIC LEVEL IN THE SLIDER ADJUSTMENT MUST BE AT MAX!. Reran Dirac w/o a hitch, no faults, and most important all of the imbalance and volume quirks were resolved.  System now sounds outstanding, and just waiting on a replacement tweeter in the Center Channel to resolve an issue that Dirac is indicating in the 1.5K Hz and up region.

[ansat]

Big and RED at the top of the FAQ -- NOTE: You must turn off AGC in the system settings as well as turn your mic gain to the max before you run Dirac. Windows is turning on AGC by default. If you have not turned it off manually, then yours is likely still on and your measurements will need to be re-taken.

[hesster]

Yeah - my Duh.  Results demonstrate the impact this will have if you don't do this, and it is a wonder it even worked with it not fully turned up.

The sampling is much more uniform between each speaker even with different speaker dynamic efficiencies, and no more clipping .  After listening to a variety of program sources yesterday, the surround sound imaging and balance with my Fronts is stellar.

[ansat]

Jan 16, 2015 12:16:45 GMT -5 hesster said:

Yeah - my Duh.  Results demonstrate the impact this will have if you don't do this, and it is a wonder it even worked with it not fully turned up.

The sampling is much more uniform between each speaker even with different speaker dynamic efficiencies, and no more clipping .  After listening to a variety of program sources yesterday, the surround sound imaging and balance with my Fronts is stellar.

Lol I just added it. You didn't miss anything

[goozoo]

I took my measurements with AGC turned off and mic gain at 50% in windows. The results look as they should and I didn't really have to turn down the mic gain in DIRAC very much to hit the -24db mark. I will try again with 100% set in windows, but wouldn't this just mean you need to turn down the gain more in DIRAC?

[ansat]

Jan 17, 2015 11:34:11 GMT -5 goozoo said:

I took my measurements with AGC turned off and mic gain at 50% in windows. The results look as they should and I didn't really have to turn down the mic gain in DIRAC very much to hit the -24db mark. I will try again with 100% set in windows, but wouldn't this just mean you need to turn down the gain more in DIRAC?

goozoo, It really shouldn't matter all that much (but there really is no gain in the windows settings, only loss for this mic), but I am designing these faq's to complement the directions given by emotiva. For my setup, I have been doing something a little different that produces reliable and more importantly repeatable results for me. I set the volume of the mains to 75 db spl when running the trst sweep, then back the gain down to where the subwoofer falls in the middle of the green zone.

Tony

[richter250]

Jan 10, 2015 14:43:39 GMT -5 melm said:

Sounds better with this last run.. speakers are more precisely balanced. I do note that compared to my non-Dirac settings I really need to sit in the sweet spot to get the best balance and image. I have Maggies with a phantom center and to get dialog right I need to be in my best spot. Not a problem, just an observation...

So better with AGC turned off.

Thanks again Tony.

Have you tried your Maggies with the tweeter on the outside?  I did this and then reran Dirac and the sweet spot is much wider for music and movies.  Dirac does work with Maggies.  Mine have never sounded better.

[melm]

Jan 21, 2015 12:26:01 GMT -5 richter250 said:

Jan 10, 2015 14:43:39 GMT -5 melm said:

Sounds better with this last run.. speakers are more precisely balanced. I do note that compared to my non-Dirac settings I really need to sit in the sweet spot to get the best balance and image. I have Maggies with a phantom center and to get dialog right I need to be in my best spot. Not a problem, just an observation...

So better with AGC turned off.

Thanks again Tony.

Have you tried your Maggies with the tweeter on the outside?  I did this and then reran Dirac and the sweet spot is much wider for music and movies.  Dirac does work with Maggies.  Mine have never sounded better.

That's the way they are oriented now. Always sounded better to me that way.

Thanks

Mel

[Thunderduck]

Back in December I had to remove my Def Tech BP-7006s because I thought I may have blown one of the woofers. After further testing, I believe I may have been wrong and was going to put them back in my system and run Dirac.

After making sure that AGC was off, I started Dirac. I had to set the Input Gain all the way up to +20 to get a reading of -24. This is normal for my settings ever since it was discovered we should be turning off AGC. So far so good.

But, when I went to setup the Output Volume for the speakers, I may have run into an issue. When setting the Output Volume to get the Center, Surrounds, and Subwoofer into the green area, the Front Speakers only have a maximum level of -24. At no time to the Front Speakers come close to the green area. Is this ok?

I did not bother with continuing because I was not sure if this would invalidate the results or if in fact it would be ok to run with the speaker volume at this level and Dirac would adjust for it.

I will say that the BP-7006s sound good in the system when using the filter for the previous front speakers (shows you how good my hearing is). However, I would like to get a filter with these speakers in the mix and see how they sound with Dirac.

So, I guess what I need to know is, is it ok to run Dirac when one speaker or a set of speakers have a significantly lower volume level than the others?

Thank you for any help or advice.

Best Regards,

Steve

[melm]

I had an interesting experience I thought I'd share.

My Dirac update worked well, and the improvement from no room equalization was considerable. The more I listened, however, I noticed that my phantom center channel between my Maggie 3.6r speakers was a bit muddled. It was most obvious on movie dialog, where the voices sounded sort of out of phase, but the speakers were in phase, and when I switched Dirac off the center (phantom) was spot on.

Flavio suggested I re-run Dirac and just do one position, at my normal center couch sweet spot. I did, and the phantom center channel image was back and was solid... as good as without Dirac. Now I may have lost something by not doing the other mic positions, but my listening tests indicate that the sound has improved as much as before, but the center image is now stable.

I have a very irregular room, which the Maggies are very sensitive to. At any rate, Flavio's idea worked. My last remaining issue is gone. Bass (from my PSA sub) is solid, and instrument placement in music is better, and movies have sharper directionality as well.

I'm usually listening on my own, unless friends are over for a movie, so I'll listen from more positions in the room, but so far, trying a few locations off center, the speakers still sound better than before.

My situation with the 3.6r's may be unique, and my room is certainly a challenge, but doing a run from only the sweet spot made a dramatic improvement. I've saved my earlier Dirac runs and can always re-load them, but this sounds really, really good. Not sure if I have given up anything, but if so, I can't hear it.

FWIW.

Mel

[ansat]

Jan 27, 2015 22:25:25 GMT -5 Thunderduck said:

Back in December I had to remove my Def Tech BP-7006s because I thought I may have blown one of the woofers. After further testing, I believe I may have been wrong and was going to put them back in my system and run Dirac.

After making sure that AGC was off, I started Dirac. I had to set the Input Gain all the way up to +20 to get a reading of -24. This is normal for my settings ever since it was discovered we should be turning off AGC. So far so good.

But, when I went to setup the Output Volume for the speakers, I may have run into an issue. When setting the Output Volume to get the Center, Surrounds, and Subwoofer into the green area, the Front Speakers only have a maximum level of -24. At no time to the Front Speakers come close to the green area. Is this ok?

I did not bother with continuing because I was not sure if this would invalidate the results or if in fact it would be ok to run with the speaker volume at this level and Dirac would adjust for it.

I will say that the BP-7006s sound good in the system when using the filter for the previous front speakers (shows you how good my hearing is). However, I would like to get a filter with these speakers in the mix and see how they sound with Dirac.

So, I guess what I need to know is, is it ok to run Dirac when one speaker or a set of speakers have a significantly lower volume level than the others?

Thank you for any help or advice.

Best Regards,

Steve

Try a power cycle of the XMC and test again. Those defTechs should be pretty high on the scale with the powered subwoofers. Also try setting the gain on the 7006's to about 50% to ensure that the subwoofer is puting out its sound.

Tony

[Thunderduck]

Thank you Tony, will give it a go tomorrow. Just realized that I had turned the gain down to about the 10 O'clock position. May have had some effect on their reading.

Thanks again and good night.

Steve

[KeithL]

As long as Dirac Live doesn't complain you're OK (the software will issue an error if any of the speakers are so low it can't use the resulting test strip). That said, if you're running that low, you should be extra careful to avoid any extraneous noise or background noise that might "contaminate" the results.

Incidentally, it looks like the Full version (yes, it's getting closer) will allow you to adjust the levels individually

Jan 27, 2015 22:25:25 GMT -5 Thunderduck said:

Back in December I had to remove my Def Tech BP-7006s because I thought I may have blown one of the woofers. After further testing, I believe I may have been wrong and was going to put them back in my system and run Dirac.

After making sure that AGC was off, I started Dirac. I had to set the Input Gain all the way up to +20 to get a reading of -24. This is normal for my settings ever since it was discovered we should be turning off AGC. So far so good.

But, when I went to setup the Output Volume for the speakers, I may have run into an issue. When setting the Output Volume to get the Center, Surrounds, and Subwoofer into the green area, the Front Speakers only have a maximum level of -24. At no time to the Front Speakers come close to the green area. Is this ok?

I did not bother with continuing because I was not sure if this would invalidate the results or if in fact it would be ok to run with the speaker volume at this level and Dirac would adjust for it.

I will say that the BP-7006s sound good in the system when using the filter for the previous front speakers (shows you how good my hearing is). However, I would like to get a filter with these speakers in the mix and see how they sound with Dirac.

So, I guess what I need to know is, is it ok to run Dirac when one speaker or a set of speakers have a significantly lower volume level than the others?

Thank you for any help or advice.

Best Regards,

Steve

[Thunderduck]

Thank you very much Keith, I appreciate the added info.

Will not be able to run Dirac until tonight when others have gone to bed and no longer using the tv. I will say that I have noticed that the added "bump" in gain on the speakers has made a noticeable difference. 

Thank you,

Steve

[Thunderduck]

Just finished running Dirac and so far everything seems good. Will not get much critical listening done tonight, but will see how it goes.

Strangely enough, even with the gain control on my front speakers at the 12 o'clock position, the best Output Volume I could get was still -24 on the scale. However, there was no problem running the sweeps and Dirac created the filter without complaint. As I say, it sounds good with a few minutes on non-critical listening but will have more tomorrow.

The mic positions I used were the low were even with the back of my chairs and the high were about 18" above that. So the speakers all definitely had line of sight to the mic.

That's it for now, just wanted to let you know how it went.

Thanks all for the help.

Steve

[nickwin]

Jan 21, 2015 12:26:01 GMT -5 richter250 said:

Jan 10, 2015 14:43:39 GMT -5 melm said:

Sounds better with this last run.. speakers are more precisely balanced. I do note that compared to my non-Dirac settings I really need to sit in the sweet spot to get the best balance and image. I have Maggies with a phantom center and to get dialog right I need to be in my best spot. Not a problem, just an observation...

So better with AGC turned off.

Thanks again Tony.

Have you tried your Maggies with the tweeter on the outside?  I did this and then reran Dirac and the sweet spot is much wider for music and movies.  Dirac does work with Maggies.  Mine have never sounded better.

Glad to hear your liking the results. I will be running Dirac for the first time with my Maggies sometime in the next week or so. I'll post my results when I have them. I run my tweeters on the outside as well.

[wilburthegoose]

Question - since the microphone is attached to the PC, how do you click the measure/proceed button and get your own body out of the picture so it doesn't affect measurements?

Back when I used Audyssey, I could initiate with the AVR's remote control.

Thanks! Looking forward to my first run of Dirac this weekend

[Thunderduck]

I use about a 10 foot usb extension cable that allows me to sit in back of my rear speakers so I am out of the way. My mic is mounted on a stand so with the extension cord I can move it around without any problems.

Regards,

Steve