**miniDSP DDRC-88A Official Thread**8-channel AI/AO Dirac Live in a box - Page 389 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 3874Likes
Reply
 
Thread Tools
post #11641 of 16635 Old 09-18-2016, 08:29 AM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by markus767 View Post


It's not wrong but aiming for -36 will give you 12dB better SNR compared to aiming for -24dB. Better is better.
If you think this is the better approach, you should recommend to MiniDSP that the help file should be changed.
AustinJerry is online now  
Sponsored Links
Advertisement
 
post #11642 of 16635 Old 09-18-2016, 08:33 AM
AVS Forum Addicted Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 11,767
Mentioned: 187 Post(s)
Tagged: 0 Thread(s)
Quoted: 5645 Post(s)
Liked: 2668
I already did.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #11643 of 16635 Old 09-19-2016, 05:12 AM
Advanced Member
 
KevinG's Avatar
 
Join Date: Apr 2002
Location: S. Jersey
Posts: 956
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 370 Post(s)
Liked: 175
I've been following what Marcus is outlining since day one.

What you are doing at that step is making the microphone a sensitive as possible *in your room* so that background noise is at the lowest possible measurement level (as defined by what the GUI will show us). In between speaker squawks, you'd like DIRAC to hear "silence". If you get this setting correct, you should see a perfectly flat line between squawks. Anything that isn't perfectly flat is absolutely "noise" (though I think DIRAC does a damn fine job at discerning what is noise and what isn't...but that's besides the point)
KevinG is offline  
Sponsored Links
Advertisement
 
post #11644 of 16635 Old 09-19-2016, 06:58 AM
AVS Forum Special Member
 
rcohen's Avatar
 
Join Date: Feb 2005
Posts: 3,298
Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 1413 Post(s)
Liked: 652
Using Dirac PC, I switched from a MiniDSP UMIK-1 to a Cross Spectrum EMM-6 with a Behringer Xenyx 302USB mic preamp.
I was surprised to see that the EMM-6 + preamp got a much better SNR than the MiniDSP USB mic.
I forgot exactly how much, but it was something like 10-20 db!

That said, I heard no audible difference with Dirac.
I guess good enough is good enough.
Now I stick with the EMM-6, anyway, due to the better numbers.

(In both cases, I also turned off the AC, pool equipment, and ceiling fan during measurements, to make sure I had perfect silence.)

I don't remember if the DDRC-88A supports different mics.
rcohen is online now  
post #11645 of 16635 Old 09-19-2016, 07:05 AM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by rcohen View Post

I don't remember if the DDRC-88A supports different mics.
AFAIK, it doesn't.
AustinJerry is online now  
post #11646 of 16635 Old 09-19-2016, 03:06 PM
AVS Forum Special Member
 
Stoked21's Avatar
 
Join Date: Sep 2015
Location: Kansas City (Lenexa)
Posts: 2,287
Mentioned: 107 Post(s)
Tagged: 0 Thread(s)
Quoted: 1526 Post(s)
Liked: 932
Fighting the notorious Mac issue.
  • DLCT 1.1.0 build 8244. Latest rev
  • 1.7b1 DDRC-88 Utility Latest rev
  • Latest FW pushed down to each 88A....Can't get into 88Util to confirm revs.
  • 88BM tool works great and connects
  • DLCT as well as DDRC-88 utility Programs won't connect.
  • Running El Capitan 10.11.6

DLCT reports can't connect and no UMIK-1. 88Util reports 3 failed attempts. Both 88A same problem.
Why does 88BM work great, connect, adjust volume, switch slots, etc????

Swear I'm going to have to buy a WinPC to keep the miniDSP in the system....
Attached Thumbnails
Click image for larger version

Name:	Screen Shot 2016-09-19 at 5.09.29 PM.png
Views:	25
Size:	309.0 KB
ID:	1670025   Click image for larger version

Name:	Screen Shot 2016-09-19 at 5.09.18 PM.png
Views:	25
Size:	314.0 KB
ID:	1670033   Click image for larger version

Name:	Screen Shot 2016-09-19 at 5.10.09 PM.png
Views:	20
Size:	73.4 KB
ID:	1670041  

Last edited by Stoked21; 09-19-2016 at 03:10 PM.
Stoked21 is offline  
post #11647 of 16635 Old 09-19-2016, 03:33 PM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by Stoked21 View Post
Fighting the notorious Mac issue.
  • DLCT 1.1.0 build 8244. Latest rev
  • 1.7b1 DDRC-88 Utility Latest rev
  • Latest FW pushed down to each 88A....Can't get into 88Util to confirm revs.
  • 88BM tool works great and connects
  • DLCT as well as DDRC-88 utility Programs won't connect.
  • Running El Capitan 10.11.6

DLCT reports can't connect and no UMIK-1. 88Util reports 3 failed attempts. Both 88A same problem.
Why does 88BM work great, connect, adjust volume, switch slots, etc????

Swear I'm going to have to buy a WinPC to keep the miniDSP in the system....
DDRC-88A Util will no longer connect if you have upgraded to BM. Delete this software now.

I have had several occasions where the DLCT fails to connect. Powering the device off and on has cleared the issue for me. Of course, you can't connect to DLCT if 88BM is running, but you knew that. Also, if you use any other MiniDSP devices like a 2x4, you can't connect to DLCT if a 2x4 plug-in is running.

No other ideas at this time.
AustinJerry is online now  
post #11648 of 16635 Old 09-19-2016, 03:36 PM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by KevinG View Post
I've been following what Marcus is outlining since day one.

What you are doing at that step is making the microphone a sensitive as possible *in your room* so that background noise is at the lowest possible measurement level (as defined by what the GUI will show us). In between speaker squawks, you'd like DIRAC to hear "silence". If you get this setting correct, you should see a perfectly flat line between squawks. Anything that isn't perfectly flat is absolutely "noise" (though I think DIRAC does a damn fine job at discerning what is noise and what isn't...but that's besides the point)
OK, Kevin, which of these looks better to you? And do you see anything that might suggest one might produce a better-sounding result?



AustinJerry is online now  
post #11649 of 16635 Old 09-19-2016, 05:05 PM
Advanced Member
 
KevinG's Avatar
 
Join Date: Apr 2002
Location: S. Jersey
Posts: 956
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 370 Post(s)
Liked: 175
Quote:
Originally Posted by AustinJerry View Post
OK, Kevin, which of these looks better to you? And do you see anything that might suggest one might produce a better-sounding result?



Well, the first one *looks* better (but only after making my browser big enough to see a difference) because the silence between the test sounds is actually represented as silence. (DIRAC is showing that it received no signal/sound between test sounds).

From experience, I can easily say that both will produce the (exact?) same results, because DIRAC seems very good at subtracting out "noise". (It doesn't seem to require a very high signal to noise ratio to figure things out.) That being said, I'd still opt for giving it zero noise if I had the choice, and I do, so I do. :-)
KevinG is offline  
post #11650 of 16635 Old 09-19-2016, 05:20 PM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by KevinG View Post
Well, the first one *looks* better (but only after making my browser big enough to see a difference) because the silence between the test sounds is actually represented as silence. (DIRAC is showing that it received no signal/sound between test sounds).

From experience, I can easily say that both will produce the (exact?) same results, because DIRAC seems very good at subtracting out "noise". (It doesn't seem to require a very high signal to noise ratio to figure things out.) That being said, I'd still opt for giving it zero noise if I had the choice, and I do, so I do. :-)
My question was a trick question. The second screen shot is with the mic gain at -36. The first screen shot is with the mic even lower, say around -50 (can't tell because the indicator only goes down to -36).

My original question to Markus was if -36 was better, why not go even lower? Where do the diminishing returns set in? Or as you say, it may not make an audible difference.

Regardlesss, I can see value in maximizing the signal-to-noise ratio, so no further arguments from me.
ellisr63 likes this.
AustinJerry is online now  
post #11651 of 16635 Old 09-19-2016, 05:46 PM
AVS Forum Special Member
 
Stoked21's Avatar
 
Join Date: Sep 2015
Location: Kansas City (Lenexa)
Posts: 2,287
Mentioned: 107 Post(s)
Tagged: 0 Thread(s)
Quoted: 1526 Post(s)
Liked: 932
Quote:
Originally Posted by AustinJerry View Post
DDRC-88A Util will no longer connect if you have upgraded to BM. Delete this software now.

I have had several occasions where the DLCT fails to connect. Powering the device off and on has cleared the issue for me. Of course, you can't connect to DLCT if 88BM is running, but you knew that. Also, if you use any other MiniDSP devices like a 2x4, you can't connect to DLCT if a 2x4 plug-in is running.

No other ideas at this time.
No 2x4sw running. No BM running. Obviously closed before dlct launch. Even rebooted 3 times and even power cycled each 88a twice. Countless USB swaps (I know USB ports can hang for several seconds). Still no connect to dlct. Retried after all fail connects and BM still connects every time. Will still adjust mv, and change slots....still plays audio no blown fuses. Even rew still works with umik. Clearly Dlct issue with Mac . Never had Mac issues with any minidsp sw before.

Jerry do u use Mac?


Btw. I'm so pissed I just bought the third 88a. I'd rather run 0 and 90 cal files from spectrum in each unit individually. May have to steal wife's win machine to cal at this rate

Marantz NR1607, Crown XLS1500 (2 in bridge mode for mains)
JTR 210RT, JTR Cap S2
Samsung 78” Curved LED 4K (KU7500)
Oppo UDP-203 UHD HDR BD player, AppleTV Gen4, Synology DS216play
STOKED Home Theater Build
Stoked21 is offline  
post #11652 of 16635 Old 09-19-2016, 05:54 PM
AVS Forum Special Member
 
Stoked21's Avatar
 
Join Date: Sep 2015
Location: Kansas City (Lenexa)
Posts: 2,287
Mentioned: 107 Post(s)
Tagged: 0 Thread(s)
Quoted: 1526 Post(s)
Liked: 932
Btw. I have began a starter thread for multi 88a BM with multi sub. It exists with minimal info now. WIP. Once third unit is wired I'll screenshot every BM snapshot, every auto curve and every REW measurement. Will be a long thread so hopefully no one posts over my 15 reserved until I at least get through working install!!!. Once everything is stable house curves. I'll post link soon.

Following @KevinG recommendation. Love Markus input of feedback loop which IMO is best method. But I don't want to kill the extra channel.
Stoked21 is offline  
post #11653 of 16635 Old 09-19-2016, 06:04 PM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by Stoked21 View Post

Jerry do u use Mac?


Btw. I'm so pissed I just bought the third 88a. I'd rather run 0 and 90 cal files from spectrum in each unit individually. May have to steal wife's win machine to cal at this rate
i do not use a Mac, sorry. Time to contact MiniDSP support. Why do you have 3 88A's?
AustinJerry is online now  
post #11654 of 16635 Old 09-19-2016, 06:10 PM
AVS Forum Special Member
 
jpco's Avatar
 
Join Date: May 2004
Location: Fort Collins, CO
Posts: 2,774
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 654 Post(s)
Liked: 303
Quote:
Originally Posted by Stoked21 View Post
Fighting the notorious Mac issue.
  • DLCT 1.1.0 build 8244. Latest rev
  • 1.7b1 DDRC-88 Utility Latest rev
  • Latest FW pushed down to each 88A....Can't get into 88Util to confirm revs.
  • 88BM tool works great and connects
  • DLCT as well as DDRC-88 utility Programs won't connect.
  • Running El Capitan 10.11.6



DLCT reports can't connect and no UMIK-1. 88Util reports 3 failed attempts. Both 88A same problem.

Why does 88BM work great, connect, adjust volume, switch slots, etc????



Swear I'm going to have to buy a WinPC to keep the miniDSP in the system....

I know nothing about 88BM. Running Sierra public beta with DLCT 1.1.0 (8244) and DDRC 1.5 with no issues. Sierra drops tomorrow. Might that help?
jpco is offline  
post #11655 of 16635 Old 09-19-2016, 08:59 PM
AVS Forum Special Member
 
Stoked21's Avatar
 
Join Date: Sep 2015
Location: Kansas City (Lenexa)
Posts: 2,287
Mentioned: 107 Post(s)
Tagged: 0 Thread(s)
Quoted: 1526 Post(s)
Liked: 932
Quote:
Originally Posted by jpco View Post
I know nothing about 88BM. Running Sierra public beta with DLCT 1.1.0 (8244) and DDRC 1.5 with no issues. Sierra drops tomorrow. Might that help?
Guessing that will compound the problem as is any os upgrade.....I know better than to run auto auto FW upgrades though.
Stoked21 is offline  
post #11656 of 16635 Old 09-19-2016, 09:08 PM
AVS Forum Special Member
 
Stoked21's Avatar
 
Join Date: Sep 2015
Location: Kansas City (Lenexa)
Posts: 2,287
Mentioned: 107 Post(s)
Tagged: 0 Thread(s)
Quoted: 1526 Post(s)
Liked: 932
Quote:
Originally Posted by AustinJerry View Post
i do not use a Mac, sorry. Time to contact MiniDSP support. Why do you have 3 88A's?
Really??? Mine goes to 11 dude!!!


Seriously though. There's a 0• and 90• mic strategy. Plus a 6 top strategy. And a 6 surround strategy. Not to mention likely 5 sub strategy. Yea 9.5.6 is goal. 22 amp channels installed. About 30 speaker locations wired. Finally my NF sub is like comparing a multizone massage chair to a dildo rubbing on your foot. I'll likely add more. They cannot be gain matched to the rest of the system. They must be level matched with proximity and intensity of ULF.

Some might call me an immersive freak as a comparison to Bass heads. I can generate images on your shoulders like the angel and devil with speaker invisibility (non localizable). Voices materialize in mid air around the room. If I can't EQ every element separately I lose my imaging control even with LFE.

More in the works. Will let you know the thread when further along.

Marantz NR1607, Crown XLS1500 (2 in bridge mode for mains)
JTR 210RT, JTR Cap S2
Samsung 78” Curved LED 4K (KU7500)
Oppo UDP-203 UHD HDR BD player, AppleTV Gen4, Synology DS216play
STOKED Home Theater Build
Stoked21 is offline  
post #11657 of 16635 Old 09-20-2016, 06:30 AM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by Stoked21 View Post
Really??? Mine goes to 11 dude!!!


Seriously though. There's a 0• and 90• mic strategy. Plus a 6 top strategy. And a 6 surround strategy. Not to mention likely 5 sub strategy. Yea 9.5.6 is goal. 22 amp channels installed. About 30 speaker locations wired. Finally my NF sub is like comparing a multizone massage chair to a dildo rubbing on your foot. I'll likely add more. They cannot be gain matched to the rest of the system. They must be level matched with proximity and intensity of ULF.

Some might call me an immersive freak as a comparison to Bass heads. I can generate images on your shoulders like the angel and devil with speaker invisibility (non localizable). Voices materialize in mid air around the room. If I can't EQ every element separately I lose my imaging control even with LFE.

More in the works. Will let you know the thread when further along.
Uh oh. Sounds like time for rehab...
LarryMagoo and Stoked21 like this.
AustinJerry is online now  
post #11658 of 16635 Old 09-20-2016, 12:38 PM
AVS Forum Special Member
 
rcohen's Avatar
 
Join Date: Feb 2005
Posts: 3,298
Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 1413 Post(s)
Liked: 652
Quote:
Originally Posted by Stoked21 View Post
Really??? Mine goes to 11 dude!!!


Seriously though. There's a 0• and 90• mic strategy. Plus a 6 top strategy. And a 6 surround strategy. Not to mention likely 5 sub strategy. Yea 9.5.6 is goal. 22 amp channels installed. About 30 speaker locations wired. Finally my NF sub is like comparing a multizone massage chair to a dildo rubbing on your foot. I'll likely add more. They cannot be gain matched to the rest of the system. They must be level matched with proximity and intensity of ULF.

Some might call me an immersive freak as a comparison to Bass heads. I can generate images on your shoulders like the angel and devil with speaker invisibility (non localizable). Voices materialize in mid air around the room. If I can't EQ every element separately I lose my imaging control even with LFE.

More in the works. Will let you know the thread when further along.
Are there any best practices when measuring 3D speaker placements with the same mic angle and calibration file? It seems like would be a major problem for timbre matching. 90 degree works great for surround, but what about Atmos?

I suppose the mic angle could be pseudo-randomized at each measurement position, although Dirac would assume that there is more seat to seat variation than there really is.

The ceiling speakers could have their target curves adjusted so that they have a matching 0 degree response, although that would be difficult to do by hand.
rcohen is online now  
post #11659 of 16635 Old 09-20-2016, 08:48 PM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
I have noticed an anomaly with the Dirac delays that I can't explain. Recall that with the bass management plug-in, we can now observe the trims and delays that Dirac is calculating. Here is what I observed:

Trims and delays for the "Auto Targets"



Trims and delays for the same project, using custom targets




Note that in the Auto Targets screen shot, the right front speaker is the "reference speaker", with delay set to zero. The other delays are offsets from the reference speaker, with the left and center speakers showing very small delays, which coincides with the physical placement of the LCR speakers, all 9ft from the MLP.

Now look at the custom targets screen shot. The "reference speaker" has now shifted to the sub, which is 13ft from the MLP. Once again, the delays for the LCR speakers are almost identical, with 3.96-4.06ms delays corresponding to ~4.5ft of delay, which is reasonably close to 13.1ft - 9ft. So far, so good.

But what is strange is why Dirac switched from one reference speaker to another, i.e. if the sub is indeed the furthest speaker, why is it not the reference speaker for the Auto Targets? And why are the trims and delays identical, whether Auto Target or custom target, if it is the same project? And does all of this make any difference? I can't come up with a good explanation.

@markus767 , is this worth asking MiniDSP support?
AustinJerry is online now  
post #11660 of 16635 Old 09-20-2016, 10:49 PM
AVS Forum Special Member
 
rcohen's Avatar
 
Join Date: Feb 2005
Posts: 3,298
Mentioned: 50 Post(s)
Tagged: 0 Thread(s)
Quoted: 1413 Post(s)
Liked: 652
My guess is that they are selecting a reference to minimize the length of the filters. Simple digital delay is much more efficient than longer filters.
rcohen is online now  
post #11661 of 16635 Old 09-21-2016, 05:22 AM
AVS Forum Addicted Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 11,767
Mentioned: 187 Post(s)
Tagged: 0 Thread(s)
Quoted: 5645 Post(s)
Liked: 2668
Quote:
Originally Posted by AustinJerry View Post
@markus767 , is this worth asking MiniDSP support?
No because they will probably tell you that they don't know or can't disclose the inner workings of DL.
The delay is most likely part of DL's phase correction. Different target curve means different filters means different delays.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #11662 of 16635 Old 09-21-2016, 06:29 AM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by markus767 View Post
No because they will probably tell you that they don't know or can't disclose the inner workings of DL.
The delay is most likely part of DL's phase correction. Different target curve means different filters means different delays.
So I assume by your response that it is probably working as designed, and there is no cause for concern. After my original post, I loaded a second project, generated both the auto targets and custom targets, and observed the Dirac delays. The results are consistent with the first project, with almost identical delays.

It is interesting that no one else noticed this. Maybe I have too much spare time on my hands...
DaveyMac and LarryMagoo like this.
AustinJerry is online now  
post #11663 of 16635 Old 09-21-2016, 10:40 AM
AVS Forum Special Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 1,031
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 417 Post(s)
Liked: 204
Quote:
Originally Posted by AustinJerry View Post
I can't come up with a good explanation.
I can't too, except if one of the measurements was wrong at sub channel calculations (due to bad S/N or some noise happened during it's measurement).
IgorZep is offline  
post #11664 of 16635 Old 09-21-2016, 10:47 AM
AVS Forum Special Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 1,031
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 417 Post(s)
Liked: 204
Quote:
Originally Posted by markus767 View Post
No because they will probably tell you that they don't know or can't disclose the inner workings of DL.
They either don't know anything about it and their bass management is completely independent of Dirac (most likely) and then something weird happened, or they know something and integrate with it, then they will be able to tell at least something high level about it without disclosing anything sensitive.

Quote:
Originally Posted by markus767 View Post
The delay is most likely part of DL's phase correction. Different target curve means different filters means different delays.
DL targets minimum phase response, per channel. So, if filters/targets are different, the excess phase part is still always the same, so the same should be also delays.
IgorZep is offline  
post #11665 of 16635 Old 09-21-2016, 10:54 AM
AVS Forum Addicted Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 11,767
Mentioned: 187 Post(s)
Tagged: 0 Thread(s)
Quoted: 5645 Post(s)
Liked: 2668
Quote:
Originally Posted by IgorZep View Post
They either don't know anything about it and their bass management is completely independent of Dirac (most likely) and then something weird happened, or they know something and integrate with it, then they will be able to tell at least something high level about it without disclosing anything sensitive.
BM is implemented separately from DL.

Quote:
Originally Posted by IgorZep View Post
DL targets minimum phase response, per channel. So, if filters/targets are different, the excess phase part is still always the same, so the same should be also delays.
DL also sets delay so there's no lag between speakers at the MLP.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #11666 of 16635 Old 09-21-2016, 11:11 AM
AVS Forum Special Member
 
IgorZep's Avatar
 
Join Date: Sep 2012
Posts: 1,031
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 417 Post(s)
Liked: 204
Quote:
Originally Posted by markus767 View Post
DL also sets delay so there's no lag between speakers at the MLP.
delay_to_set = DELAY_REF - excess_group_delay
Changing target curves doesn't change excess_group_delay as all targets are minimum phase, so it should not change the delay_to_set too.
IgorZep is offline  
post #11667 of 16635 Old 09-22-2016, 08:31 AM
AVS Forum Addicted Member
 
markus767's Avatar
 
Join Date: Apr 2009
Posts: 11,767
Mentioned: 187 Post(s)
Tagged: 0 Thread(s)
Quoted: 5645 Post(s)
Liked: 2668
As said before, DL is a black box. How it's achieving its results is unknown. Obviously delay is introduced in the process which has to be compensated for. When you change target curves, speaker delay applied by DL changes. Nevertheless, in the end the delay between speakers at the MLP is still 0.

Markus

"In science, contrary evidence causes one to question a theory. In religion, contrary evidence causes one to question the evidence." - Floyd Toole
markus767 is offline  
post #11668 of 16635 Old 09-22-2016, 12:53 PM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by markus767 View Post
As said before, DL is a black box. How it's achieving its results is unknown. Obviously delay is introduced in the process which has to be compensated for. When you change target curves, speaker delay applied by DL changes. Nevertheless, in the end the delay between speakers at the MLP is still 0.
I conducted another quick test that seems to confirm your theory about how Dirac behaves. I took my current project and crafted some quick "crazy targets" completely different from my custom targets. After optimizing and storing the results on the 88BM, I opened the BM utility and observed the Dirac delays. Sure enough, they were completely different from my usual custom target delays. BTW, trims stayed consistent, which is expected behavior.

So I think we can conclude that Dirac is indeed adjusting delays depending on the shape of the target curve. While we may not understand the proprietary logic behind how it does it, we can be confident that it is working as designed. This may be a clue towards understanding why Dirac sounds so good to many of us.
LarryMagoo likes this.
AustinJerry is online now  
post #11669 of 16635 Old 09-23-2016, 04:35 AM
Advanced Member
 
KevinG's Avatar
 
Join Date: Apr 2002
Location: S. Jersey
Posts: 956
Mentioned: 11 Post(s)
Tagged: 0 Thread(s)
Quoted: 370 Post(s)
Liked: 175
Quote:
Originally Posted by AustinJerry View Post
So I think we can conclude that Dirac is indeed adjusting delays depending on the shape of the target curve.
This does have implications, though, for anyone that is using the "post-dirac run some other room correction to get delays right for non dirac channels" method, if they happen to use more than one stored filter.

I know some people like to keep a music and a movies setting, for example. If the delays are different between them, the other channels will no longer be "perfectly" aligned for one of those filters. I put quotes around "perfectly" because, in reality, I'd guess that no one could hear the difference anyway.
Svenibaer likes this.
KevinG is offline  
post #11670 of 16635 Old 09-23-2016, 05:41 AM
** Man of Leisure **
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 20,495
Mentioned: 374 Post(s)
Tagged: 1 Thread(s)
Quoted: 11034 Post(s)
Liked: 6577
Quote:
Originally Posted by KevinG View Post
This does have implications, though, for anyone that is using the "post-dirac run some other room correction to get delays right for non dirac channels" method, if they happen to use more than one stored filter.

I know some people like to keep a music and a movies setting, for example. If the delays are different between them, the other channels will no longer be "perfectly" aligned for one of those filters. I put quotes around "perfectly" because, in reality, I'd guess that no one could hear the difference anyway.
That could be a good point, Kevin. It had not occurred to me since I don't have that type of configuration. And, of course, if the user doesn't have the BM plug-in, they may be completely unaware of the changes in delays. Whether or not this issue would result in audible problems or not, knowing the issue exists would cause me some concern.
AustinJerry is online now  
Sponsored Links
Advertisement
 
Reply Receivers, Amps, and Processors

Tags
DDRC-88A , dirac live , minidsp , req

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off