Optimizing subwoofers and integration with mains: multi sub optimizer - Page 47 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 494Likes
Reply
 
Thread Tools
post #1381 of 1682 Old 02-12-2019, 11:03 AM
AVS Forum Special Member
 
Nalleh's Avatar
 
Join Date: Aug 2014
Location: Norway
Posts: 3,453
Mentioned: 274 Post(s)
Tagged: 0 Thread(s)
Quoted: 2404 Post(s)
Liked: 3134
Quote:
Originally Posted by andyc56 View Post
Here's a sample biquad text file. I put it in zip format so that it stays binary, to avoid hidden end-of-line conversions that might happen in some file operations.
It worked fine

So that means it should work doing the MSO work on a PC, and the rest on Mac, right?
Attached Thumbnails
Click image for larger version

Name:	BF779B18-3DD2-4BB3-8939-A15AA13B8FC3.jpg
Views:	31
Size:	167.5 KB
ID:	2524744  

Dual Atmos Receivers - Atmos 13.1.8/DTS X 9.1.8/Auro 3D 13.1 - Denon AVCX8500H+AVRX7200WA - Klipsch+KEF - 6xSI18" - 8xJBL 12" BOSS - 4xJBL 12" w/SLAPS M12" VNF - 3x2 stacked Crowson MA - 4xBK-LFE - 6xNU6K(fan&trig mod) - Minidsp 10x10HD - Oppo UDP203 - XBox OneX - Apple TV4K - JVC RS600 Dreamscreen V2 120"- Philips 65OLED873.
Nalleh’s HT
Nalleh is online now  
Sponsored Links
Advertisement
 
post #1382 of 1682 Old 02-12-2019, 11:31 AM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by Nalleh View Post
It worked fine

So that means it should work doing the MSO work on a PC, and the rest on Mac, right?
Yes.
andyc56 is online now  
post #1383 of 1682 Old 02-12-2019, 11:38 AM
AVS Forum Special Member
 
Nalleh's Avatar
 
Join Date: Aug 2014
Location: Norway
Posts: 3,453
Mentioned: 274 Post(s)
Tagged: 0 Thread(s)
Quoted: 2404 Post(s)
Liked: 3134
Quote:
Originally Posted by andyc56 View Post
Yes.
Awsome. I’ll see if i can get a start on this in the next couple of weeks. Seems daunting at first, but curious what it can do

Dual Atmos Receivers - Atmos 13.1.8/DTS X 9.1.8/Auro 3D 13.1 - Denon AVCX8500H+AVRX7200WA - Klipsch+KEF - 6xSI18" - 8xJBL 12" BOSS - 4xJBL 12" w/SLAPS M12" VNF - 3x2 stacked Crowson MA - 4xBK-LFE - 6xNU6K(fan&trig mod) - Minidsp 10x10HD - Oppo UDP203 - XBox OneX - Apple TV4K - JVC RS600 Dreamscreen V2 120"- Philips 65OLED873.
Nalleh’s HT
Nalleh is online now  
Sponsored Links
Advertisement
 
post #1384 of 1682 Old 02-18-2019, 01:18 PM
AVS Forum Special Member
 
Join Date: Nov 2016
Posts: 1,141
Mentioned: 36 Post(s)
Tagged: 0 Thread(s)
Quoted: 941 Post(s)
Liked: 275
Dear MSO experts, I have a question. I have 2 subs and I am optimizing only MLP so basically 1 position only. For that, I added Global EQs and got what I needed. Perfect. Worked exactly as I expected. The issue though is that I came across this BEQ thread https://www.avsforum.com/forum/113-s...ed-movies.html. It requires to enter EQ on Global. Since I already have Global EQ from MSO, I can't apply BEQ eq as global. Obviously one option is to use individual EQ on subs but I didn't like it. MSO, boosted one sub and cut another to get flat FR and that resulted in one sub playing Low frequencies a lot more and other Midbass.

I there anyway that I can get MSO to equalize both subs using same EQ without using Global filters? Here is my MSO project.

https://www.dropbox.com/s/pdwhfmp5j5...urve.msop?dl=0

Thx.
harrisu is offline  
post #1385 of 1682 Old 02-19-2019, 09:43 AM
AVS Forum Special Member
 
richardsim7's Avatar
 
Join Date: Apr 2015
Posts: 2,039
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1312 Post(s)
Liked: 1682
So I need some advice, or rather, affirmation that what I'm doing isn't dumb


This is my 2 subs with no EQ combined


And this is what MSO is proposing to get them as flat as possible
Problem is, here's the EQ


As you can see, this pulls all the power out of the rear sub at 10Hz, which is kinda where I don't want to lose any power if I can avoid it


So I started tinkering, and came up with this: 1
Which leaves the natural power to 30Hz then starts correcting, as you can see from the propsed EQ


I figured I may as well push this slightly further and came up with this: 2 (EQ)


I tried going a little further and added a target curve (not sure it worked) and ended up with this, but not entirely convinced about this one: 3 (EQ)


This is my Dirac curve - which I modified using the 6dB Harman Curve from here



I figured there's no point cutting the subs down then boosting them back up in Dirac, I may as well try and retain as much of the house curve and then adjut from there, right?


So which of these should I go for? Or should I try something else?


Imgur album here, if it's easier: https://imgur.com/a/pjLxeEI
Let me know if you want me to post any MSO projects
pixphipau and andyc56 like this.

Last edited by richardsim7; 02-19-2019 at 11:55 AM.
richardsim7 is online now  
post #1386 of 1682 Old 02-19-2019, 10:11 AM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by richardsim7 View Post
So I need some advice, or rather, affirmation that what I'm doing isn't dumb
Not dumb at all! The problem of automating the maximization of output while getting flat response at multiple listening positions is even harder than just doing the latter by itself.

I've been trying to come up with a way of doing this using using a metric I call "pseudo-sensitivity". This is a generalization of the idea of sensitivity to a multiple-input, multiple-output system, where each "input" is a power amp input (the output of the DSP), and each "output" is a listening position. The worry is that this might make the optimization so slow as to be unusable. But I digress.

Quote:
Originally Posted by richardsim7 View Post
I figured there's no point cutting the subs down then boosting them back up in Dirac, I may as well try and retain as much of the house curve and then adjust from there, right?

So which of these should I go for? Or should I try something else?
Both of these results are exceptional. If I were to venture a guess (and that's all it is), I'd say maybe the first one with equal levels to the two subs at 10 Hz (assuming identical subs and power amps).
pixphipau and richardsim7 like this.
andyc56 is online now  
post #1387 of 1682 Old 02-19-2019, 10:26 AM
AVS Forum Special Member
 
richardsim7's Avatar
 
Join Date: Apr 2015
Posts: 2,039
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1312 Post(s)
Liked: 1682
Quote:
Originally Posted by andyc56 View Post
Not dumb at all! The problem of automating the maximization of output while getting flat response at multiple listening positions is even harder than just doing the latter by itself.

I've been trying to come up with a way of doing this using using a metric I call "pseudo-sensitivity". This is a generalization of the idea of sensitivity to a multiple-input, multiple-output system, where each "input" is a power amp input (the output of the DSP), and each "output" is a listening position. The worry is that this might make the optimization so slow as to be unusable. But I digress.



Both of these results are exceptional. If I were to venture a guess (and that's all it is), I'd say maybe the first one with equal levels to the two subs at 10 Hz (assuming identical subs and power amps).

Ahh, the magician himself! I must thank you for your wonderful program you made


Yes, they're identical subs and amps in a tiny room - only aiming for one listening position (because I'm selfish )


I've edited my post with numbers to make it easier, are you saying to go for solution 1, or 2? I was thinking 2 looked the best
andyc56 likes this.

Last edited by richardsim7; 02-19-2019 at 10:33 AM.
richardsim7 is online now  
post #1388 of 1682 Old 02-19-2019, 10:56 AM
Member
 
pixphipau's Avatar
 
Join Date: May 2009
Posts: 128
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 74 Post(s)
Liked: 36
Quote:
Originally Posted by richardsim7 View Post
Ahh, the magician himself! I must thank you for your wonderful program you made


Yes, they're identical subs and amps in a tiny room - only aiming for one listening position (because I'm selfish )


I've edited my post with numbers to make it easier, are you saying to go for solution 1, or 2? I was thinking 2 looked the best
Wow, fascinating, thank you. Makes it easy to understand what you’re looking at doing.
pixphipau is online now  
post #1389 of 1682 Old 02-19-2019, 12:24 PM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by richardsim7 View Post
I've edited my post with numbers to make it easier, are you saying to go for solution 1, or 2? I was thinking 2 looked the best
That would be number 2 per your revised post (the one with equal sub levels at 10 Hz), but 3 looks good too.
pixphipau and richardsim7 like this.
andyc56 is online now  
post #1390 of 1682 Old 02-19-2019, 12:34 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 26,907
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 1460 Post(s)
Liked: 891
Quote:
Originally Posted by andyc56 View Post
That would be number 2 per your revised post (the one with equal sub levels at 10 Hz), but 3 looks good too.
Holy cow! After reading, and maybe hearing it from you as well, that MSO brings little improvement to only two subs I see @richardsim7 's results and am blown away. I had used the "little improvement with only two subs" as a rationale to not take the time to try it. Even more embarrassing is that I am even already using a MiniDSP 2x4 Balanced, have REW on a laptop and a calibrated mic, so deploying MSO would involve an investment of only my time.

I know MSO's "proposal" is theoretical, so I have to ask how much of that improvement will actually be realized in his room?

Jeff
pixphipau likes this.

"The future is already here, it's just not evenly distributed." W. Gibson

"I like the future, I'm in it." F. Theater
pepar is offline  
post #1391 of 1682 Old 02-19-2019, 12:39 PM
AVS Forum Special Member
 
richardsim7's Avatar
 
Join Date: Apr 2015
Posts: 2,039
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1312 Post(s)
Liked: 1682
Quote:
Originally Posted by pepar View Post
Holy cow! After reading, and maybe hearing it from you as well, that MSO brings little improvement to only two subs I see @richardsim7 's results and am blown away. I had used the "little improvement with only two subs" as a rationale to not take the time to try it. Even more embarrassing is that I am even already using a MiniDSP 2x4 Balanced, have REW on a laptop and a calibrated mic, so deploying MSO would involve an investment of only my time.

I know MSO's "proposal" is theoretical, so I have to ask how much of that improvement will actually be realized in his room?

Jeff

Keep in mind I was only aiming for one listening position, rather than flat across multiple, but will apply both the flat results, and the 2nd curve and take some real-world measurements and let you know
pixphipau likes this.
richardsim7 is online now  
post #1392 of 1682 Old 02-19-2019, 12:42 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 26,907
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 1460 Post(s)
Liked: 891
Follow up here ... would anyone know of a piece of software ... MSO perhaps? ... that would assist me in checking for the best locations along my two subs' respective walls for said subs? IOW, one sub is on the front wall and the other on the rear, and I have some space to move them laterally if it would improve the raw room response. "If" ... is where I need some help. Further complicating the issue, in a misguided attempt 29 years ago to eliminate modes, I skewed the walls by one degree.

Jeff

"The future is already here, it's just not evenly distributed." W. Gibson

"I like the future, I'm in it." F. Theater
pepar is offline  
post #1393 of 1682 Old 02-19-2019, 12:44 PM
AVS Forum Special Member
 
richardsim7's Avatar
 
Join Date: Apr 2015
Posts: 2,039
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1312 Post(s)
Liked: 1682
Quote:
Originally Posted by pepar View Post
Follow up here ... would anyone know of a piece of software ... MSO perhaps? ... that would assist me in checking for the best locations along my two subs' respective walls for said subs? IOW, one sub is on the front wall and the other on the rear, and I have some space to move them laterally if it would improve the raw room response. "If" ... is where I need some help. Further complicating the issue, in a misguided attempt 29 years ago to eliminate modes, I skewed the walls by one degree.

Jeff

If you measure each sub in each position (with a timing reference) you can use REW to combine responses using 'trace arithmetic' and figure out the best combination that way?
andyc56 likes this.
richardsim7 is online now  
post #1394 of 1682 Old 02-19-2019, 12:45 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 26,907
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 1460 Post(s)
Liked: 891
Quote:
Originally Posted by richardsim7 View Post
Keep in mind I was only aiming for one listening position, rather than flat across multiple, but will apply both the flat results, and the 2nd curve and take some real-world measurements and let you know
I have single a seat that 95% of the time holds the only theater user, me. I'd like to optimize for the two seats just left and right of me in the same row if I could. The three seats in the front rows only ever hold non-audiophile spouses.

Follow-up: How much of a difference does 2' to the left and 2' to the right (of my MLP) make in sub response?

Jeff

"The future is already here, it's just not evenly distributed." W. Gibson

"I like the future, I'm in it." F. Theater
pepar is offline  
post #1395 of 1682 Old 02-19-2019, 12:49 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 26,907
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 1460 Post(s)
Liked: 891
Quote:
Originally Posted by richardsim7 View Post
If you measure each sub in each position (with a timing reference) you can use REW to combine responses using 'trace arithmetic' and figure out the best combination that way?
So, I'd measure the two subs individually for their current locations. Then move them, repeat the measurements and then compare?

"The future is already here, it's just not evenly distributed." W. Gibson

"I like the future, I'm in it." F. Theater
pepar is offline  
post #1396 of 1682 Old 02-19-2019, 12:53 PM
AVS Forum Special Member
 
richardsim7's Avatar
 
Join Date: Apr 2015
Posts: 2,039
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1312 Post(s)
Liked: 1682
Quote:
Originally Posted by pepar View Post
I have single a seat that 95% of the time holds the only theater user, me. I'd like to optimize for the two seats just left and right of me in the same row if I could. The three seats in the front rows only ever hold non-audiophile spouses.
You can prioritize the MLP by "weighting" it against the other positions in MSO so it will get the MLP as flat as possible whilst making the others better than they were before



Quote:
Originally Posted by pepar View Post
Follow-up: How much of a difference does 2' to the left and 2' to the right (of my MLP) make in sub response?

Probably a lot, but you've got the mic + REW, only you can confirm it


Quote:
Originally Posted by pepar View Post
So, I'd measure the two subs individually for their current locations. Then move them, repeat the measurements and then compare?

Yup, and you can combine different measurements to find the best combination in REW
richardsim7 is online now  
post #1397 of 1682 Old 02-19-2019, 01:05 PM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by pepar View Post
Holy cow! After reading, and maybe hearing it from you as well, that MSO brings little improvement to only two subs I see @richardsim7 's results and am blown away. I had used the "little improvement with only two subs" as a rationale to not take the time to try it. Even more embarrassing is that I am even already using a MiniDSP 2x4 Balanced, have REW on a laptop and a calibrated mic, so deploying MSO would involve an investment of only my time.
When I wrote the original documentation about using two subs, I was thinking more about the possibility of improvement in the seat-to-seat variation of frequency response than anything else. Also, I hadn't tried using only two subs, nor had I seen data from anyone who had done so. Later, genesplitter did a nice writeup of his experience with two subs, and I incorporated his results into the documentation.

Quote:
Originally Posted by pepar View Post
I know MSO's "proposal" is theoretical, so I have to ask how much of that improvement will actually be realized in his room?
I'm not completely sure I understand the question. If it's about calculated results vs measured (after applying the calculated filter parameters, attenuations and delays), there's really only two assumptions made by MSO:
  1. That the electro-acoustical system is linear, meaning that it has negligible distortion
  2. That the MSO filter responses, in both magnitude and phase, accurately represent the actual filter responses in the DSP

These are essentially the same assumptions made by e.g. REW, except that in REW the phase responses of the filters don't enter into its calculations. The fact that MSO's accuracy also depends on the phase accuracy of the filters makes MSO more sensitive to filter inaccuracies than REW.

At very low frequencies, the miniDSP 2x4 non-HD versions have some problems with actual response vs. theoretical, because they use fixed-point DSP processing. This is demonstrated using the 4x10 here (scroll down to the graph that says "miniDSP 4x10 HD low frequency filter stability on 48 kHz plugin"). The HD versions of the 2x4 use floating-point DSP processing and are said to be far more accurate at low frequencies, but I have not seen actual data supporting the claim (which makes complete sense according to the theory). The "HD" designation of the 4x10 is confusing, because it uses a fixed-point DSP, rather than the floating-point DSP of the 2x4 HD.
richardsim7 likes this.
andyc56 is online now  
post #1398 of 1682 Old 02-19-2019, 01:31 PM
AVS Forum Special Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 3,826
Mentioned: 243 Post(s)
Tagged: 0 Thread(s)
Quoted: 2298 Post(s)
Liked: 2391
@richardsim7 is it literally a single mic position or a tight cluster of readings? If the former I would be inclined to look at the sensitivity of the result to the mic position. I imagine this simplest way to do this is measure at those positions with that EQ in place.

Ultimately how does it sound and feel?
andyc56 likes this.
3ll3d00d is online now  
post #1399 of 1682 Old 02-19-2019, 01:49 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: On the beach in Quintana Roo
Posts: 26,907
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 1460 Post(s)
Liked: 891
@andyc56 , I needed 2x4 balanced for the run to the amp rack in front from the "head end" in the rear.

I will revisit your documentation and it appears that I need to re-refresh myself with REW wrt it's use in conjunction with MSO.

1. The six boundaries of my room flex; that is considered "distortion", right?

Anyway, enough procrastination!

"The future is already here, it's just not evenly distributed." W. Gibson

"I like the future, I'm in it." F. Theater
pepar is offline  
post #1400 of 1682 Old 02-19-2019, 02:23 PM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by pepar View Post
1. The six boundaries of my room flex; that is considered "distortion", right?
If you keep the SPL levels reasonable, you'll be fine. Also, REW's log sweep technique for computing frequency response acts to suppress the effects of distortion on the frequency response measurement. This isn't true of other, older techniques for doing frequency response measurements, such as MLSS.
andyc56 is online now  
post #1401 of 1682 Old 02-21-2019, 11:37 AM
AVS Forum Special Member
 
cdy2179's Avatar
 
Join Date: Dec 2005
Location: central Louisiana
Posts: 3,340
Mentioned: 38 Post(s)
Tagged: 0 Thread(s)
Quoted: 865 Post(s)
Liked: 642
Quote:
Originally Posted by andyc56 View Post
The HD versions of the 2x4 use floating-point DSP processing and are said to be far more accurate at low frequencies, but I have not seen actual data supporting the claim (which makes complete sense according to the theory). The "HD" designation of the 4x10 is confusing, because it uses a fixed-point DSP, rather than the floating-point DSP of the 2x4 HD.
The 4x10 HD is crippled with 7 ms of delay just as the regular and balanced 2x4 units are. The 2x4 HD is the one you want for delay and far better eq resolution.

When loading eq files on the standard 2x4 the eq isn't as tight to the target and always need massaging afterwards. The HD 2x4 will suck the response to the target often +/- 1 db and need absolutely no touch up. It's basically auto EQ.

Here's 3 seats with the standard 2x4 after EQ file and lots of massaging.



And here's 2 seats with an HD 2x4 and only the EQ file (house curve). No massaging at all. This is everytime if you know how to setup the target. EQ with the HD takes less than 2 minutes and you're done.


All measurements shown with no smoothing. Seat to seat consistency acquired by proper alignment via good placement and delay.
Attached Thumbnails
Click image for larger version

Name:	House Curve.jpg
Views:	280
Size:	1.14 MB
ID:	2529168   Click image for larger version

Name:	3 seats back.jpg
Views:	205
Size:	45.9 KB
ID:	2529170  

Last edited by cdy2179; 02-21-2019 at 11:46 AM.
cdy2179 is offline  
post #1402 of 1682 Old 02-25-2019, 12:34 PM
AVS Forum Special Member
 
jcmccorm's Avatar
 
Join Date: Mar 2001
Location: Madison, AL, USA
Posts: 4,492
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 597 Post(s)
Liked: 444
I've got a question about filters. Please excuse me if this has been answered. A search didn't show what I was looking for.

There's a long list of filters that MSO provides that are compatible with the NU6000D, however, only 8 can be used at once in the 6000D.

I assume that MSO will utilize every filter that I add to the config panel for each subwoofer. How do I reconcile the case where it uses all available filter types which exceeds the number of filters that the 6000D can actually implement?

Furthermore, in the 6000D, all 8 filters could be PEQ or all 8 can be LS12. DO I add all of those into MSO?

Thanks!
jcmccorm is offline  
post #1403 of 1682 Old 02-25-2019, 01:32 PM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by jcmccorm View Post
I've got a question about filters. Please excuse me if this has been answered. A search didn't show what I was looking for.

There's a long list of filters that MSO provides that are compatible with the NU6000D, however, only 8 can be used at once in the 6000D.

I assume that MSO will utilize every filter that I add to the config panel for each subwoofer. How do I reconcile the case where it uses all available filter types which exceeds the number of filters that the 6000D can actually implement?

Furthermore, in the 6000D, all 8 filters could be PEQ or all 8 can be LS12. DO I add all of those into MSO?
It's been a while since I've looked at the Behringer offerings, and I see that they have both the original NU6000DSP and a newer NX6000D now. These use different software, so I downloaded the software for the newer NX and tried it out. It looks like there's a limit of 8 filters per channel. Are you saying that there is a grand total limit of 8 specified somewhere? I couldn't find any reference to this. The documentation seems pretty sparse.

Assuming the limit is actually 8 per channel, then you would have a maximum of 8 PEQs per channel in MSO if your target is a flat response. If the target has a boosted LF response, you'd need to either use a room correction product like Dirac that supports that or reduce the number of PEQs used in the 6000 and use either a single LS6 or LS12 per channel to achieve the desired LF boost.

In MSO, you generally start out with a smaller number of PEQs per channel (say 3 or 4), because the optimization time becomes quite long when lots of filters are used.

If the limit is 8 total, then you'd use 4 PEQs per channel for a flat target, or 3 PEQs plus one LF shelf per channel for an LF boost if the room correction software doesn't support LF target curves.
andyc56 is online now  
post #1404 of 1682 Old 02-25-2019, 01:54 PM
AVS Forum Special Member
 
jcmccorm's Avatar
 
Join Date: Mar 2001
Location: Madison, AL, USA
Posts: 4,492
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 597 Post(s)
Liked: 444
Quote:
Originally Posted by andyc56 View Post
It's been a while since I've looked at the Behringer offerings, and I see that they have both the original NU6000DSP and a newer NX6000D now. These use different software, so I downloaded the software for the newer NX and tried it out. It looks like there's a limit of 8 filters per channel. Are you saying that there is a grand total limit of 8 specified somewhere? I couldn't find any reference to this. The documentation seems pretty sparse.

Assuming the limit is actually 8 per channel, then you would have a maximum of 8 PEQs per channel in MSO if your target is a flat response. If the target has a boosted LF response, you'd need to either use a room correction product like Dirac that supports that or reduce the number of PEQs used in the 6000 and use either a single LS6 or LS12 per channel to achieve the desired LF boost.

In MSO, you generally start out with a smaller number of PEQs per channel (say 3 or 4), because the optimization time becomes quite long when lots of filters are used.

If the limit is 8 total, then you'd use 4 PEQs per channel for a flat target, or 3 PEQs plus one LF shelf per channel for an LF boost if the room correction software doesn't support LF target curves.
Thank you very much!

I'm using the older 6000DSP but I beilieve the FW is the same (just a new chassis).

It has a total of 8 filters available (independent of HP, LP, and polarity). Each of these filters (8) can be either LS6, LS12, HS6, HS12, or PEQ.

So if may paraphrase, you're saying that I should choose the filters before optimization that would best achieve my goal. My target curve does have a boosted LF so I will choose an LS12 along with 7 PEQs for a total of 8 filters.

The LP and HP are independent so I can add those as well.

I added the polarity filter (since I can set it in the 6000DSP to 0 or 180 degrees) but it appears that when adding it, MSO reverses the polarity rather than using it in the mix of variables (I thought adding the polarity would perhaps give MSO the ability to deal with limited delay options). Anyway, I took out the polarity filter.

Allow me to thank you for this tool. I've been working on my two sub "groups" and mains integration on and off for weeks. I got to what I believe to be a good place but I just recently decided to give MSO a try. I'm excited about what this program is doing and anxious to see the results. What an enormous time saver this will be and I can't thank you enough.
jcmccorm is offline  
post #1405 of 1682 Old 02-25-2019, 02:16 PM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by jcmccorm View Post
The LP and HP are independent so I can add those as well.
You generally don't want to use a sub LPF in MSO at all unless you're doing a less common arrangement like the Geddes approach, which does not use a conventional crossover. Ordinarily, you'd just use the sub LPF and mains HPF that already make up your AVR's crossover network, with no corresponding filter blocks added in MSO at all (because their effect is already present in the measured data).

You would use an HPF for the sub only if you had, say, a DIY ported sub that needs protection from signals whose frequency is below the box tuning frequency.

The polarity inverters in MSO are fixed because (unfortunately) if the optimizer were allowed to change their state it would play havoc with convergence.
andyc56 is online now  
post #1406 of 1682 Old 02-25-2019, 03:03 PM
AVS Forum Special Member
 
jcmccorm's Avatar
 
Join Date: Mar 2001
Location: Madison, AL, USA
Posts: 4,492
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 597 Post(s)
Liked: 444
Quote:
Originally Posted by andyc56 View Post
You generally don't want to use a sub LPF in MSO at all unless you're doing a less common arrangement like the Geddes approach, which does not use a conventional crossover. Ordinarily, you'd just use the sub LPF and mains HPF that already make up your AVR's crossover network, with no corresponding filter blocks added in MSO at all (because their effect is already present in the measured data).

You would use an HPF for the sub only if you had, say, a DIY ported sub that needs protection from signals whose frequency is below the box tuning frequency.

The polarity inverters in MSO are fixed because (unfortunately) if the optimizer were allowed to change their state it would play havoc with convergence.
Thank you again! I had just deleted the HPF and LPF filters after checking my results and before reading your reply. Some of the PEQs were working hard and I assumed that the LPF and HPF were being used and needed to be overcome.

My subs are sealed (DIY) so no need for the HPF.

I have yet to try out the filters on my system (this is my first time to use MSO) and I'm learning that the magic (on my end at least) is the filters that you allow MSO to use.

I understand the polarity. Toggling the polarity would really reset things drastically. I imagine that I could just do two optimization runs; one with 0 degrees and one with 180 degrees for the polarity and compare the results.
jcmccorm is offline  
post #1407 of 1682 Old 02-25-2019, 03:29 PM
AVS Forum Special Member
 
Join Date: Feb 2013
Posts: 1,514
Mentioned: 34 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 662
Quote:
Originally Posted by jcmccorm View Post
I understand the polarity. Toggling the polarity would really reset things drastically. I imagine that I could just do two optimization runs; one with 0 degrees and one with 180 degrees for the polarity and compare the results.
You can use the "configurations" feature for this. You'd clone the original configuration, duplicating all graphs, then add the inversion to the new configuration. That way you can compare results between the two configurations.

Also, you'll probably want the allowed minimum value for the center frequency of the PEQs to be 20 Hz. It's currently 40 Hz by default.
andyc56 is online now  
post #1408 of 1682 Old 02-25-2019, 06:22 PM
AVS Forum Special Member
 
jcmccorm's Avatar
 
Join Date: Mar 2001
Location: Madison, AL, USA
Posts: 4,492
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 597 Post(s)
Liked: 444
Quote:
Originally Posted by andyc56 View Post
You can use the "configurations" feature for this. You'd clone the original configuration, duplicating all graphs, then add the inversion to the new configuration. That way you can compare results between the two configurations.

Also, you'll probably want the allowed minimum value for the center frequency of the PEQs to be 20 Hz. It's currently 40 Hz by default.
Thank you again! I did change the minimum frequency to 20Hz in Tools->Application Options so that whenever I drop a new PEQ filter in, the default is what I need. Nice to have that

I had some success and failure.

First off, here's where I ended up with manual tweaking the subs together and then adjusting distances to integrate with the mains (center channel in this case). Notice the dip near 200Hz. I don't have it well integrated yet and tweaking the distance for both sub groups (together) manually just shifted the dip and made it deeper. If I were doing this manually, I need to go back and work on just the two sub groups together again before trying to integrate with the mains.



Here's what I ended up with using MSO. Wow. I limited MSO to an LS, HS, and 3 PEQ's to work with. Looking at the resulting filters I realized that there was no way, ever, that I'd come up with that on my own. I was really curious what it would look like. To my delight, it looks great!



The subs are blended nicely with each other and they are integrated with the mains perfectly. MSO followed my target curve.

There is a problem with this however that I discovered right away. I was excited about the results and anxious to listen to something familiar and I chose a concert video. This problem is unique to my setup though; my first sub group is up front against the wall, while my second group is right behind the seating. In my manual tweaking, I had purposely kept the volume of the near-field subs low with respect to the front subs. MSO had no such constraints so now I felt every pluck of a bass string. I felt it so much that I couldn't really hear it. It was too much. Luckily, I'm sure I can constrain MSO from adding any gain to that near-field sub group.

Anyway, overall, I'd call it a glorious success. I've just got to work on some constraints for MSO that are particular to my sub positioning and perhaps tame that target curve a bit.

Thank you again for your help @andyc56 . This is a great tool that worked like a champ for me.
Attached Thumbnails
Click image for larger version

Name:	Manual.jpg
Views:	168
Size:	101.0 KB
ID:	2531132   Click image for larger version

Name:	MSO.jpg
Views:	170
Size:	95.2 KB
ID:	2531134  
andyc56 likes this.
jcmccorm is offline  
post #1409 of 1682 Old 03-01-2019, 01:29 PM
Advanced Member
 
AdamAttewell's Avatar
 
Join Date: Sep 2014
Location: UK
Posts: 835
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 652 Post(s)
Liked: 135
Hello all,

I am currently trying to use MSO with my 2x4HD to improve the response over my four seating locations spread over two rows.

I have taken over 300 measurements & have finally settled on the positions for my two SVS PB13's.

This is the response I have over my four seating locations without any EQ or crossovers active.






I ran MSO overnight & got this result.





I dont know what I have done wrong but the result seems worse after MSO.

In case I have missed any information I have uploaded my MSO project to my Dropbox here if anyone wants to take a look:

https://www.dropbox.com/s/tc4b6i5ig5...3.19.msop?dl=0

If anyone can point me in the right direction I would really appreciate it.

Thanks Adam

Last edited by AdamAttewell; 03-01-2019 at 01:55 PM.
AdamAttewell is online now  
post #1410 of 1682 Old 03-03-2019, 05:46 AM
Advanced Member
 
AV_mike's Avatar
 
Join Date: May 2009
Posts: 591
Mentioned: 16 Post(s)
Tagged: 0 Thread(s)
Quoted: 237 Post(s)
Liked: 184
Quote:
Originally Posted by AdamAttewell View Post
Hello all,

I am currently trying to use MSO with my 2x4HD to improve the response over my four seating locations spread over two rows.

I have taken over 300 measurements & have finally settled on the positions for my two SVS PB13's.

This is the response I have over my four seating locations without any EQ or crossovers active.

I ran MSO overnight & got this result.

I dont know what I have done wrong but the result seems worse after MSO.

In case I have missed any information I have uploaded my MSO project to my Dropbox here if anyone wants to take a look:

https://www.dropbox.com/s/tc4b6i5ig5...3.19.msop?dl=0

If anyone can point me in the right direction I would really appreciate it.

Thanks Adam
Could you upload the "optimized" project - after the over-night run?

Also, do compare traces using the same vertical scale spacing.

Regards, Mike.
AdamAttewell likes this.

System: Pioneer KRP-500M, Lumagen Mini3D, Denon AVR-4520 (custom modified to allow use of >>>), miniDSP DDRC88BM, Oppo 103EU, Sky+HD DRX895, Humax HDR-FOX-T2 (x2), Apple ATV3
Bowers & Wilkins CM8 (left & right) - CMC2 (centre) - Kef HTS3001SE Surrounds, miniDSP 2x4HD, SVS SB13 Ultra, SVS SB2000 (x2) <<< Perfectly blended by MSO
AV_mike is offline  
Sponsored Links
Advertisement
 
Reply Subwoofers, Bass, and Transducers

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