If you've really read the manual then you probably didn't understand it. What exactly are you struggling with? You measure each sub full range at multiple positions (with a common timing reference), import those measurements into MSO, optimize the response in MSO and export the settings to your miniDSP 2x4.

I am a newbie to this, so I need some help understanding this procedure. That is correct.
If I understand you and the manual correct, I will use the the left front channel as a timing reference. The subs is to be measured full range. No adjustments shall be done in the 2x4 before running MSO. Routing will of course be set. Subs measured as described in the manual in several positions, both all on and seperately. 4 subs is at least measurements in 4 difference positions. Filters generated in MSO to be applied in EQ of each individual sub. These filters then will correct both level and timing on each sub when applied?
In my case, no MSO filters to be applied in the DDRC88BM?
Maybe stupid questions for you pro's, but for a newbie these things mat not that obvious.

Here's a short summary of the process: https://www.minidsp.com/applications...-sub-optimizer

Helpful threads:
https://www.avsforum.com/forum/91-aud...et-graphs.html
https://www.avsforum.com/forum/90-rec...-live-box.html

I have a copy of the short summary already.
But it basically link the MSO to the DDRC88. My questions is about the 2x4HD in combination with the MSO. Anyone who can give me an "how-to" answer regarding the combination MSO/2x4HD? Let us imagine that I do not have a ddrc88. Only a minidsp 2x4HD. Can the MSO be used on the 2x4HD? Can anyone try to give me a short explanation in their own words how I shall do this? Can the procedure described for ddrc88/MSO be directly transformed to the combination 2x4HD/MSO?
I obiously do not understand the manual good enough, so maybe it will be easier if anyone could try to give me a short explenation instead of linking to various manuals?

https://www.minidsp.com/applications...-sub-optimizer

Steps 10 & 11 look to be the exact same as just about any of the MiniDSP devices as they use very similar core software. So it would work the same as the DDRC88. Also, look at Appendix A (Appendix A. Notes on using an external 2x4 for the subs) as it discusses using an external 2x4 if you need more channels for subs.

Everything seems to be there but at some point you need to start trying it for you to truly understand it. Give it a shot and then if you have troubles, ask specific questions. But really, I think you've looked at the manuals enough and should start trying it as everything should become clear once you start actually doing it.

Can I just quickly double check as I thought I would peruse the minidsp instructions (not that I have one). There's no mention of including one main channel reading. Is that no longer a requirement with MSO. Apologies if I missed it on the thread.

At one point Andy added "sub only" configurations to MSO. No measurements of the sats required.

Here is the MSO documentation, which is quite detailed and feature-packed.
http://andyc.diy-audio-engineering.org/mso/html/

MSO originally focused on sub integration with mains (L/R or C) but after several requests Andy added Sub-only optimization. This page on optimization strategies speaks to this.
http://andyc.diy-audio-engineering.o...trategies.html

And this page talks about taking the measurements for L/R, C, or sub-only optimizations.
http://andyc.diy-audio-engineering.o...measuring.html

So if it doesn't fix your results after running it, what could be some problems? Bad room, Bad positioning? I've ran the measurements twice. The first time I forgot to kill all the delays in my amps and I had Anthem ARC on, so I reran all the measurements with Anthem off and the amps and sub delays at 0 ms.

My results are pretty horrible.

I'm guessing it's sub position because it wants me to reduce three subs by -9 to -11db while boosting the back right by 12.3db. Seems like it would kill headroom and there's no way it'll pressurize the room like this.

Set up:
4 sealed 18's in the corners. It's a strange set up in that three are on the floor and one is up in the loft above the kitchen in the back right. When measuring the subs, there isn't a huge gain difference between the 1 versus the 3.

Attached Thumbnails

 

I can't see using a software that recommends 3 subs 20 db down from one. Tough pill to swallow

You can constrain the range of allowable gain of the gain blocks by selecting each gain block in the Config View on the left, then entering the minimum and maximum values of the allowed gain in the boxes in the Properties Window on the right.

I thought about that, and recalculated, but it still gave the same gain for some reason. Still, not really seeing the improvement I was hoping for versus my trial and error results, but I'll keep playing with it.

I'm not following. If you have a gain block whose current gain value is, say -10 dB and you set its minimum allowable value to, say, -6 dB, its current value will be forced to -6 dB in this case. You can see that happen in the Properties Window. IOW, if the existing value is outside the new allowable range, it will be changed to the closest possible value that's within the new range. Further tuning by the optimizer will disallow its value from being outside the new range.

What DSP are you using? How many PEQs per sub? Are you using a timing reference for your measurements? With REW, this could be a loopback with an analog mic or the acoustic timing reference with a USB mic.

Also, the optimization needs to be run for about a half hour when you have a lot of filter blocks. The default optimization time is a very short number.

I was going to the options and changing the limits to -6 to 6 db in the GAIN, but that wasn't changing the gain blocks in the model. I had to manually change all four gain blocks and now it's taking.

I'm using MiniDSP 2x4HD and I set modeling software to spit out biquads (10). Haven't made it that far because I'm still trying to get a simulation I want to input.

Timing reference is Rew's "use acoustic timing reference" option with a USB mic. I spit out the HF out of the left speaker (amp on) and measure out the right speaker (right speaker amp off, subs on).

I am now getting better results, Stdev of 2db with some unfortunate areas in 60-80hz. I shortened the correction windows to 10-120hz and I'm running the 30 minute simulation now. I'll report back.

Thanks!

Ah, I see. You were using the Application Options, not the Properties Window. The app options are mainly for configuring the software to match the limits of particular hardware. For instance, the maximum allowable Q of the PEQ filters in the iNuke is 10, so it's best to set it to 10 in the application options so all newly-created PEQ filters will be compatible with the iNuke. And you're right, changing these limits in Application Options won't affect existing filters, only newly-created ones. The application options make less sense for a miniDSP, as its biquad import facility gives you so much flexibility that there are few constraints. In this case, the default maximum Q is just there to lessen the chances of having solutions with a spiky response.

You can think of the Application Options as being specific to the hardware, and the parameter limits set in the Properties Window to be specific to an optimization.

BTW, if you want to compare different setups, you can clone a configuration and make the modifications to the cloned copy. This allows you to compare results with other, different filter setups. The second graph on the main MSO page is that of two different configurations: before and after optimization.

I think I'm good, man. I don't want to sit here and act like I nailed it with trial and error, but it was good. That said, your software took it up another level. I'm pretty happy with the results until I get these in ported boxes some day.

Thanks!

Appreciate the support as well.

You're welcome!

I searched the thread for "placement" and didn't find anything that would point me in the right direction. I apologize if my search wasn't done properly to yield helpful results. I have two collocated tube subs (extra gain) w/one amp in the front of my room approx 1/3 room width from one side wall and the same setup in the rear 3/5 the room width. I recently purchased a miniDSP to present a single mono sub to Dirac and am researching the MSO to set up my subs prior to running Dirac. Playing while I am typing this is a Mr. Geddes Youtube video on multiple subs. Does his advice that sub placement does not matter apply to MSO, or is there something inherently better (read: easier to optimize) about one placement pattern over another?

Jeff

Jeff, if I read your setup correctly, you will not be able to take advantage of MSO and you are limited in your placement as you have two sub drivers sharing a single amp, which means a single signal from DSP/receiver/preamp. With a single amp, you'll need to locate your subs consistently within the room in regards to the primary listening position. For instance, you could put your subs flanking a center channel speaker or the front two corners or just back from the corners but you'll need to make sure that each sub is the same distance from the primary listening position. If you have one sub up front, presumably farther away, and one in the back, presumably closer, the bass from the rear sub would reach your ears before the front sub's bass. This would likely result in muddy or boomy bass.

You'll either need to stick to equal-distant positions with your currents subs, add a second sub amp to your current sub system, or buy another sub to put in the back as an example. With multiple amps, you'll be able to use DSP on each signal (to sub amp) and more importantly, you'll be able to time align each sub with it's own amplifier.

It I had written my setup correctly, you would have read it as I have two Hsu TN-1220HO subs and a (single channel) Hsu amp in the front and two Hsu TN-1220HO subs and a second Hsu amp in the rear. (The collocation of two subs acts as a single sub.) Does two subs and a miniDSP allow me to use MSO?

Jeff

Yes. This is on the main MSO page under System Hardware Requirements. The relevant portion is:

Thanks. I will give it a fiddle. It's free and could improve my system.

Could anyone offer any sub placement advice? Go with thirds/fifths as I have it, or midwall-midwall front-back?

Jeff

It sounds like your pair of subs in the front are right next to each other and the ones in the rear are right next to each other (collocated). Collocation is often discussed in terms of the additional output provided but really, even sharing an amp, I'd at least try to spread the front subs some and the rear subs some, as long as each set of subs is the same distance to the listening position. Front left and front right both 12' from main listening position as an example. So this might give you some additional locations to try and measure to see what works best in your room.

Regarding your specific question, the normal best case with two subs is mid-wall front and rear but that assumes a symmetrical room and in the real world it may or may not be the best. Would definitely use MSO and start running a few scenarios to see what works best in your room.

I have been using two dual opposing Active subs for a while with Dirac. Had decent results with Dirac itself (10-150Hz). Now I just 4 more subs. Instead of trying to figure out the delay etc... for all subs, planning to give MSO a shot. First timer. About to go read the tutorial.

I'm planning to use 2 at MLP and 2 at other locations. These 4 subs are passing and using external amp. Issue is that amp has 2 outputs. Therefore, I'm placing these subs in pair with equal distance from MLP. If I measure all these subs individually, MSO might set their distance different but since I'm running two subs from one output, only 1 distance can be set. Should I run two subs together in this case since I won't be able to set the distance on them individually?


Also, I'm assuming that I should set the distance to 0 for all subs in minidsp?. Avr for sure should be set to 0 distance. I'll be crossing over at 80-100 but I'm assuming that I should take full range measurements for subs (0-300) wihtout any bass management set?

If I understand you correctly, you only have two independent amplifier channels, and each channel drives two subs? If that's true, measure both subs that are on the same amplifier channel simultaneously and treat them as one sub in MSO. Before measuring, it's best to set the miniDSP delays and attenuations to zero msec and zero dB initially. The calculated results from MSO represent the change that must be applied relative to the condition at which the measurement was performed. It can get very confusing applying the needed delays if they weren't zero when the measurement was performed.

Since you're using Dirac, you'll want to use a "sub-only configuration" in MSO. In this case, you can use the LFE channel to energize the subs. Set the "LPF of LFE" as high as it will go.

There is a new version 1.21 up now. The only changes are a fix to the bug reported by @rumpeli . This bug occurred when renaming a filter channel after creating a measurement group having measurements associated with that channel. In that case, the associated measurement group(s) would become invalid. The other change is making the CHM help file match the latest updates from the web documentation.

Thx. Makes sense. I don't have "LPF to LFE" in avr. I use Yamaha 3050. What I do generally when I'm checking only subs is set the cross-over as high as I can on Center and then use that channel to take reading and keeping Center channel disconnected from amp. This way only sub is played to highest frequency (200Hz) before avr switches to C. Since C is turned off on amp, C doesn't play any sound or effect in any way with sub output.

I'm using 1.20 and one issue I found is with Help window. When I have help window maximize, there is no way for me to switch main window. (Ctrl+tab) or even clicking in task bar doesn't bring main window to front. The only option left for me is to minimize help window to go to main window. Don't think its that big of a deal but it does make it hard to follow tutorial on one screen. BTW, I'm using windows 10.

A workaround for now is to find the multi-sub_opt.chm file (usually in c:\Program Files (x86)\Multi-Sub Optimizer) and make a shortcut to it somewhere, like the desktop. When you launch the help file outside MSO from a shortcut, it doesn't have this annoying behavior.

Or you can use the MSO web site in your browser. That doesn't give you the tree view navigation though.

Perfect thx. You asked me to set the distance/gain ... to 0. What about phase. I know that the front subs are not in phase with back subs. I saw the phase in impulse. But considering that each sub is measured independently, its ok to keep all the phase to normal. MSO will figure it out if there is a phase issue b/w two subs?