2 sub adjustment w/ miniDSP 2x4 HD & REW with an XMC-2?

[tcoble3]

I've got an XMC-2 and two HSU VTF-3 MK5 subs, with a miniDSP 2x4 HD and a UMK-1 on order so that I can run REW to get the subs dialed in. A couple of questions for anyone else with an XMC-2 or comparable model who has tackled this process :

1. I've currently got my subs connected via the Left and Right outputs on the XMC-2. Do I need to change this and run a single cable/input into the miniDSP 2x4 HD? Or can I keep things as-is and run two inputs into the miniDSP? Is there a functional difference between these two approaches, or does it not really matter?

2. All of the tutorials that I've seen indicate that prior to running REW it's necessary to go into the AVR or pre/pro setup and turn off all room correction (in this case, Dirac), set crossovers at their highest settings, set all speaker trims to zero, etc. Is there any reason to NOT to take any of these steps, or is this sound advice? Also, are there any other settings on the XMC-2 that I should adjust before starting?

3. Any other general advice (re: connections, configurations, settings, best practices, etc.) for making the most of this process - particularly as it relates to the XMC-2? If there's an Emo-specific tutorial here on the forums that I haven't found, I'd appreciate it if someone could point me to it. Thanks!

[marcl]

Dec 27, 2021 6:14:06 GMT -5 tcoble3 said:

I've got an XMC-2 and two HSU VTF-3 MK5 subs, with a miniDSP 2x4 HD and a UMK-1 on order so that I can run REW to get the subs dialed in. A couple of questions for anyone else with an XMC-2 or comparable model who has tackled this process :

1. I've currently got my subs connected via the Left and Right outputs on the XMC-2. Do I need to change this and run a single cable/input into the miniDSP 2x4 HD? Or can I keep things as-is and run two inputs into the miniDSP? Is there a functional difference between these two approaches, or does it not really matter?

2. All of the tutorials that I've seen indicate that prior to running REW it's necessary to go into the AVR or pre/pro setup and turn off all room correction (in this case, Dirac), set crossovers at their highest settings, set all speaker trims to zero, etc. Is there any reason to NOT to take any of these steps, or is this sound advice? Also, are there any other settings on the XMC-2 that I should adjust before starting?

3. Any other general advice (re: connections, configurations, settings, best practices, etc.) for making the most of this process - particularly as it relates to the XMC-2? If there's an Emo-specific tutorial here on the forums that I haven't found, I'd appreciate it if someone could point me to it. Thanks!

There can be a bit more to this but here are some basics:

• All three sub outputs - when set to Mono/Dual Mono - put out the same signal.  To use your miniDSP, connect the Center Sub output to one of the inputs of the miniDSP, and your two subs to two miniDSP outputs.  You route the input to the two outputs in the miniDSP app.
• Yes, turn off Dirac and any PEQ in the XMC-2, and set the gain on your two subs about the same, and the phase switch or control to 0 on the subs also.  Turn the crossover on the subs to maximum.
• When you use REW you will test the subs using output #4.  This will send the signal to the XMC-2 as a LFE output and the XMC-2 will roll off the response above 120Hz.  If you're using the subs only for LFE, or if you are crossing over to small speakers below 100Hz, this is fine.  If you need to see the sub response higher than 100Hz, you need to use a different output.
• Much more to the alignment of the two subs in the miniDSP than I'll write here.  It involves matching levels, aligning phase (time delay) using REW, etc.  Check out Austin Jerry's REW tutorial here:

www.minidsp.com/support/community-powered-tutorials/295-minidsp-tutorials-by-austinjerry

And miniDSP sub alignment tutorial here:

www.dropbox.com/s/w7jmpozner6muzw/MiniDSP%202x4%20Set-Up%20Guide%20Latest%20Version.pdf

[tcoble3]

Thank you, sir! This is exactly the info that I was seeking. I think I'm making good progress on understanding the basic alignment and EQ processes, but this helps with some of the preliminary setup issues that were eluding me.

A couple of quick clarifying questions:

1. Once I'm done running REQ, I'll obviously turn Dirac back on - but do I keep the subs' crossovers turned up to max or return them to their normal settings? (I assume the latter, but just want to confirm.)

2. My understanding is that if I use the center sub output for both speakers (via the miniDSP) I'll have the option to set that center output as either "Mono" or "LFE". As I want my subs to do as much of the low-end work as possible, am I correct in assuming that I should go with the Mono setting, so as to not limit sub output to just the LFE channel? I've got may mains all set to "small" and crossovers set at or above 80. (And if I'm not limiting the output to just the LFE channel, then there's no need to use output #4 in REW, per your third point, correct?)

Thanks again!

[marcl]

Dec 27, 2021 9:14:19 GMT -5 tcoble3 said:

Thank you, sir! This is exactly the info that I was seeking. I think I'm making good progress on understanding the basic alignment and EQ processes, but this helps with some of the preliminary setup issues that were eluding me.

A couple of quick clarifying questions:

1. Once I'm done running REQ, I'll obviously turn Dirac back on - but do I keep the subs' crossovers turned up to max or return them to their normal settings? (I assume the latter, but just want to confirm.)

2. My understanding is that if I use the center sub output for both speakers (via the miniDSP) I'll have the option to set that center output as either "Mono" or "LFE". As I want my subs to do as much of the low-end work as possible, am I correct in assuming that I should go with the Mono setting, so as to not limit sub output to just the LFE channel? I've got may mains all set to "small" and crossovers set at or above 80. (And if I'm not limiting the output to just the LFE channel, then there's no need to use output #4 in REW, per your third point, correct?)

Thanks again!

Okay, we're on the right track, now for a couple more details 

• Always keep the sub crossovers set to max.  You do crossovers in the processor and you want the full range of LFE to get to the subs.
• Let's assume that all you do in miniDSP (for now anyway) is use the single Center Sub output to send LFE and bass management (small speaker bass) to the two subs.  And, that the only adjustments you make in the miniDSP are to match the sub output levels to each other, and adjust time delay to get the two subs in phase with each other.  Let's assume you do no PEQ or crossovers in the miniDSP.
• After aligning the subs with the miniDSP, you rerun Dirac.  Dirac will treat the subs as one sub, and since they are aligned to each other and Dirac aligns them to all the other speakers, everything is in alignment.
• To the point about setting Center Sub to LFE vs Mono, since all your speakers are small you must set Center Sub to Mono and have both LFE and bass management go to the subs.  Using the LFE setting for Center Sub requires some other way to do bass management, so let's leave that alone for now.
• You can use REW Output #4 for your sub alignment.  After you run Dirac you can measure all the channels with REW and the measurements for all the small speakers will reflect the combination of the small speaker above the crossover, and the subs below the crossover.  these measurements might suggest tweaking your crossover points to get a smooth transition with the subs.

[ttocs]

Dec 27, 2021 6:14:06 GMT -5 tcoble3 said:

I've got an XMC-2 and two HSU VTF-3 MK5 subs, with a miniDSP 2x4 HD and a UMK-1 on order so that I can run REW to get the subs dialed in. A couple of questions for anyone else with an XMC-2 or comparable model who has tackled this process :

1. I've currently got my subs connected via the Left and Right outputs on the XMC-2. Do I need to change this and run a single cable/input into the miniDSP 2x4 HD? Or can I keep things as-is and run two inputs into the miniDSP? Is there a functional difference between these two approaches, or does it not really matter?

2. All of the tutorials that I've seen indicate that prior to running REW it's necessary to go into the AVR or pre/pro setup and turn off all room correction (in this case, Dirac), set crossovers at their highest settings, set all speaker trims to zero, etc. Is there any reason to NOT to take any of these steps, or is this sound advice? Also, are there any other settings on the XMC-2 that I should adjust before starting?

3. Any other general advice (re: connections, configurations, settings, best practices, etc.) for making the most of this process - particularly as it relates to the XMC-2? If there's an Emo-specific tutorial here on the forums that I haven't found, I'd appreciate it if someone could point me to it. Thanks!

1. Run a single cable from the Left Sub Output to the miniDSP, then out to all the subs from there. Yes, there is a functional difference. The goal is to make it look to the XMC-2 like there is only one subwoofer on each output.

2. This is what I did. Turn up the XO all the way for a channel, turn off the amp to the speaker on that channel so only the subs are working. Tweak the subs to your delight.

3. Try not to be confused about things for the thread I'm showing here, just look at the subwoofer routing only. But, go to Finding Subwoofer at:
emotivalounge.proboards.com/thread/57730/subwoofer

Optional
I use two subwoofer outputs which both go the miniDSP, then out to the same subwoofer group. I do this so Bass Management is separate from LFE. Read about it in the thread. Look at the end of second post with diagram of subwoofer path only.

Left Sub Output = Mono
Center Sub Output = LFE

But you can simply just use Left Sub Mono for simplicity if you like. I like being able to adjust the Level of each output independently, which is why I do it using both outputs.

edit: Man, my brain is still not in gear yet (sick last week). Didn't notice Marc's responses. He's got good advice. I do what I suggested because the subs can be tested to a higher frequency range. And yes, change the XO after testing so each channel is crossed where you want them.

[tcoble3]

Got it, thanks.

- My HSU subs have the ability to bypass their internal crossovers in favor of letting the processor handle that function - which I'm already doing - so I'm covered on that point. I'll make sure that they're set at the highest level in the XMC-2. (Which is 200 Hz, right?) And as you say, I can tweak as/if necessary after the fact to ensure a smooth transition with the subs.

- Using the miniDSP primarily for aligning levels and time delay makes sense, though I'm slightly surprised by your recommendation to forego using it for PEQ in favor of re-running Dirac for that. While I'm not well versed in the details of Dirac's bass equalization capabilities, given that so many people have been bemoaning the lack of the Dirac bass 'module' on Emo products I had assumed that it didn't have much to offer in the bass management department. But if I'm reading your comment correctly, you're saying that Dirac can handle EQ duties for the subs as well, yes? So there's no compelling advantage to using REQ/miniDSP to EQ the subs? Interesting...

- So, theoretically, since I've already run Dirac - subs included - if I were to run REW sweeps without first disabling Dirac, the frequency responses for the subs should look reasonably good? (Level matching and time delay matching notwithstanding.)

- Center Sub to Mono to get both LFE and bass management sent to the subs - got it. And since I don't have any speakers configured as Large, the Enhanced Bass option (which I DON'T want) doesn't come into play, correct?

[marcl]

Dec 27, 2021 10:47:34 GMT -5 tcoble3 said:

Got it, thanks.

- My HSU subs have the ability to bypass their internal crossovers in favor of letting the processor handle that function - which I'm already doing - so I'm covered on that point. I'll make sure that they're set at the highest level in the XMC-2. (Which is 200 Hz, right?) And as you say, I can tweak as/if necessary after the fact to ensure a smooth transition with the subs.

- Using the miniDSP primarily for aligning levels and time delay makes sense, though I'm slightly surprised by your recommendation to forego using it for PEQ in favor of re-running Dirac for that. While I'm not well versed in the details of Dirac's bass equalization capabilities, given that so many people have been bemoaning the lack of the Dirac bass 'module' on Emo products I had assumed that it didn't have much to offer in the bass management department. But if I'm reading your comment correctly, you're saying that Dirac can handle EQ duties for the subs as well, yes? So there's no compelling advantage to using REQ/miniDSP to EQ the subs? Interesting...

- So, theoretically, since I've already run Dirac - subs included - if I were to run REW sweeps without first disabling Dirac, the frequency responses for the subs should look reasonably good? (Level matching and time delay matching notwithstanding.)

- Center Sub to Mono to get both LFE and bass management sent to the subs - got it. And since I don't have any speakers configured as Large, the Enhanced Bass option (which I DON'T want) doesn't come into play, correct?

BTW, ttocs and I are friends and talk a lot about this stuff.  We may have some slightly different methods but we're on the same page :-)

• When aligning the subs in the miniDSP and using Output #4 in REW, crossovers in the XMC-2 have no effect.   Leave the crossovers on the subs bypassed, and leave the crossover points for your other speakers set to whatever is appropriate.  Note ... crossover settings have no effect on LFE.
• DLBC is mostly for multiple subs.  You're using the miniDSP to accomplish this.  If and when we have DLBC as an option it probably would replace the miniDSP.  But in the meantime, Dirac does just fine EQ'ing subs ... much more sophisticated filter algorithms than miniDSP.  It's best to let Dirac do all the EQ.
• Once you align the subs using the miniDSP, you have to rerun the Dirac calibration so it can properly align and EQ the subs.  

[tcoble3]

Good to know! Given various interactions on the firmware beta testing board over the past year or so, I've got HUGE respect for both of you. Seeing that you're on the same page on this topic puts a big smile on my face.

So in your second bullet are you saying that the primary benefit of DLBC will be aligning levels and time delays on multiple subs, rather than EQ functionality, which our current version of Dirac already provides? If so, I'm more than happy to let Dirac handle EQ duties for the subs. Thanks for pointing that out and saving me from spending countless hours of manual tweaking and self-recrimination over not getting everything just exactly perfect.

Cheers!

[marcl]

Dec 27, 2021 11:12:55 GMT -5 tcoble3 said:

Good to know! Given various interactions on the firmware beta testing board over the past year or so, I've got HUGE respect for both of you. Seeing that you're on the same page on this topic puts a big smile on my face.

So in your second bullet are you saying that the primary benefit of DLBC will be aligning levels and time delays on multiple subs, rather than EQ functionality, which our current version of Dirac already provides? If so, I'm more than happy to let Dirac handle EQ duties for the subs. Thanks for pointing that out and saving me from spending countless hours of manual tweaking and self-recrimination over not getting everything just exactly perfect.

Cheers!

The big problem up until now has been that Dirac measures each speaker and each sub separately.  The first issue with this is that the advantage of multiple subs is how they work together in a room.  Dirac will align them to each other, but since it doesn't measure their response together is doesn't EQ the sum of their response.  We fix this using miniDSP.  DLBC fixes the problem by aligning the multiple subs and correcting the SUM of the response ... but it uses the same Dirac algorithms and filters that we have in Dirac Live.  Using a miniDSP and then letting Dirac correct the sub channel works really well.

DLBC also looks at the combined response of small speakers where they cross over to the subs, and aids in fine tuning that crossover point.  And I believe it also does some specific phase alignment of pairs of speakers in their bass region.

[tcoble3]

That actually makes a lot of sense. I can see how EQ'ing both subs as a single, summed unit would be vastly preferable. Sounds like we're actually getting the best of both worlds by doing it in the way that you've described. Now I just have to be patient for my miniDSP to arrive in mid-January...

Again, can't thank you enough for your help and wisdom! (And you too, ttocs.)

[marcl]

Dec 27, 2021 11:39:45 GMT -5 tcoble3 said:

That actually makes a lot of sense. I can see how EQ'ing both subs as a single, summed unit would be vastly preferable. Sounds like we're actually getting the best of both worlds by doing it in the way that you've described. Now I just have to be patient for my miniDSP to arrive in mid-January...

Again, can't thank you enough for your help and wisdom! (And you too, ttocs.)

I forgot to mention a couple things.  The research on multiple subs by Todd Welti talks about how many, and where to place them, irrespective of EQ.  EQ is the final step to try to fix whatever placement and room treatment can't.

miniDSP HD has the capability to implement IIR filters similar to what's available in the XMC-2.  It can do IIR all-pass filters which the XMC-2 can't do, and these can be helpful for some issues.  It can also do FIR filters, which XMC-2 can't do.  FIR filters can affect phase and amplitude independently.  Dirac uses a proprietary combination of filters and algorithms to determine what to fix and how.  This is something that can't be replicated in a miniDSP.  

I actually have tried doing some PEQ in the miniDSP to knock down the big bass peaks, and then let Dirac fine tune.  In the end I stopped using PEQ in the miniDSP because I was not convinced that the combination was any better than using Dirac alone, and it seemed that using PEQ in the miniDSP affected the phase alignment of the subs.  And Dirac alone is a simpler solution.

[ttocs]

As Marc says, PEQ anywhere in the chain can be goofy (technical term). With that said, I've been able to use some PEQ for one subwoofer and none for the other two in the miniDSP and get away with it. This was just to help with an issue with that one sub's interaction with the other two. But go to far with any PEQ and Dirac will get real goofy in its filter as measured after the fact in REW. It didn't show any issues within Dirac's interface, but the final filter being applied and the measured was where anomalies could show up. Once I reduced the PEQ down to a tolerable amount, things seemed ok.

Bottom line. PEQ? Use it sparingly, if at all when using Dirac as the end game.

Oh, I should also mention that I use Reference Stereo for two channel, and use Dirac up to just a few hundred Hz, for only my Left/Right Main speakers, for two channel as well to compare the two. My speakers are finicky with regards to Dirac, so I've been working through being able to at least use it for bass. So when I use Reference Stereo, this is where using that PEQ is most beneficial for me.

[tcoble3]

Interesting. My sub placement options are a bit limited, but I've done my best to optimize location within those parameters. And thanks to some extremely helpful guidance from the GIK Acoustics team, I've made a lot of progress with regard to room treatments - particularly with lower-end absorption, as my room tended to be a bit "boomy" in certain areas. While I'm sure there's more to be done in both areas, I'm fairly comfortable in saying that I'm using EQ fairly appropriately - i.e., to 'fill in the gaps' that haven't been fully rectified by other means.

My knowledge of the various filter types that you mentioned is rudimentary at best, but I can definitely see how Dirac would have the advantage over the miniDSP in terms of providing a more nuanced, refined and powerful solution. A scalpel vs. a good chef's knife, so to speak. In any case, I suspect that once I've used both tools in the way that we've discussed here, I'll be very close to the line beyond which I'd be hard pressed to discern any further improvements were I to continue to tinker. (Not that I'm likely to stop trying, though.)

Now, to go find that Todd Welti research...

[foggy1956]

Dec 27, 2021 13:01:38 GMT -5 ttocs said:

As Marc says, PEQ anywhere in the chain can be goofy (technical term). With that said, I've been able to use some PEQ for one subwoofer and none for the other two in the miniDSP and get away with it. This was just to help with an issue with that one sub's interaction with the other two. But go to far with any PEQ and Dirac will get real goofy in its filter as measured after the fact in REW. It didn't show any issues within Dirac's interface, but the final filter being applied and the measured was where anomalies could show up. Once I reduced the PEQ down to a tolerable amount, things seemed ok.

Bottom line. PEQ? Use it sparingly, if at all when using Dirac as the end game.

Oh, I should also mention that I use Reference Stereo for two channel, and use Dirac up to just a few hundred Hz, for only my Left/Right Main speakers, for two channel as well to compare the two. My speakers are finicky with regards to Dirac, so I've been working through being able to at least use it for bass. So when I use Reference Stereo, this is where using that PEQ is most beneficial for me.

Doesn't reference stereo remove all EQ? Can't remember

[ttocs]

Dec 27, 2021 15:10:43 GMT -5 foggy1956 said:

Dec 27, 2021 13:01:38 GMT -5 ttocs said:

As Marc says, PEQ anywhere in the chain can be goofy (technical term). With that said, I've been able to use some PEQ for one subwoofer and none for the other two in the miniDSP and get away with it. This was just to help with an issue with that one sub's interaction with the other two. But go to far with any PEQ and Dirac will get real goofy in its filter as measured after the fact in REW. It didn't show any issues within Dirac's interface, but the final filter being applied and the measured was where anomalies could show up. Once I reduced the PEQ down to a tolerable amount, things seemed ok.

Bottom line. PEQ? Use it sparingly, if at all when using Dirac as the end game.

Oh, I should also mention that I use Reference Stereo for two channel, and use Dirac up to just a few hundred Hz, for only my Left/Right Main speakers, for two channel as well to compare the two. My speakers are finicky with regards to Dirac, so I've been working through being able to at least use it for bass. So when I use Reference Stereo, this is where using that PEQ is most beneficial for me.

Doesn't reference stereo remove all EQ? Can't remember

Yes it does, which is why I have used it mostly, up until recently. But with Dirac being limited to under 300-400Hz (varies by my mood) it helps with the bass issues and leaves the electrostatic panels alone - mostly, so that's what I've been comparing, and it's been working out very well!

For all the other channels I let Dirac do its thing.

[tcoble3]

Just wanted to follow up with a quick thank you for all of your extremely helpful input. I spent much of the day yesterday with the new miniDSP 2x4 HD, REW and Dirac Live, and thanks in large part to your guidance, the entire process was surprisingly straightforward and productive.

Long story short, I tested many, many combinations of positions for my two subs and was ultimately able to zero in on one combination of placement, time alignment and phase that produced a much stronger (positively summed), flatter combined response. There were a few small frequency ranges where, due to the limitations of my room, I had to make some of the usual trade-offs in terms of making a small sacrifice in one area for the sake of improving another, but overall I was able to either improve or maintain the bass response across almost all of the 20-150 Hz range.

When I then re-ran Dirac, having it measure the two subs as a single source seems to have made a noticeable improvement. While not a night and day difference, I'm definitely noticing the effects of the two subs working together more effectively. There's probably a bit of confirmation bias in play, but after a few hours of listening, my system's bass response no longer feels like an itch that I need to scratch. Ahhhhh....

So again, thanks for so patiently sharing your knowledge and guidance. I owe you one!

[jjkessler]

Wanted to confirm my understanding of the physical connections. For the Mini DSP 2x4 HD physical connections to align 2 subs into a single virtual sub (pre Dirac). I have my RMC left sub connected to MiniDSP input 1. One sub connected to MiniDSP output 1 and one connected to MiniDSP output 2. The MiniDSP then connects to my computer via the MiniDSP provided USb cable. UMIK microphone also connects to the computer to run/ measure REW. Then, I need another connection from my computer (IMAC with no HDMI output and will have to use a USB to HDMI converter ) to an open HDMI input to the RMC for test tone generation

So, three connections at the computer (MiniDSP, UMIK Mic, and RMC)

[ttocs]

Dec 19, 2022 21:03:49 GMT -5 jjkessler said:

Wanted to confirm my understanding of the physical connections. For the Mini DSP 2x4 HD physical connections to align 2 subs into a single virtual sub (pre Dirac).

I have my RMC left sub connected to MiniDSP input 1.
One sub connected to MiniDSP output 1 and one connected to MiniDSP output 2.

The MiniDSP then connects to my computer via the MiniDSP provided USb cable.
UMIK microphone also connects to the computer to run/ measure REW.
Then, I need another connection from my computer (IMAC with no HDMI output and will have to use a USB to HDMI converter ) to an open HDMI input to the RMC for test tone generation

So, three connections at the computer (MiniDSP, UMIK Mic, and RMC)

YES.

[jjkessler]

Anybody able to get all three connections running on a MAC, especially the HDMI to HDMI from the MAC to the RMC to generate test tones in REW?

[ttocs]

Apr 25, 2024 14:35:03 GMT -5 jjkessler said:

Anybody able to get all three connections running on a MAC, especially the HDMI to HDMI from the MAC to the RMC to generate test tones in REW?

If you're referring to the 3 connections from Mac to RMC-1 & UMIK-1 & miniDSP, then yes, I do this all the time.