MSO : minidsp vs Arendal 1723 dsp

[Lifer]

Hello everyone,

I am going to acquire 2 subs and for proper integration I was planning to use MSO.

When looking at the integrated DSP features of Arendal sub that can be set with the remot app, I wonder if a minidsp is really mandatory.

Arendal 1723 Sub DSP:
- LPF frequency and slope
- HPF frequency and slope
- 0-180 phase
- 7 band PEQ (1Hz increment, +3/-10db gain)
- 0-50ms Time delay (0,25 ms increment)

Would I miss something for MSO that the minidsp would offer on top ?

Thanks in advance !

 

[abdo123]

Does your audio processor already offer two LFE outputs?

 

[TurtlePaul]

Have you measured the speakers in your room?

Some rooms are OK with just sub-100 hz equalization. However, a lot require work up to 300-400 hz. This can be heavily dependent on speaker placement. For example, I know that the shelf my speakers are on is creating issues up to 335 hz, which I am addressing with DSP. Can't do that with just the sub settings.

 

[anotherhobby]

Hello everyone,

I am going to acquire 2 subs and for proper integration I was planning to use MSO.

When looking at the integrated DSP features of Arendal sub that can be set with the remot app, I wonder if a minidsp is really mandatory.

Arendal 1723 Sub DSP:
- LPF frequency and slope
- HPF frequency and slope
- 0-180 phase
- 7 band PEQ (1Hz increment, +3/-10db gain)
- 0-50ms Time delay (0,25 ms increment)

Would I miss something for MSO that the minidsp would offer on top ?

Thanks in advance !

I don't know anything about the Arendal, but you do get more bands of PEQ (and thus likely a flatter curve) with the miniDSP 2x4HD. That said, I only used 5 bands of EQ in my MSO setup. The miniDSP also has a lot more gain/cut per PEQ. The gain limits of +3/-10 on the Arendal seems like it would be limiting, and I can say it would not have been enough for my MSO correction. Every setup is different though, and it doesn't hurt to try. The other thing I'd consider (which I don't see there) is if you can adjust the Q of each band. I would assume you can, but it's not listed in your specs above. At any rate, why not just give it a try first and see what you get?

 

[TurtlePaul]

Looking at the manual, it is a parametric EQ, so you can adjust the Q (from 0.3-10.0). It is limited to below 200 hz, so it only seems to apply to the sub and not eq the mains which are just high passed (but I could be wrong on this).

Another thing to consider: the sub has line level ins and outs, so you will need your stereo volume control before the sub and a power amp after the sub. Other arangements - e.g. most receivers - would be harder to arrange to have the volume control and crossover work correctly.

Also, it would be interesting to see the measurement of the DSP output for the mains to see if it is a 70 dB SINAD situation or if it is 90-100 dB+.

 

[Lifer]

Hi,
Thanks for all your answers.

The processor would be a Denon X4700 which has two LFE outputs but not independent (of my understanding).
Therefore, the minidsp would be only for Subs integration and not for full EQ since this would be taken care by Audissay.

Arendal PEQ can indeed set the Q for each band.

So if I understand correctly, the minidsp would just provide more PEQ bands and more gain.

I am still wondering if the DSP features of Arendals are a strong enough argument to chose them instead of other subs

 

[anotherhobby]

Hi,
Thanks for all your answers.

The processor would be a Denon X4700 which has two LFE outputs but not independent (of my understanding).
Therefore, the minidsp would be only for Subs integration and not for full EQ since this would be taken care by Audissay.

Arendal PEQ can indeed set the Q for each band.

So if I understand correctly, the minidsp would just provide more PEQ bands and more gain.

I am still wondering if the DSP features of Arendals are a strong enough argument to chose them instead of other subs

I personally would not factor the sub's DSP features into the equation considering the relatively low price of the miniDSP. Even in the cheap $100 2x4 (which is what I use) you can get more bands if you use the crossover biquads and input PEQ bands. However, if I already had those subs, I'd try the built in EQ before buying the miniDSP (if that makes sense?).

 

[Sancus]

I use all 7 Arendal PEQs with settings from Multi Sub Optimizer and it works great. Would a miniDSP be incrementally better? Yeah probably. It's easy enough to add later if wanted though.

You can specify boost/cut and Q limits so MSO won't go outside the limits when calculating its solution.

The only way to determine if it would produce a better solution with a miniDSP is to run the optimization twice, under the different limitations of each. So nobody can tell you how much improvement you would get in your room.

 

[abdo123]

I think you would be more than fine using the built-in EQ.

Make sure that you minimize the seat to seat variation first by modeling the room and figuring out the best location for the subs. MSO cannot do miracles.

 

[Lifer]

Thanks!
Except if I find a good deal on a SVS then I will go with the Arendals and if not enough I will add a minidsp later.

 

[GalZohar]

Don't you still get 3 bands of PEQ with SVS? Granted it isn't 7, but I would expect that with Audyssey on top you wouldn't need too many? I wonder if there is a good way to somehow configure MSO to just apply correction that is different between the subs, and let Audyssey apply any filters that are equal to both subs. Maybe that's what the "minimize seat to seat variation" does? Minimize the differences between seats so that Audyssey can correct the resulting response better?

 

[abdo123]

Don't you still get 3 bands of PEQ with SVS? Granted it isn't 7, but I would expect that with Audyssey on top you wouldn't need too many? I wonder if there is a good way to somehow configure MSO to just apply correction that is different between the subs, and let Audyssey apply any filters that are equal to both subs. Maybe that's what the "minimize seat to seat variation" does? Minimize the differences between seats so that Audyssey can correct the resulting response better?

Yes that's exactly what it does.

 

[GalZohar]

Yes that's exactly what it does.

So if there is no seat to seat variation, is there no benefit for individual sub EQ? That is, under no circumstances will individual sub EQ do a better job than just EQ them as 1, except for seat to seat variation? Or is there still some benefit applying MSO for flattening the response with individual sub EQ before running the receiver's room EQ (ex: Audyssey XT32)? That is, assuming we neglect differences between seats and the (probably minor) differences between the PEQ and whatever the room EQ can do.

 

[abdo123]

So if there is no seat to seat variation, is there no benefit for individual sub EQ?

In principal there isn't.

But again if you have your own personal concert hall or anechoic chamber and as a result you have no seat to seat variation please invite me over for champagne and caviar sometimes

 

[GalZohar]

I meant as in the only benefit is reducing seat to seat variation, and that the rest can be done just as well by external EQ that sees both subs as one. And also if the seat to seat variation is small and you have such external EQ, then you probably won't need many filters and hopefully even the 3 bands from SVS will suffice? And basically an extra sub with optimized delay can help flatten the response at the MLP but individual subwoofer EQ can't really do that any better than just EQing both subs as one? Then again MSO also optimizes delay, doesn't it? And for that one would still want flattening and not just seat to seat variance reduction?
Sorry but I am a bit confused about what exactly can achieve what, and what doesn't actually help.

 

[Sancus]

I meant as in the only benefit is reducing seat to seat variation, and that the rest can be done just as well by external EQ that sees both subs as one. And also if the seat to seat variation is small and you have such external EQ, then you probably won't need many filters and hopefully even the 3 bands from SVS will suffice? And basically an extra sub with optimized delay can help flatten the response at the MLP but individual subwoofer EQ can't really do that any better than just EQing both subs as one? Then again MSO also optimizes delay, doesn't it? And for that one would still want flattening and not just seat to seat variance reduction?
Sorry but I am a bit confused about what exactly can achieve what, and what doesn't actually help.

MSO sets overall gain and delay in addition to PEQs. 3 is not a lot but it's better than nothing.

In general if you're using something else as well like Audyssey, I would run MSO on the uncorrected subs first. First of all, that means you can tweak Audyssey without re measuring and re running MSO every single time. Plus, Audyssey performs best when it's addressing what it thinks is one sub. So you want to use a splitter and one sub out. It did provide an additional improvement for me even with 2 subs and 7 PEQs each.

If you have an EQ system that actually addresses multiple subs correctly in a similar manner to MSO(like Dirac with the extra bass control module) then I wouldn't use MSO at all. But most of them don't do this anyways.

 

[abdo123]

I meant as in the only benefit is reducing seat to seat variation, and that the rest can be done just as well by external EQ that sees both subs as one. And also if the seat to seat variation is small and you have such external EQ, then you probably won't need many filters and hopefully even the 3 bands from SVS will suffice? And basically an extra sub with optimized delay can help flatten the response at the MLP but individual subwoofer EQ can't really do that any better than just EQing both subs as one? Then again MSO also optimizes delay, doesn't it? And for that one would still want flattening and not just seat to seat variance reduction?
Sorry but I am a bit confused about what exactly can achieve what, and what doesn't actually help.

MSO can work in two modes.

Either reduce seat to seat variation, in this mode you would still need global EQ or input EQ to make sure that the response is at the end flat. This mode does not in the slightest focus on a flat response in ANY of the seats. but just that the response is more or less the same in All seats.

Or the general optimization mode which focuses on both reducing seat to seat variation AND achieving a flat response at the same time (or whatever target curve you provide) for all seats.

Your comments about 'What If there is not much seat to seat variation' are mostly academic, which is why i'm finding it difficult to understand why this would be of interest to you.

I'm going to give an example of my own living room. I have taken 5 measurement points over a small couch (180cm X 80cm), there is no EQ but the location of both the couch and my two subs have been EXTREMELY optimized over the years, the delay between the two subwoofers is also ideal.

Even in such a small space, with two subs, and countless hours of optimizing the location of everything there is still a huge seat to seat variation at the ~35Hz room mode (a whooping 20 dB variation). Luckily very little material have energy this low so i found it to be the best compromise for my room.

The only way i could minimize variation at 35Hz at this point is if i buy two more subs, which is what i'm planning to do in the future.

Most people have several seats that spread over several dimensions (unlike my small couch) and would exhibit much more variation with two subwoofers than I do.

By the way your comment about an extra sub inherently flattening the response is generally incorrect. it will only reduce seat to seat variation and perhaps filling up deep holes in the response. but the peaks remain untouched.

 

[GalZohar]

If the couch is on a long axis and not near either the center nor the side, then the lowest frequencies on that width axis (those that cause serious variations between the seats) should be less significantly different. Length and height modes should be similar for the seats. Not that this is an optimal layout, but that's what I have.
Also in general if you care mostly about the MLP then minimizing seat to seat variation is also somewhat less important.

In any case that wasn't the main purpose of my question. What I was trying to understand is what is the correct way to optimize when there is a global EQ that gets to EQ the subwoofers as a single sub (like Audyssey when not using SubEQ HT which is, as I understand, less effective than manual delays, not to mention MSO optimzied delays). Supposedly time/phase alignment between the subwoofers still needs to be optimized for MLP by MSO/manual DSP, but if I understand correctly the actual flattening can be safely left to the AVR as individual subwoofer EQ won't help with that? And if that's correct, how is MSO supposed to be used to achieve it, as we would want delays/phase for the flattest response but individual sub PEQ for minimizing seat to seat variance?

Another thing is that while I understand 3 PEQ is probably not enough for flatting a sub, in a case without extreme seat to seat variation and external EQ (so need the PEQ just for reducing variation), would it suffice in that case? Anyone experimented with MSO and SVS subwoofers without a minidsp?

 

[abdo123]

If the couch is on a long axis and not near either the center nor the side, then the lowest frequencies on that width axis (those that cause serious variations between the seats) should be less significantly different. Length and height modes should be similar for the seats. Not that this is an optimal layout, but that's what I have.
Also in general if you care mostly about the MLP then minimizing seat to seat variation is also somewhat less important.

In any case that wasn't the main purpose of my question. What I was trying to understand is what is the correct way to optimize when there is a global EQ that gets to EQ the subwoofers as a single sub (like Audyssey when not using SubEQ HT which is, as I understand, less effective than manual delays, not to mention MSO optimzied delays). Supposedly time/phase alignment between the subwoofers still needs to be optimized for MLP by MSO/manual DSP, but if I understand correctly the actual flattening can be safely left to the AVR as individual subwoofer EQ won't help with that? And if that's correct, how is MSO supposed to be used to achieve it, as we would want delays/phase for the flattest response but individual sub PEQ for minimizing seat to seat variance?

Another thing is that while I understand 3 PEQ is probably not enough for flatting a sub, in a case without extreme seat to seat variation and external EQ (so need the PEQ just for reducing variation), would it suffice in that case? Anyone experimented with MSO and SVS subwoofers without a minidsp?

Okay so the number of PEQs is not really vital, MSO is a brute force algorithm that will work with whatever you feed it.

Even with 3 PEQs and delay and gain values you will get far.

Basically you would assign MSO whatever digital processing you have left after the input/global EQ. But generally that is only useful for reducing the seat to seat variation.

Whatever room correction you're using in 2022 should be more than capable (horsepower wise) to reach whatever RMS average you decide on with your target curve during the calibration step.

 

[GalZohar]

I wasn't refering to lack of horsepower of the room correction but rather its lack of ability to EQ the subs individually (which most not-super-expensive systems can't do), but supposedly that's not actually needed if MSO already reduced seat to seat variation to a minimum?