Though I'm too ignorant to offer a "valid" opinion, I agree that one would first treat the room as good as feasible. Then the multiple subwoofers would be the "last chance" to fix problems that the treatment did not solve.

In my specific room and equipment found that "trying nearly every setting, even the dumb ones" was productive finding as good possible subwoofer adjustments. There is currently one sealed 18" sub under the right main. Later I want to build an identical twin to put under the left main.

I had another sub in the room rear but currently am not using it because it raised group delay and decay time along with fixing the null. Currently the flattest setting seems rather weird but it sounds pretty good--

The coax horns are crossed with the 12" coax woofers at 1 kHz 24 dB Linkwitz-Riley, and the 18" sub is low-pass filtered at 500 Hz. There is no highpass filter on the mains 12" woofer at all, though the sealed cabs start acoustically rolling off around 100 Hz or whatever.

I would never have theoretically decided to try that setting though it works pretty good. It was only discovered by trying a lot of settings even if it seemed stupid. Kind of "guided stupidity" or "successive approximation". If one stupid setting had surprising results, started tweaking in the ballpark of that stupid setting to see if it could be further optimized.

Which is of course specific to my room and equipment. Inapplicable to any other setup.

Hi JCJR- thanks for the great questions and positive attitude!

- Subwoofer LF specs- this one is weird because 2 or 3 we tried have specs that say they roll off below say 40hz, but there is definitely energy produced lower than you can play with using DSP/EQ. I really don't understand this, my only thought is that they want to differentiate from more expensive models?

- No trouble tweaking SPL, crossover, EQ, DSP etc you just have to get a little experience with it, it is not difficult but requires a lot of trial and error. My friend is way better at it than me and he does it in something like an hour

- This room has no low frequency treatment. It has fiberglass panels in first reflection, a simple cloud and a couple of other spots but nothing that has any effect on these low frequencies. In general we are finding that multiple subwoofers are many times more effective than any practical low frequency treatments. Even the PSI AVAA which works is about 1/10 as effective as (4) properly configured subwoofers

- the response definitely changes as you add 1-2-3-4 subwoofers, Toole and Welti at JBL/Harmon suggest 2 is the big gain and 4 is ideal. This is exactly what we see

- Decay time is greatly improved (attached screenshots before and after)
-Group Delay is greatly improved (attached screenshots before and after)
- Distortion- I dont really understand this but I included a screenshot choosing 90hz which is the biggest dip in the "before" and it is 0.574%

In short, the results are very counter-intuitive. You would assume that pouring more LF frequency energy into a small room would make things a lot worse. However, what seems to be happening is a very complex interaction of waves, both constructive and destructive, which ends up smoothing out the room response at these difficult frequencies. It also does this in a pretty wide physical area which makes the practical "sweet spot" pretty large.

I think I address what you asked, looking forward to more discussion!!

here is the distortion screenshot

I have JBL LSR 305, in a very small room . . . I also use Dirac audio processor. . . . is this software capable of correcting the bass to the extent that a sub would?

Dirac can EQ the speakers but it can't address room modes like a multiple subwoofer approach can. When you place the (4) subs properly most of the work is done. The amount of EQ needed at that point is minimal. Without them you would have to EQ sometimes 30dB which is not practical for several reasons, perhaps the biggest being it would only be effective at one spot in the room.

if you dont plan to use multiple subwoofers than your best available tool is to experiment with placement and try to find a location that is the most even, then Dirac can likely provide some help to optimize it

Dirac are not standing still. They have been and are developing multiple source solutions. I have been privileged to see some pretty amazing results. Much of it has to do with high end cars, aircraft, and such. But I think we can hope to see something of that ilk, of use in the Studio sometime.
DD

Agreed, live sound, home theater, cinema, etc are way ahead of the studio world on this stuff. On the multi-channel stuff you saw did they support multiple subwoofers in the auto-alignment system??

I think I could benefit from two subs underneath my monitors in my new room (decently treated now).

What software or something could I use for highpassing the monitors and do the adjustments needed for a good integration? A lot of subs offer only say a 80Hz highpass of mains, I'd want to experiment further as most of my "problems" is in the 100-300Hz area.

Hello,

if the problem is modal
Two sub in mono or chained : bass correlated (each sub reproducing the same signal) yes.
Two sub in stereo : bass uncorrelated (each sub reproducing a signal different) i doubt

with the phase stereo : no.

@ Steve , I haven't run DL multichannel, but I am pretty sure the current multi channel version is full range. The R&D I saw definitely had multiple processed LF. Very multiple. The LF response from 200 down was flat and fast according to the data. No room treatment but I didn't see particulars of the actual room. Maybe it was a tent......;-)

@ Deckdaddy , your response, particularly if you used more smoothing, looks pretty flat. That would sound very bright to my ear. You have a consistent dip at 400Hz. This could be caused by a human body close to the measurement mic.
That peak could well be a height mode or combination. It could probably be softened successfully with Eq. Your LF looks good but a bit more would be nice. Turn it up, or use a Shelving Eq. I would demo Dirac Live.
DD

My only problem with multi sub setups is that the DSP has to be 'just right' time alignment / phase + amplitude wise and it just cannot fix room decay issues.

Having a flat FR is good - but still muddy because much too long doesn't really help the guys that need to hear their LFOs automations, fast edits and notes, kick vs bass pumping, stereo effects etc.

In this case, headphones are still needed - so would the 'car and living room system check'.

Hi Thomas. Clearly the only way to achieve the most neutral decent sized listening area is by acoustic design and treatment. But when treatment is inadequate due to money, space, and other logistical restrictions, these last resort techniques are heaven sent. DL's primary work comes from spaces which have no possibility of treatment, cars, airplanes. But I do note pretty LF transparent boundaries.

Historical development aside, DL software in normal room use, shows improvements in the Impulse Response, and actual shortening of modes.
Way back we had the BagEnd Active LF trap, now the PSI and many DIY efforts.
If we consider increasing the number of LF sources, say to a dozen or more, and consider polarity reversals and time alignment, well then it looks like the Harmon Multi sub benefits combined with those of Active cancellation.
DD

But as Thomas pointed out, there is still the room decay issue. Multiple active subs isn't a replacement for room damping unfortunately, but given the choice of using one or the other, subs should fair well.

Hi, DD and thank you for responding to my room thread as well! I stood ~3 meters away from the mic when measuring and I did averaging for better noise floor. I don't think it sounds very bright, but I feel the bass could be a little stronger to balance it (OK, so bright in a way), a solid foundation for the rest of the image. It sounds a little thin down there, and as it slopes down starting a little high in this room. I thought a sub could make it more present and flat. I can't push much more bass out of my ~6" PSI A17-Ms.

I have tried Soundworks Reference 3 tje last few weeks, and it certainly made the low mids and bass flatter and less boxy, but I hope the HD version will be a little more accurate, and I wish I could adjust low end/mids/high end wet/dry seperately as I think it sounds a little too thin in the low mids, and a bit too bright up top at 100% wet. Go down to 50% and the correction is not the same elsewhere. I will check out Dirac Live! I could get a sub or two highpassed at 80hz down the line to get a flat and strong bottom end for my clear and detailed PSI image. I tried moving monitors but they must be at wall towards corner trap, I can try to put them higher up, but not lower.

Seems like I can get a little closer to were I want to be with a gentle tilt in Reference 3. I found the curves a little too heavy for what I'm used to, I think there should be a seperate amount knob on these curves, so I could have 100% wet calibrarion but with 50% or so home curve. But then we're approaching what we can adjust by tilt/bass boost anyways. Best to mix and see what translates best.

One newb-ish question, I assume in this 4 subs configuration, it's a single low frequency channel, like you'd have in 2.1 configuration, distributed to 4 subwoofers, right ?

If yes, what's the general stand on using 2 subs ? Should they be the same channel as well, and not one the extension of the L and the other of the R ?

Yes, the low frequency energy is summed to mono and sent to all 4 subwoofers

There are multiple approaches to 2 subwoofers, but in the one that is connected to the Harmon/JBL research which recommends 4 as ideal ,you would also sum them with the same signal. In this case the best results are often achieved by having one at the midpoint of the front wall and one at the midpoint of the back wall.

Other approaches include having them under the mains facing each other or treating them as separate signals so that L and R mains and subs are full range, discrete signals

Hey, thanks for the fast reply !

I got one more question meanwhile, how do you daisy chain the LSR310s ? I don't see LFE outputs on it ?

Thanks !

in our configuration we are turning off the on-board processing and using an outboard DSP so we have a direct line out to each sub. We have tried some other models but this is the only mid priced one we could find that actually lets you bypass the onboard processor - this is important for fine tuning the DSP side of things

In some cases just having the 4 subs in the correct arrangement does wonders for smoothing room modes, but we like to take it further and adjust spl, phase, eq etc.

What outboard DSP are you using?

Only a few years ago, I had the same opinion … but I was wrong.

There are some situations where you actually can reduce modal decay using DSP correction, especially if lower order axial modes:

How to tame a resonance at 22hz

In this example, we deliberately choose to ignore the first axial mode (in terms of treatment) and focus on higher order modes (that where more problematic) and then managed to fix the 1,0,0 mode with a filter that matched the Q and gain of the mode.

Having said that; I would naturally still opt for as much treatment as possible before reaching for that DSP …

I´m not a big fan of multi sub systems either (if “multi” means more than two). The best sounding systems I´ve heard has always consisted of two subs in the front (below the mains, carefully integrated using DSP) in a well-treated room.


Sincerely Jens Eklund

Nice one.

I can see how if there is only one well identified variable that could work to a certain extent without too much collateral damage, especially on a sub. More than that and you'll see me slowly frown.

But yes, design & treatment first. We actually never use DSP correction anywhere, but the ATC amps allow fine tuning of phase (crossover) and Woofer/Midrange/tweeters levels. Which is plenty enough to get the system to perform 100% in the rooms.

What struck me with even recent DSP correction systems and DSP controlled monitors, is that as soon as it does more than one wee little thing, it has audible artefacts - that are often very hard to nail on a measurement but are audible. Some systems, turning DSP on without DSP doing anything yet is audible. Sometimes it's like a weird veil, a shift in stereo and depth, loss of realism of the soundstage.

The only guy I know of that has managed to create a purely DSP driven speaker system that really works and sounds amazing is Bruno Putzeys, with Kii Audio. (Also the man behind the Hypex Class D amps and Grimm speakers.)
He also happens to be one of the most intelligent human being I have met. I have a lot of respect for his work.

These are also cardioid speakers - and not only on paper. It works very very well. In untreated rooms, it helps with a lot of SBIR related issues.

Kii Audio GmbH - High End Active Speakers for HiFi and Professional use.
Kii Loudspeakers: for studio

If not careful, then yes; correction can easily make things worse, especially if making individual correction to L & R channel (something I tend to avoid) that will often result in what you describe. But if done correctly and carefully; I only see (and hear) benefits of gentle correction (assuming well treated room so you don’t need massive amounts of correction in the first place).

What needs to be remembered however, is that if you make corrections that improves the response in the sweet spot area, the situation often degrades in other areas (naturally), so it´s a balancing act. Most (if not all) of my clients favor a better response in sweet spot area, even if the response in other areas degrades. But so far all of my clients care less about how it sounds elsewhere in the room as long as it´s perfect in the sweet spot. I would make the same decision.


Sincerely Jens Eklund

Thanks !

So, if I understand well, the only way I could run two subs, without external dsp processing, is by connecting L and R outs from my interface to the L-input of the first sub and the R-input of the second sub, and then connecting left satellite to one and the right satellite to the second sub, right ? And in that case, the only way I could place them is in front ?

I don't see how I could distribute the same mono LFE signal to 2 subs without external processing.

Does that also mean that I am dependant on LSR310's crossovers, meaning that I won't be able, for example, to run the satellites full range and just add the subs for the lowest octaves ?

Thanks !

As soon as I heard Digital I immediately abandoned Analogue. I bought a Sony F1 system. At one point I decided to question my preferences and did an actual blind ABCD test. The AMS AudioFile, yes a Dalek, and a Technics DAT using complementary AD/DA, beat very high end analogue, Dolby SR included.
Half a dozen people concurred on the preference, although the actual sonic differences were minuscule, probably not measurable. The resulting album became a test CD for the Hi Fi internationale.
For years I used George Massenburg's analogue Eq's whenever budget allowed.
Until the software came along. It sounds better.

Minimal circuit path will always sound better. KM84 vs 184. The ultimate being simply wire. So if the job to be done can be done without DSP, great. But DSP can do things wire and iron and electrolytes cannot.
There is no comparison possible with Altiverb, DMG, Sonoris, Dirac Live, Trinnov.

To the topic. The Bag End and PSI active traps, the large number of speakers in the Dirac Live scenarios, some DIY approaches here on GS https://gearspace.com/board/7473483-post17.html
I think it's about time to lose the speaker cone and drive the wall pistonically.

DD

We have tried a few different ones. The good ones are transparent and the bad have a sound to them

For example the super cheap Behringer DCX2496 changes the sound of the mains a bit for the worse, buts the subs are fine.

The high end boxes, mostly designed for live sound, really are transparent and work great.

We listen through DSP all the time, it's only the studio market that resists this idea, live sound, live theater, professional cinema, high end home theater, etc are way down the road on this and it sounds great. When you use a DAW, use plug ins, listen through most active studio monitors, go to a live show, etc you are listening through DSP

Well Steve, hopeful the CR and those in it are the best qualified to assess sonic issues. Dirac Live had a couple of issues which became clear only in the Pro environment. e.g. Headroom is stressed by the high levels that occur in Mastering. Speaker position correction by delaying one side simply doesn't wash given the vast preponderance of Mono in a typical stereo mix.
One hardware box has no facility to disengage this delay, and even has a very audible 1.42K tone constantly.
When Massenburg software moved to AAX format, bypass caused full level clicks. FS.
Another App I am testing has L and R reversed.....
No harm in some healthy scepticism.
DD

Completely agree!! This is exactly why we have been testing so many different subs, DSP boxes, placements, etc. One thing I can tell you for sure is that great DSP systems sound great and not all are great!!

Self powered pro monitors and subs often use on-board DSP to implement their crossovers, phase alignment, eq, etc. Just like with rooms, proper design and materials should be the primary approach, but there is nothing wrong with using DSP as long as it doesn't introduce issues. The examples you gave above are terrible, sounds like some deaf people are responsible for those audio products!

You, Thomas and others are successful working at a high level and can achieve world class results using construction design and advanced acoustic treatment strategies. There is no doubt that what you guys do is the Gold Standard, my musings and interest in much more focused on working with an existing room, moderate budget and making it the very best it can be.

I hope one day I can hear the rooms you and Thomas have created!

Hey Steve, have you tried some of these miniDSP boxes ?

I see a lot of home theather guys are praising it ?

There is plenty of high end really bad analogue electronics out there too.
Neumann's current mics have better spec and seem more robust, better built, than ever. But the sound of the capsule simply does not make it to the XLR. Multiple active and passive components. The U47, U67, KM, Capsule, one active stage, one transformer. Less is more.

I Beta test. I have the Inverted Midas Touch. Everything I touch turns to s...
Dirac Live has the speaker position correction. On by default. I have been encouraging them to reconsider that. But this is really only audible in the CR. It is a vital feature in a car.
Unfortunately this feature cannot be disengaged when Dirac is implemented in the Minidsp box. Nor can it be viewed or turned off during the analysis procedure while creating the actual filter.

Essentially all of the examples I gave were of great people designing great products. Many of them have no idea what goes on in a CR though. Often, they come out with the product first, then address the Beta issues. Which in hardware, is too late. It would be harsh to call them deaf, but they should definitely PFL..... Pre Failure Listen!

This also applies to Speaker and CR designers. There is a huge reluctance to accept that Flat won't translate.

I have worked and continue to do so at a high sonic level in Recording/Mix/Mastering. Way back in the days or proper Hi Fi, Mary Black kinda took over from Ricky Lee Jones, internationally.

During the glory days of Recording many of us were repeatedly frustrated by pervasive failure of translation in commercial studios. We brought our own speakers in. We kept checking on headphones, in the car, on the home Hi Fi.
Ultimately I noticed the problem was being caused by the solution. i.e. The treatment and tuning of the response. I studied acoustics personally, followed by formal training. I became an Acoustics and Noise Control Consultant. I have designed only one green field studio. But have treated or re treated many many. Some commercial, most ProSumer. Much of my work is on the other side of the coin, Noise Control. Or remedial Acoustics. Many of my clients have buildings which are unusable because of the Architects' utter lack of understanding of the most basic acoustics. You cannot have a public event or even a meeting in a space which sounds like a swimming pool. One of my current clients has a vast lovely looking cafe/meeting area. Bare wood glass etc. This is the HQ of the Guide Dogs for the Blind. Can you imagine trying to communicate in noisy swimming pool with no visual clues.
This type of work, as opposed to the Arts, I call a Toilet Paper Industry. Something real people need and use every day.

There is no real similarity or equivalence to the work of my friend Thomas.
I have never heard Thomas' work yet, but based on discussions on fora and personally, I have recommended him to friends, albeit unnecessarily. Sooner or later I will visit Brussels or LA, but I have been promised a 'preview' remotely. Using Binaural Recording and, well, DSP......
DD