You're
sort of right - at least
some of the time - although it's a bit more complicated than that... and it also depends on the "target
AREA" you're considering.
In a "perfect simple world" you would in fact be wrong. In that perfect world, if you individually time-aligned sub #1 with your left front main, then you time aligned sub #2 with your left front main,
they would in fact also be time aligned with each other. It's basic math; if you time align #1 with A, and you time align #2 with A, then #1 and #2 must also be time aligned with each other. (If A=C and B=C then A=B.) That being the case, they will add perfectly and, if they are individually correct, then the combination will also be correct. (
READ THAT AGAIN - CAREFULLY. If two speakers that are indeed time aligned to the same reference - and so to each other - are both corrected to flat, then, when you add them together, the resulting combination will also be flat. There may
ALSO be other combinations that add up to flat. And, as long as both are time-aligned relative to your listening position, there is no specific requirement that they be arranged symmetrically.)
To elaborate on that with a simple example. Put a chair in an empty field (no reflections). Put two identical subs the same distance from the chair.
AT THAT CHAIR those two subs will be time aligned - they will be perfectly "in phase" at every frequency. One could be located to the left of the chair, and one directly behind it - not symmetrical but still time aligned. (This will be true at that chair; the arrangement may in fact be terrible when considered from other listening positions.)
Unfortunately, in "real life" there are several problems already:
1) Whenever you time align
ANYTHING, you can only do so for one specific observer location. The only possible way two subs could be "correctly time aligned" for all listeners at
ALL possible listening positions would be if they were located in the exact same spot (in three dimensions). Again, it's basic geometry; if you want to pick two spots
that will be the same distance as each other from every other point in the room then it's pretty obvious they
MUST be located at the same point in space. Therefore, you can
only actually time-align two or more subs correctly for one specific listener position.
2) We aren't considering things like room modes, and reflections, and other "higher order effects". Even considering a single listening position, it's quite possible that you could have cancellations or reinforcements due to reflections interacting with the "main signal", or even with other reflections...
However, so far I haven't mentioned the 800 pound gorilla in the room..... namely that nobody really wants to adjust their room for a single listening position. And, once you accept that you really would like to make things correct at more than one specific location in the room, things become a lot more complicated. Now you're dealing with averages, and compromises, and variations from ideal.
Now that you've read this far I'm going to get to the whole point I was leading up to....
All of the various "methods" for figuring out how to position and correct multiple subs are
COMPROMISES. And all compromises end up leading to choices. The reason why the various methods seem complicated, and why there are in fact so many different methods, is that every one of them is a compromise. Unlike what many people seem to think, it's
NOT like: "Gee, why aren't they doing it the right way?" Rather, it's really: "There is no way that is entirely right; but, in my room, with my particular requirements, method XYZ happens to work best for me."
NONE of the various "procedures" you're talking about works perfectly well, for everyone, in every room, all the time. (Although, to flip that around, many of them may work pretty well for most people most of the time.)
My point in all this is that individually correcting each sub
SEPARATELY, then simply adding them together (turning them both on) is a perfectly valid way of correcting them - which is why it works pretty well for most people. There are all sorts of
other ways of doing it, and some of them may work better for some people, in some rooms, but that doesn't mean that the simplest way is wrong... it also doesn't mean that those other ways, even though they may work especially well in certain situations, are "best" - and what that even means is going to depend on what choices you make in terms of what compromises
YOU prefer to make - which will depend on your goals and requirements.
I'm going to throw you one more "simple" example.... which will illustrate this.....
Let's assume you have one huge sub that has excellent frequency response and oodles of low bass output. You put that sub in the only location in your room where it fits (and where the floor has been properly reinforced). However, when you measure things, you find that there's a major room mode at 43 Hz - the room booms like a big empty box at that frequency. Now let's assume you find yourself another, smaller, but still decent sub. What's "the correct" way to set up those subs (or 'the best way") and fix that room? Well, we could simply use our EQ to turn down the peak at 43 Hz (and leave the second sub packed up). That way we don't have to worry about interactions between them. We could also find a good spot for sub #2, EQ each of them separately in their current locations, and hope that they add together in such a way that they complement each other. Another way would be to leave our big sub without EQ, then position the second sub at the point in the room where that 43 Hz energy builds up (maybe a convenient corner), and then through that second sub play
ONLY a band of frequencies around 43 Hz - specifically calculated and phased so as to cancel out the peak at that frequency. (It's called an active bass trap; Bag End makes them - and maybe some other vendors). And, of course, there are endless other combinations of where we could place our two subs, and how we could EQ each to work well together. And we didn't mention passive bass traps yet....
As for which of those solutions is "better" or "right"...... well...... (The active bass trap method will probably give you the cleanest bass, but will also probably give you the
LEAST bass output overall, since you're actually subtracting the output of one sub from the other. Equalizing each sub separately to be as flat as possible is probably a good compromise. If the big sub is the much more powerful of the two, then EQing it first, and then EQing the smaller one to "fill" any significant gaps or notches might also work well. That would depend on the specific characteristics of each.) My point, however, is that there is no one "right"... because even the best compromise is going to depend on your requirements (cleanest bass; most bass; most reasonably clean bass; etc). I'm certainly not discouraging anyone from trying the miniDSP, or the "swarm of little subs" method, or the "two wall centers and a corner method"; I'm sure all of them work the best in at least some situations, and I have no idea statistically which one "works best most of the time". But, as it so happens, the way we do it is to EQ each sub separately - and that way works for a lot of people. And Dirac Live happens to do a very good job of getting that right - which makes it work even better. And, unfortunately, we can't easily accommodate other methods (which is probably not such a bad thing because, if we could, then we'd have to decide which other methods to support). Variety is a good thing; and the world is complicated; which is why we have cool little boxes like the miniDSP, and Bag End active bass traps, and DEQX processor boxes that let you EQ and individually time-align six individual channels.
(Oooops... I almost left out one important detail..... Since virtually all filters, and all speakers themselves, have phase responses that vary with frequency, all this talk of time aligning things also may only be true at one
FREQUENCY. If you have two different subs, even different but similar models, odds are that their phase plots will be different - so you can only time align them correctly with each other, relative to one position,
at one frequency. Of course Dirac is able to correct that somewhat - because it can adjust phase response - but any simple "add delay to align the times" solution won't do it... because, whatever time adjustment you make, it will only be "correct" at one frequency anyway.)
The issue is that if you eq each sub separately and then combine them together it will not result in a flat response unless the subs are positioned symmetrically. If you have 4 subs like many do in dedicated rooms they need to be time aligned then eq'd.
).
