LS50 meta's plus sub vs More expensive speakers

[abdo123]

Correction : Earl not Tom

I tend to agree with that too. We may need to come back to thins but that phase issue in the bass could be a non-issue.. Point for another discussion. in the bas the room is minimum-phase .. In simple termsif you get a smooth FR .. It is mostly solved. behavior in music is similar to steady-tone.. In the low bass .
On this
Time to fire REW and try to "Geddes" my current room ... Again

Thanks for the correction, I haven't fully 'Geddes' my room yet too as I still question whether i should leave the subwoofers closest to the main extending towards the 'statistical region' of the room.

my brain is telling me that will definitely sound wrong unless the sub was verically aligned with the woofer of my bookshelf.

What do you think? did you try doing that?

 

[Ultrasonic]

Correction : Earl not Tom

I tend to agree with that too. We may need to come back to thins but that phase issue in the bass could be a non-issue.. Point for another discussion. in the bas the room is minimum-phase .. In simple termsif you get a smooth FR .. It is mostly solved. behavior in music is similar to steady-tone.. In the low bass .
On this
Time to fire REW and try to "Geddes" my current room ... Again

Get the relaitive timing of two subs and the main speakers wrong and you'll get a cancellation dip in the frequency response in the crossover region. As above, I think it's important to distinguish between two subs and lots of subs.

 

[Ultrasonic]

It may help if I just add something. I don't believe that I'm disagreeing with Earl Geddes about anything but rather that his excellent research is inappropriately being extrapolated to a two-subwoofer situation here.

 

[abdo123]

It may help if I just add something. I don't believe that I'm disagreeing with Earl Geddes about anything but rather that his excellent research is inappropriately being extrapolated to a two-subwoofer situation here.

I didn't want to initially respond as you used a straw-man argument that practically cannot even happen in real life.

Anyway lets backtrack a bit.

you mentioned that theoretically, two subwoofers can be completely out of phase to the point that turning one off would be better off for the listerner.

Well for that to happen, you need to place both subwoofers in a corner or in a symmetrical location (Something Dr. Geddes does not recommend) and for you to sit smackdown in the middle of the room, and then, intentionally reverse the polarity of one of these subwoofers.

and even then, this would only significantly effect the frequency response below the first mode, which is usually at 30 to 40Hz in most rooms, and at that range, most if not all music have no fundamental notes there.

What about below the modal region, i.e. in the first mode region? One cool thing is that subs sum coherently (in-phase) below the modal region. So while small subs may not individually have very much output, efficiency goes up tremendously through the use of multiples.

basically, there is no way someone would get a worse sound out of two subwoofers, placed randomly in a room, unless they intentionally go out of their way for it to sound bad.

 

[Ultrasonic]

I didn't want to initially respond as you used a straw-man argument that practically cannot even happen in real life.

Anyway lets backtrack a bit.

you mentioned that theoretically, two subwoofers can be completely out of phase to the point that turning one off would be better off for the listerner.

Well for that to happen, you need to place both subwoofers in a corner or in a symmetrical location (Something Dr. Geddes does not recommend) and for you to sit smackdown in the middle of the room, and then, intentionally reverse the polarity of one of these subwoofers.

and even then, this would only significantly effect the frequency response below the first mode, which is usually 30 to 40Hz in most rooms, and at that range, most if not all music have no fundamental notes there.

basically, there is no way someone would get a worse sound out of two subwoofers, placed randomly in a room, unless they intentionally go out of their way for it to sound bad.

I really didn't present a strawman argument at all, and I gave a worst case example to make the point that the timing matters. 'Random' positioning of two subwoofers won't somehow magically ensure perfect phase matching at the main listening position, so it's a question of how good/bad will it be on a spectrum of perfectly good to perectly bad.

Do just take a moment to consider my full range speaker example. Let's assume a multi-way design that means the lowest frequency driver only operates below 100 Hz. Would you argue it makes no difference if these drivers are run out of phase, and if not, why not?

(For info. I have a research backround in acoustics, but as my username suggests this is at a rather different frequency range to what we're discussing here! I'm always happy to learn by being proved wrong but I really don't believe I am here.)

 

[ROOSKIE]

After using two subs in stereo crossed often higher than typical I will never go back to one - ever.
With music I am not worried about sub 30hrz bass nor the best positioning for such bass.
I am also not worried about maximum output as two or more subs for music can handle just about everything even in slightly sub optimal locations (for even high output levels.)
You might push the subs back against the wall and keep the monitors forward.
I over lap the frequency range quite a bit - this helps fill in dips and to some degree nulls - then I mitigate any resulting peaks via DSP/PEQ.
Finally if necessary I time align the woofer in the DSP (a inexpensive miniDSP 2x4HD easily can do all of this)
The time alignment is not as needed in a stereo set-up of subs as long as the subs are close to the monitors (with a few feet) actually you can skip it if you want but since the miniDSP can easily do it might as well.

I often listen at higher volumes. This is where crossing a small speaker like the LS50 really helps. With similar speakers using stereo subs I will usually cross at 120-175hrz and fully overlap the sub 1 octave. (based on measurements not guessing and always using 24db slopes again measured not assumed)
This measurably reduces distortions and really sounds good. I have to say that I would bet I could easily pass a blind test, the many monitors I have tested monitors sound much cleaner and stress free almost all of them have)

Anyway, no way to get the best sound without at least REW(free) and UMIK USB mic ($80-100usd). Measure your work no matter what set-up you end up with.

 

[abdo123]

I really didn't present a strawman argument at all, and I gave a worst case example to make the point that the timing matters.

That's literally the definition of a strawman argument. read more here:

Essentially, the person using the strawman pretends to attack their opponent’s stance, while in reality they are actually attacking a distorted version of that stance, which their opponent doesn’t necessarily support.

'Random' positioning of two subwoofers won't somehow magically ensure perfect phase matching at the main listening position, so it's a question of how good/bad will it be on a spectrum of perfectly good to perectly bad.

Like i mentioned before, the modes of the room determine the response above the first mode, and the phase of the subwoofers determine the overall output (complete cancellation, complete summation or somewhere in between) only below the first mode.

There will be a huge overall loss of output of course, but the error rate of the response should remain largely the same.

This still doesn't matter because there will never be a situation where the two separate subwoofers in a room are completely out of phase unless you go out of your way and set it that way.

Anyway just to entertain that fantasy, I looked up some old measurements of mine where I know the subwoofers are time aligned at the measurement position, and i used the alignment tool in REW to calculate an aligned sum (simulation) in which one of the subwoofer has inverted phase.

 

[FrantzM]

Hi

I apologize in the name of my fellow ASR to the OP for veering the discussion toward subwoofer integration.
It is however an important, even determinant issue in home audio reproduction. You get the bass right and you are 30% there

Get the relative timing of two subs and the main speakers wrong and you'll get a cancellation dip in the frequency response in the crossover region. As above, I think it's important to distinguish between two subs and lots of subs.

Could be true but a cancellation dip may be inaudible, it depends on the Q of that dip, teh Frequency of the dip itself dependent of many factors...
The relative timing of the subs can also be quite different. A reversal of phase of one subs , could well result in better (smoother and more extended) bass FR within the listening volume. @abdo123 post 47 is an interesting illustration
That leaves in my mind no doubt. More subwoofers are better than one. @ is the minimum but 3 optimum. Above 3 .. potential very small improvements perhaps not worth the hassle.
By the way the 3 subs do not have to be the same. Geddes preconize a strong, one in a corner, the two others on the other walls. Currently playing with 2 (cheap) subs with one in the back.. I can't for the life of me localize the one in the back when music or movies play, yet its level is rather high about .5 less than the Front subwoofer in the Audyssey App. ... Audyssey did a great job... I find however on careful listening that I am missing some things in the 100 to 150 Hz region.. I will later return to a Geddes-style 2-sub arrangement using the miniDSP 2 x4 HD and presenting this to the Audyssey as one subwoofer.

@abdo123.

I don't think it matters. The vertical alignment won't make much of a difference. Once the bass is smooth at the listening position.. One take it and align the other speakers to that combined subs. That is where a good AVR helps.. I have the suspicion that if the bass is properly handled, that is properly smoothed and EQ'd, Dirac or Audyssey should provide very similar results in the system
I am off to the Geddes thing.. Waiting for the noise ( I live in the city ) to abate.

Peace

 

[Ultrasonic]

@abdo123 I'm considering a stiuation with subwoofers run to frequencies significantly above he lowest room mode, which I'm sure most reading this thread had in mind for the OP. Also 'huge overall loss of output' matters. A lot. Possibly it's this last point that means we're most disagreeing actually. Note my original posts related to what is optimal and throwing away headroom / increasing distortion is absolutely something I'd want to avoid.

 

[Ultrasonic]

View attachment 148788

Exactly what FDW has been applied above will influence what we see but let's park that. How are you defining 'in phase' and 'out of phase' for your above example? And what does the graph look like up to say 100 Hz?

Fundamentally though I'd argue that this graph proves my point that relative timing matters, since your two combined curves are very different from each other.

 

[abdo123]

How are you defining 'in phase' and 'out of phase' for your above example?

matching Impulse response.

And what does the graph look like up to say 100 Hz?
.

The infinite cancelation moves up in frequency, there is an output penalty.

 

[Ultrasonic]

I'm conscious this discussion is probably becoming less useful to @Eternals so I'm going to do my best to reply in a way to try to help them...

Where this current discussion started was my suggestion that effectively using two subwoofers is more complicated than just setting up one. I stand by this but @abdo123 has been suggesting otherwise, with reference to the work of Earl Geddes. Below is a YouTube video where Earl discusses some of this, which @abdo123 posted on another thread recently. At heart of the method proposed is the use of a software optimisation to calculate a frequency dependent amplitude correction for each subwoofer. This is additional complexity of the type I was referring to, and requires hardware to be able to implement this (such as the excellent miniDSP 2x4 HD mentioned earlier in this thread). The most commonly used, and free, software I'm aware of to do something similar is Multi-Sub Optimizer. I have used this but it was a couple of years ago now.

Getting into slightly more detail, Earl basically argues in the video that his experience is that he has been able to get decent results by applying only amplitude adjustments to the signal to each sub, and to not worry about initial time-alignment or phase matching considerations. I would argue that it would be superior to do so, and I think it's relevant to note that doing so today is considerably easier then when Earl carried out his original research. Essentially in situations where signals from different subwoofers would cancel to a significant degree, I'm sure the optimisation programme will limit the output of the sub with the lower output and make the other basically do all of the work. If you have enough head-room then you can get away with this but it's at best inefficient, both in output terms and financially.

There is also the question of subwoofer placement, which whilst Earl tries not to be too prescritpive about he does clearly acknowledge can make a difference. Most obviously if the response from one sub has some significant dips in the in-room frequency response then ideally you'd like the second sub in a location that doesn't have the same dip. The more subs that are used then the more likely this will happen by chance but I'd strongly suggest there is a smarter way to do this .

Finally I'll just comment that the video example only uses amplitude adjustments to sub outputs to optimise the response. It is perfectly possible to also calculate and apply (with the right hardware) frequency-dependent phase shifts to further improve the situation. This is yet further complexity though.

So as I tried to say initially, I very much understand and appreciate the potential benefits of multiple subwoofer use but I don't accept that there isn't additional complexity to doing so optimally.

 

[abdo123]

I'm conscious this discussion is probably becoming less useful to @Eternals so I'm going to do my best to reply in a way to try to help them...

Where this current discussion started was my suggestion that effectively using two subwoofers is more complicated than just setting up one. I stand by this but @abdo123 has been suggesting otherwise, with reference to the work of Earl Geddes. Below is a YouTube video where Earl discusses some of this, which @abdo123 posted on another thread recently. At heart of the method proposed is the use of a software optimisation to calculate a frequency dependent amplitude correction for each subwoofer. This is additional complexity of the type I was referring to, and requires hardware to be able to implement this (such as the excellent miniDSP 2x4 HD mentioned earlier in this thread). The most commonly used, and free, software I'm aware of to do something similar is Multi-Sub Optimizer. I have used this but it was a couple of years ago now.

Getting into slightly more detail, Earl basically argues in the video that his experience is that he has been able to get decent results by applying only amplitude adjustments to the signal to each sub, and to not worry about initial time-alignment or phase matching considerations. I would argue that it would be superior to do so, and I think it's relevant to note that doing so today is considerably easier then when Earl carried out his original research. Essentially in situations where signals from different subwoofers would cancel to a significant degree, I'm sure the optimisation programme will limit the output of the sub with the lower output and make the other basically do all of the work. If you have enough head-room then you can get away with this but it's at best inefficient, both in output terms and financially.

There is also the question of subwoofer placement, which whilst Earl tries not to be too prescritpive about he does clearly acknowledge can make a difference. Most obviously if the response from one sub has some significant dips in the in-room frequency response then ideally you'd like the second sub in a location that doesn't have the same dip. The more subs that are used then the more likely this will happen by chance but I'd strongly suggest there is a smarter way to do this .

Finally I'll just comment that the video example only uses amplitude adjustments to sub outputs to optimise the response. It is perfectly possible to also calculate and apply (with the right hardware) frequency-dependent phase shifts to further improve the situation. This is yet further complexity though.

So as I tried to say initially, I very much understand and appreciate the potential benefits of multiple subwoofer use but I don't accept that there isn't additional complexity to doing so optimally.

I think this is only worth the effort when there is one listening position to optimize for, but usually multiple subs are used to optimize for a wide spaces or multiple seats. whatever point of reference you use in space or time you're measuring, you're optimizing the phase in that spot only.

However if someone buys me a Dirac live bass control license where the 'average phase' is calculated at the crossover frequency for all the seats (it's quite cool, check Erin's review with the Dirac employee) then i would likely change my mind. but right now it's too much effort to optimize the phase for multiple seats, and the point behind multiple subs is to optimize the response for multiple seats.

 

[Eternals]

I'm conscious this discussion is probably becoming less useful to @Eternals so I'm going to do my best to reply in a way to try to help them...

Where this current discussion started was my suggestion that effectively using two subwoofers is more complicated than just setting up one. I stand by this but @abdo123 has been suggesting otherwise, with reference to the work of Earl Geddes. Below is a YouTube video where Earl discusses some of this, which @abdo123 posted on another thread recently. At heart of the method proposed is the use of a software optimisation to calculate a frequency dependent amplitude correction for each subwoofer. This is additional complexity of the type I was referring to, and requires hardware to be able to implement this (such as the excellent miniDSP 2x4 HD mentioned earlier in this thread). The most commonly used, and free, software I'm aware of to do something similar is Multi-Sub Optimizer. I have used this but it was a couple of years ago now.

Getting into slightly more detail, Earl basically argues in the video that his experience is that he has been able to get decent results by applying only amplitude adjustments to the signal to each sub, and to not worry about initial time-alignment or phase matching considerations. I would argue that it would be superior to do so, and I think it's relevant to note that doing so today is considerably easier then when Earl carried out his original research. Essentially in situations where signals from different subwoofers would cancel to a significant degree, I'm sure the optimisation programme will limit the output of the sub with the lower output and make the other basically do all of the work. If you have enough head-room then you can get away with this but it's at best inefficient, both in output terms and financially.

There is also the question of subwoofer placement, which whilst Earl tries not to be too prescritpive about he does clearly acknowledge can make a difference. Most obviously if the response from one sub has some significant dips in the in-room frequency response then ideally you'd like the second sub in a location that doesn't have the same dip. The more subs that are used then the more likely this will happen by chance but I'd strongly suggest there is a smarter way to do this .

Finally I'll just comment that the video example only uses amplitude adjustments to sub outputs to optimise the response. It is perfectly possible to also calculate and apply (with the right hardware) frequency-dependent phase shifts to further improve the situation. This is yet further complexity though.

So as I tried to say initially, I very much understand and appreciate the potential benefits of multiple subwoofer use but I don't accept that there isn't additional complexity to doing so optimally.

Ultrasonic, thank you for pulling that together. I'm on a steep learning curve wondering what the energy, effort and cost would gain for me. I can appreciate that there are benefits, but as you've advised, I'm using the one fixed listening seat, so less of an issue. You've really helped me, thank you.

 

[Eternals]

I think this is only worth the effort when there is one listening position to optimize for, but usually multiple subs are used to optimize for a wide spaces or multiple seats. whatever point of reference you use in space or time you're measuring, you're optimizing the phase in that spot only.

However if someone buys me a Dirac live bass control license where the 'average phase' is calculated at the crossover frequency for all the seats (it's quite cool, check Erin's review with the Dirac employee) then i would likely change my mind. but right now it's too much effort to optimize the phase for multiple seats, and the point behind multiple subs is to optimize the response for multiple seats.

abdo123, thank you for that clarity. It is very reassuring. The future may be more complex, possibly. But for now, the simple single sub option looks promising. I am very grateful.

 

[Eternals]

The Hegel is a great integrated amplifier and will last you years, don’t have buyers remorse! Since it doesn’t have any room correction you should get a sub that does. Lastly just remember people have been enjoying music forever with far lower level equipment than you and this is a bit of an elitist group in that the knowledge base here is the 1% of the 1%.

TankTop, as well as from your username being fun, your summation is incredibly valuable. You're certainly right about the level of expertise. This has been a swan dive from my amateur ideas into a world of PhD level consultations. I really am very grateful.

 

[Ultrasonic]

I think this is only worth the effort when there is one listening position to optimize for, but usually multiple subs are used to optimize for a wide spaces or multiple seats. whatever point of reference you use in space or time you're measuring, you're optimizing the phase in that spot only.

I'd agree that I think a key strength of using multiple subs is to give a move even response over multiple seats. Any approach will be able to give better results over a smaller region that a larger one though. As @Eternals' focus was music I was taking it that single seat optimisation was the priority but perhaps they could clarify?

However if someone buys me a Dirac live bass control license where the 'average phase' is calculated at the crossover frequency for all the seats (it's quite cool, check Erin's review with the Dirac employee) then i would likely change my mind. but right now it's too much effort to optimize the phase for multiple seats, and the point behind multiple subs is to optimize the response for multiple seats.

I'm sure it's possible to do anything that DLBC does manually if you're sufficiently interested . I do use Dirac Live (but not DLBC) and still somewhat struggle with the lack of full control compared to doing everything manually as I have before. What I have learnt through experience though is to move away from focusing on single point optimisation to some sort of volume average, even when considering just a single listening position. I can get an essentially completely flat response at a single point but this isn't what works best.

 

[Eternals]

I'd agree that I think a key strength of using multiple subs is to give a move even response over multiple seats. Any approach will be able to give better results over a smaller region that a larger one though. As @Eternals' focus was music I was taking it that single seat optimisation was the priority but perhaps they could clarify?



I'm sure it's possible to do anything that DLBC does manually if you're sufficiently interested . I do use Dirac Live (but not DLBC) and still somewhat struggle with the lack of full control compared to doing everything manually as I have before. What I have learnt through experience though is to move away from focusing on single point optimisation to some sort of volume average, even when considering just a single listening position. I can get an essentially completely flat response at a single point but this isn't what works best.

Thanks @Ultrasonic I have only a single listening seat and looking at the SVS SB 1000 pro (or similar) to augment a pair of KEF ls 50 meta's. My room is classed as small according to Audioholics, so may need a smaller sub. My thoughts are to a potential future with a double subwoofer set up, subject to my confidence with learning enough to implement the miniDSP arrangement.

 

[Ultrasonic]

Thanks @Ultrasonic I have only a single listening seat and looking at the SVS SB 1000 pro (or similar) to augment a pair of KEF ls 50 meta's. My room is classed as small according to Audioholics, so may need a smaller sub. My thoughts are to a potential future with a double subwoofer set up, subject to my confidence with learning enough to implement the miniDSP arrangement.

There isn't really any such thing as a sub being 'too big' in performance terms, just physically and financially. An SB1000 isn't a big sub anyway .

Were you thinking of buying a miniDSP 2x4 HD now, or considering it for the future? If you're not going to buy one now then the (limited) EQ capability of the SB1000 Pro will be beneficial, but if you will have a miniDSP then it would be essentially pointless and you may want to consider alternative subs where part of your money isn't going to pay for this. If you mentioned where you live then possbily someone here familiar with the local market may be able to make a suggestion or two. If you're buying new then a demo or sale-or-return option may be a good place to start as well.

 

[abdo123]

Thanks @Ultrasonic I have only a single listening seat and looking at the SVS SB 1000 pro (or similar) to augment a pair of KEF ls 50 meta's. My room is classed as small according to Audioholics, so may need a smaller sub. My thoughts are to a potential future with a double subwoofer set up, subject to my confidence with learning enough to implement the miniDSP arrangement.

If you're planning on adding more subwoofers in the future then going sealed subwoofers (the SB, instead of the PB) is an excellent choice indeed, while ported subwoofers provide more output, the output is 'slower' sort of speaking. but two or three SB 1000 pro would provide a very very respectable experience for music.