[maty]

Well, there are a good compromise: passive speakers with very good selection of speakers with great slope => first order crossover.

If we know all the parameters of the speakers, frequency response, phase, impedance of the box ... with minimum impedance > 4 Ohms, better.

And if they are small coaxial, much better (or tweeter with a good waveguide).

So I bought the "cheap" KEF Q100  They are sounding incredible after my cheap improvements.

In a BIG room => we need two (or three) 12/15" active subwoofers (with adjustable phase 0...180º) at 180 - 200 Hz.

[undertow]

Well, there are a good compromise: passive speakers with very good selection of speakers with great slope => first order crossover.

If we know all the parameters of the speakers, frequency response, phase, impedance of the box ... better.

And if they are small coaxial, much better (or with a good waveguide).

So I bought the "cheap" KEF Q100  They are sounding incredible after my improvements.

In a BIG room => we need two (or three) 12/15" active subwoofers (with adjustable phase 0...180º) at 180 - 200 Hz.

I agree which was the point of my post. You can achieve excellent results with a 2 way passive design, and match if not exceed active systems as long as you use a fairly high efficiency driver selection such as horns even running the Mids, and Tweets off a good passive filter, and if it has a crossover point keeping everything above 50 hz, with a solid efficiency of 100 db that is not that difficult to achieve running only mid and tweet drivers. You can also easily drive them to insane volume pressures that active systems of 120 db achieve using a simple 50 watt stereo amp with even better quality with manageable distortion.

A key I will say is that keeping your main amp from having to drive low impedance, and under 50 hz will open up a HUGE gap missing in most systems. Any amp that is taxed with driving under 50 hz has to double down on almost everything from current to power, and the impedance drops from the woofer can, and will as well diminish performance. So running stereo subs, with a high efficiency 2-way that does not have to go to sub bass levels will make a lot of people happy before going down the rabbit hole of trying to actively tri-amp an entire system for similar if not less results in most cases.

[DaveC113]

I want to hear what perfect zero phase response top to bottom sounds like.

IMO, it's not a big deal, our ears aren't sensitive enough for it to matter outside of the midrange. Also IMO, speakers tend to start sounding worse when DSP is used above the bass frequencies.

It's hard to argue that DSP isn't extremely useful below the transition frequencies, but it's also hard to argue it's worthwhile above that if the speaker is designed properly to begin with. Speakers like the M2 are different in that they are required to have specific performance levels in certain areas like frequency response and dispersion pattern, which is greatly facilitated by the use of DSP, in fact the M2 absolutely requires it to perform properly. Personally, I don't think the tradeoffs are worth it for a home system you aren't mastering on, I'd rather have a wide band midrange that is efficient and only requires a simple 1st order passive xo. As bad a rap as passive xos get imo they are better than DSP in most cases and there's a very good reason why active speakers aren't being embraced by audiophiles... the results aren't typically as good. Pretty simple, really...   

undertow, I'd agree but I think the main speakers need to reach fairly low otherwise subs are just a bad compromise for main speakers without enough extension.

[undertow]

DaveC113,

I agree, however I was specifically pointing to "Compete" with an active system. I know of very few passive speakers with a woofer dragging on the bottom end @ 20hz, 30hz, 40hz,  even with 1000 watts that could possibly be efficient enough to achieve outputs like an active system.

Hence to do it I think you would want to focus on a very efficient easy to drive 50hz, and up high impedance driver array plus subwoofers. But not saying good full range speakers won't do better running much lower than 50hz with the right amp. In anycase I also don't know of any passive 2 ways that would likely get much below 50hz either!

[macrojack]

I have a 2 inch co-axial compression driver hanging on the back of a large wooden horn. Underneath is a 5 cu.ft. ported cabinet with a 15 inch pro woofer. I used to operate this system actively with an XTA digital speaker management unit, a Halo JC-1 preamp, and a pair of Halo 23 power amps. I was unable to ever get this setup to perform to my satisfaction. That might well be due to my incompetence but it was a hassle nonetheless. So I got in touch with the guy who designed my system and had him design a pair of passive 3 way networks which were then assembled by the team at Yorkville sound. The results are purely wonderful sounding and the there is no complication at all. In fact, all the aforementioned electronics are long gone. Now I use a Yamaha A-S2000 integrated. Plug and Play. And more often than not I don't even use the tweeter.

[DaveC113]

I know of very few passive speakers with a woofer dragging on the bottom end @ 20hz, 30hz, 40hz,  even with 1000 watts that could possibly be efficient enough to achieve outputs like an active system.

Woofers usually aren't padded down and actually something like a 24 dB LR filter comprised of 2 inductors in series and 2 caps in parallel is actually pretty efficient. Zobels can eat some power but aren't a requirement for every woofer and don't reduce efficiency that much.

There are actually quite a few passive designs that will play low and be very efficient as well... pretty much any horn speaker, even ones with dynamic bass like JBL, Geddes, etc...

OTOH, if the speaker is using the woofer <400 Hz or so than I do think the speaker is a good candidate for a DSP/active woofer as the advantages of being able to tune the bass for the room and personal preference is nice, and our ears just aren't that sensitive in these frequency ranges so any shortcomings won't be noticed nearly as much as designs where the woofer is run up an octave + higher.

[roscoeiii]

Also IMO, speakers tend to start sounding worse when DSP is used above the bass frequencies.

Like much in audio, it depends on the implementation. I haven't yet tried an active set-up, but the DSP in my DEQX has greatly improved the sound of my system, from top to bottom.

[DaveC113]

Like much in audio, it depends on the implementation. I haven't yet tried an active set-up, but the DSP in my DEQX has greatly improved the sound of my system, from top to bottom.

In your system, with your speakers, I definitely believe that! But I wouldn't say that will universally apply to any system or speakers....

[undertow]

Like much in audio, it depends on the implementation. I haven't yet tried an active set-up, but the DSP in my DEQX has greatly improved the sound of my system, from top to bottom.

You could just use a good 15 or 20 band equalizer for cheaper and easier to use if this is the case. Your simply boosting frequency response, and contouring with gain actively "essentially", but still just running thru a passive crossover in your speakers. Really all your doing is correcting room response anyway, a better room is always the real answer, but impossible for most to pull off.

Everybody will argue added distortion, loss of purity, blah, blah, but truth is whatever makes you happy. I don't like putting band aides on 10,000 dollar preamps and amps though so luckily since my days of cheap receivers, and equipment no EQ is needed.

[maty]

I have a 2 inch co-axial compression driver hanging on the back of a large wooden horn. Underneath is a 5 cu.ft. ported cabinet with a 15 inch pro woofer...

Fusion-15: The Sentinel?

-> http://www.diysoundgroup.com/waveguide-speaker-kits/fusion-15.html

[brj]

One caution... remember than an active speaker system doesn't necessarily require the crossover to be implemented with any DSP at all.  You can have a completely analog line-level crossover in an active speaker system.

(And the cost of it will be far less than the cost of an analog speaker-level crossover, as the component values are far smaller.)

[roscoeiii]

You could just use a good 15 or 20 band equalizer for cheaper and easier to use if this is the case. Your simply boosting frequency response, and contouring with gain actively "essentially", but still just running thru a passive crossover in your speakers. Really all your doing is correcting room response anyway, a better room is always the real answer, but impossible for most to pull off.

Everybody will argue added distortion, loss of purity, blah, blah, but truth is whatever makes you happy. I don't like putting band aides on 10,000 dollar preamps and amps though so luckily since my days of cheap receivers, and equipment no EQ is needed.

Gotta disagree there. The DEQX addresses the speaker phase and freq response separately from the room (of course below a certain freq it is largely the room). And then there is the customization that a 10 to 15 band EQ can't match. Will I eventually try active with the DEQX?  Yep, it is on the to-do list.

I moved from a DSPeaker Dual Core to the DEQX and it was a nice step up for the sound of my system.

[rajacat]

I built an active/passive? 2 way + subwoofers setup that just uses an L Pad to attenuate the compression driver to match it better with the 15" midwoofer. Otherwise the CD is running free range.
The 18" SEOS waveguide with a BA 750 CD reaches down to 650hz or so where it matches up with the AE TD15M mid woofer. The BA750 1" CD is being powered by rebuilt/modded Heathkit W5m tubed monoblocks. I'm using a Crown XLS 1500 with its adjustable filter set @ ~670hz LP to power the TD15M's. They're running free range on the  on the lower side at this time.
I'm also employing a couple of DIY Anarchy folded horn 6-1/2"subs that are running full range. They roll off naturally @ ~100hz and go down to 20hz. I might put an inline filter to roll these off @65hz because the TD15M's go that low despite the sealed 3.5 cu ft cabinet. I also have a traditional 12" sealed sub and another 12" push-pull sub that I've yet to set up.
Also, as you see in the photo, the waveguide/cd is built in a separate cabinet that can be moved to adjust for phase alignment.

The speakers kick azz, image very well, can handle large classical works and are good at low volume as well at earbleed levels. I do have a minidsp that I might use although I'm not sure that it would improve the sound and would make everything more complex. Both of the drivers are fairly flat FR anyway and given all the other variables room, source, cables, etc., I might just leave everything as is, aside from minor tweaks, and concentrate on improving my source.

[richidoo]

IMO, it's not a big deal, our ears aren't sensitive enough for it to matter outside of the midrange.

Hearing research is done with untrained listeners who never focus on sound quality. So I take blanket statements about the generic human performance envelope with a grain of salt. Most audiophiles are skilled listeners and can hear phase error and group delay at any audible freq. And in my experience they tend to prefer lower phase error when they hear it.
https://trueaudio.com/post_010.htm
https://trueaudio.com/post_011.htm

Also IMO, speakers tend to start sounding worse when DSP is used above the bass frequencies.

Implementation matters. 'DSP' is just a non-physical concept, and there are many opportunities in the implementation of DSP to ruin the signal quality. They can be solved, but with refinement comes higher price. Better filter algorithms, more processing power, better quality i/o.

[undertow]

This is the only reason I jumped in if you read all the following comments from there after "Proves you need to up the complexity of the system going active"...

This is not that simple, nor exactly a guaranteed solution for most people in this game. There is more than one way to skin a cat no question, sometimes the simplest, and cheapest though nets the best results I guess was the point.

[DaveC113]

Hearing research is done with untrained listeners who never focus on sound quality. So I take blanket statements about the generic human performance envelope with a grain of salt. Most audiophiles are skilled listeners and can hear phase error and group delay at any audible freq. And in my experience they tend to prefer lower phase error when they hear it.
https://trueaudio.com/post_010.htm
https://trueaudio.com/post_011.htm


Implementation matters. 'DSP' is just a non-physical concept, and there are many opportunities in the implementation of DSP to ruin the signal quality. They can be solved, but with refinement comes higher price. Better filter algorithms, more processing power, better quality i/o.

I'm not citing hearing research, I'm citing my own experience and those of many people I've had over to listen, and that says you're very wrong about our hearing sensitivity at "any audible frequency", and you just said you've never actually experienced it yourself. I have...

Below a certain point and above a certain point getting phase perfect doesn't matter nearly as much and unless there is an issue well outside the range of typical you won't hear it. You won't hear group delay of ANY properly designed bass reflex cabinet. The midrange to lower treble is critical though, and while everyone will have their own sensitivities we're all not the special little snowflakes we think we are and no matter how well trained we think our hearing is.

And "implementation matters"? ...yes of course it does, most designs work well, be it tubes or solid-state, bass reflex or sealed-box, if the implementation is done correctly. On DSP though, in the critical midrange to upper treble, many people agree it should be avoided and that passive xos are better. Of course it depends on the speaker, but if we're talking about a well-designed speaker to begin with then I'll stand by my statements, if the speaker needs help and DSP can help it, then it'll be a far greater advantage but I'd argue you're just compensating for a flawed design, or you're using DSP to get away with a flawed design.

IMO, the key to speaker design is understanding psychoacoustics. If you don't this makes it impossible to choose the best compromises. Why prioritize things that can't be heard or we're hardly sensitive to over things that matter far more? Getting phase perfect at the frequency extremes is one of those things that just don't matter that much. Try it for yourself if you don't believe me.

[DaveC113]

There is more than one way to skin a cat no question, sometimes the simplest, and cheapest though nets the best results I guess was the point.

This I 100% agree with, design choices that result in a simpler speaker or component often sound better and cost less, but too simple and you start compromising performance in other ways.

But some folks love single driver speakers, I do... but I'll take a woofer and tweeter as well if it's implemented seamlessly. Some people like simple triode preamps with no feedback, they have their charms, but a slightly more complicated design with an active triode load can be far better for my tastes.

As far as DSP, for mid-treble I think it's far too complicated and can't beat a decent design that only requires a fairly simple passive xo.

[undertow]

True... So it all comes at a cost, but I do 100% agree with your overall philosophy.

Funny thing sometimes the simpler = better quality = more money!

Just a few examples like Lowther drivers, yes full range, yes crossover-less in many designs, but the drivers cost more than full fledged 4 way speakers in many cases! Add in the complex cabinets such as transmission lines, and you may end up in a wash. Whether this is better or not is of course in the "Ear" of the beholder.

Triode amps same deal, you can spend 2000 bucks just on good tubes Sure they are simpler, but yeesh the cost does not show that sometimes.

I actually heard one of the best amps I ever did, and it was a triode amp based on 211 tubes I believe... But they cost about 22,000 for the pair, only 50 watts, and the tubes are really really hard to get, for good ones anyway and you will pay up for them. Same goes for quality 300b etc... as well.

Actually the best Active designs would not use DSP, and would be more of the type using analog designed crossover circuits like something from "Marchand Electronics", but again the cost, and complexity is very high even higher than a 500 dollar DSP.

[richidoo]

This is the only reason I jumped in if you read all the following comments from there after "Proves you need to up the complexity of the system going active"...

This is not that simple, nor exactly a guaranteed solution for most people in this game. There is more than one way to skin a cat no question, sometimes the simplest, and cheapest though nets the best results I guess was the point.

I agree, the vast majority of audiophiles choose passive xo, because it's good enough for them. But there is more to be had with active/DSP, so its pursuit is valid, imo.

[srb]

I've had many great sounding combos with passive speakers and an amplifier, yet at least in the budget realm, quite a few people think for $300 the JBL LSR305 sounds much better than $200 passive speakers and a $100 amp, so maybe for those non-critical budget applications it makes more sense?

Steve