ASR Directiva Open Source Speaker Review

[restorer-john]

I have used seas er18rnx and w300 waveguide with a scanspeak radiator in my active dsp loudspeaker HYBRID.
Crossover at 2,3 kHz 24 dB/oct Linkwitz Riley . They sound and measures very similar to my Genelec 8340.
HYBRID is a closed box loudspeaker made to play with dual subwoofers crossed at 80 Hz.
Knowing er18rnx very well, it has low distortion in the bass, but not at the same low level as the purifi midbass.
View attachment 157250View attachment 157251

Now that is a good looking speaker.

 

[AudioJester]

I might have missed it, but can you buy this as a kit somewhere? Looking for another project

 

[ROOSKIE]

I have used seas er18rnx and w300 waveguide with a scanspeak radiator in my active dsp loudspeaker HYBRID.
Crossover at 2,3 kHz 24 dB/oct Linkwitz Riley . They sound and measures very similar to my Genelec 8340.
HYBRID is a closed box loudspeaker made to play with dual subwoofers crossed at 80 Hz.
Knowing er18rnx very well, it has low distortion in the bass, but not at the same low level as the purifi midbass.
View attachment 157250View attachment 157251

I see the JBL 530's are there as well.
How do these two sets compare, your DIY and the JBL's?

 

[AudioJester]

I have used seas er18rnx and w300 waveguide with a scanspeak radiator in my active dsp loudspeaker HYBRID.
Crossover at 2,3 kHz 24 dB/oct Linkwitz Riley . They sound and measures very similar to my Genelec 8340.
HYBRID is a closed box loudspeaker made to play with dual subwoofers crossed at 80 Hz.
Knowing er18rnx very well, it has low distortion in the bass, but not at the same low level as the purifi midbass.
View attachment 157250View attachment 157251

Can you share measurements?

 

[tktran303]

People seem to question the value of this thing. I mean that is a different question.

Is the Adam S2V worth 8-10 times (depending on the country you live in) the amount of similarly sized the T7V?

Is the Genelec 8050 worth twice as much as the 8040, which is twice as much as the 8030?

Is the Dynaudio M5P 10 times better than a speaker that costs 10 times less?

How come the JBL stage 140 measures seemingly well; surely it must sound just as good as this Directiva… or something else..

How come you didn’t use x woofer with y tweeter?

Well one could- but that wasn’t the design brief.

This is a no holds all out assault on the SOTA for this size…

 

[HooStat]

From one perspective, the estimated $1300-$1400 cost is a worst-case scenario because it is the cost of the parts at retail. I imagine with a group buy of some sort, one could get the parts for less.

 

[preload]

Yes , the violent 27 KHz resonance in the tweeter is a real issue - especially if you play high resolution material . With 44,1 or 48 KHz sampling material theres no problem.

Why would a resonance above the audible range be an issue?

 

[thewas]

For more context, let's compare it to one of the best values out there, the Neumann KH310:

Spins looks almost equal(honestly can't decide which I like more).
KH310 extends ~3Hz deeper, being -1.9dB at 37Hz.
Distortion @96 is close, but Directiva comes out on top. This is especially true if we assume the Directiva spikes are a product of the miniDSP vibrating on top and the 27kHz breakup.

I that comparison I would love to see though also multitone IMD measurements like from Sound & Recording which unfortunately get often ignored, although they can be more audible and there a 3-way design like the KH310 has usually inherent advantages.

 

[aac]

I that comparison I would love to see though also multitone IMD measurements like from Sound & Recording which unfortunately get often ignored, although they can be more audible and there a 3-way design like the KH310 has usually inherent advantages.

It'll probably have better IMD for some tones actually. Woofer in kh310 needs more excursion due to sealed design and still plays up to 800 hz.
But I still feel like woofer this price and quality is kinda a waste for small 2 way design. The result is certainly great but it is probably possible to add a 12 ot 15 inch woofer for a price less than this purify woofer costs (I've checked BMS offerings as "made in Germany" company that's not cheap and their 15 inch costs about the same) and end up with appropriate directivity for "far field".

 

[xarkkon]

Just saw this and hadn't realised there was such an initiative going about. Man, this is interesting! I want to build one too! Over to the other thread to do quite a lot of catch up reading. Awesome work everyone!

 

[Richard Berg]

Seems to me the obvious comparison isn't Genelec (onboard amp, strictly nearfield) but the unpowered JBL 70x series. They've basically built a 706i

705i - 5" woofer, 8L, 45Hz extension (-3dB), $550
708i - 8" woofer, 32L, 42Hz, $1050

Directiva - 6.5", 15L, 39Hz, $1350

Size is smack in the middle of those well known competitors. Price isn't, but the bass does outperform, so the value prop isn't too far off especially if you enjoy tinkering with such things.

DSP setup complexity is comparable, though JBL does give the option of single-wire operation (reducing the # of amp channels if you don't need 100% max output). It would be nice to consider a minimalist passive crossover along the same lines.

 

[aac]

Hole? Dead?

We must be looking at different measurements. This is probably one of the 3 best 2 way speakers we've ever seen. There is a very slight directivity increase, but it's fantastically tiny. 90%+ of speakers have a much worse dip.



Comparing side by side with the 708p, this speaker digs about 7Hz deeper. 708p is down 1.9dB at ~47Hz; Directiva is down 1.9 at ~40Hz. Comparing the distortion side by side at 96dB:

View attachment 157232

Directiva has much better 96dB distortion performance where it actually matters. 708p has lower bass distortion where we can't really hear it. I'm assuming those 2 spikes seen on the Directiva graph aren't real(in the sense that they exist in the real listening situation). 708p might look better at 106dB(guessing), but 96dB is already loud enough for most, especially for a speaker this small.

Looking at the spins:

View attachment 157234

Directivity is about equal, but the Directiva is clearly better in terms of frequency response.

The Directiva is ~$1350 to build, while the 708p is ~$3,800(on Sweetwater with big sale, normally $4,600). I own the 708p, and I think it's amazing for the price, but this speaker looks even better(at least at 96dB). Seeing the measurements of DIY projects here, I've been somewhat disappointed, but this is right up there with Genelec in terms of performance, while being cheaper(to build). The 8330 cost more than this and measures worse.

For more context, let's compare it to one of the best values out there, the Neumann KH310:

Spins looks almost equal(honestly can't decide which I like more).
KH310 extends ~3Hz deeper, being -1.9dB at 37Hz.
Distortion @96 is close, but Directiva comes out on top. This is especially true if we assume the Directiva spikes are a product of the miniDSP vibrating on top and the 27kHz breakup.
View attachment 157235

The KH310 is an amazing speaker for the price, and one of the best values out there(if not the best) above $4,000. Not saying it assuredly sounds better than the KH310(I'm a huge fan of 3 way), but the fact that it arguably measures better for less than 1/3 of the price is amazing. Even at $4-5k full package price, I still think it's an incredible value. At 96dB it measures as well or better than the 708p and KH310, which are both in that same range.

I get that it's expensive, but at $1400 to build, this thing is a steal, and a huge win for DIY. The DIY designs measured up to this point have been bad to good in terms of value, but never exceptional. This is the first DIY design I've seen that really makes me want to try it.

It's really an interesting comparison, but it looks like jbl woofer prioritises other things other than distortion? It has a pretty large ribbed aluminum heatsink which makes me wonder if it's actually important. Surely jbl would've cut the cost on it if it wasn't.

 

[JohnBooty]

My wishes for the next design:
-. with a passive crossover (i am a fan of active, but is not very easy to get on a cheap (under USD10K) multichannel setup.

All you need is a MiniDSP (~$100 for the analog version, or ~$225 for digital) and four channels of amplification. You don't even need a 4-channel amplifier; it can be a pair of stereo amps.

I have achieved this in the past for $200 total with a $100 MiniDSP and two SMSL SA-50 amps @ approx $50 each.

 

[abdo123]

Not that subs are more eco-friendly, but don't you get better sensitivity using a crossover?

what are you referring to exactly?

 

[TimVG]

Seems to me the obvious comparison isn't Genelec (onboard amp, strictly nearfield) but the unpowered JBL 70x series. They've basically built a 706i

705i - 5" woofer, 8L, 45Hz extension (-3dB), $550
708i - 8" woofer, 32L, 42Hz, $1050

Directiva - 6.5", 15L, 39Hz, $1350

Size is smack in the middle of those well known competitors. Price isn't, but the bass does outperform, so the value prop isn't too far off especially if you enjoy tinkering with such things.

DSP setup complexity is comparable, though JBL does give the option of single-wire operation (reducing the # of amp channels if you don't need 100% max output). It would be nice to consider a minimalist passive crossover along the same lines.

1350 is for a pair. So 675 per speaker.

 

[abdo123]

Use a renewable electricity provider. Managing carbon footprints with speaker sensitivity is not useful. We are not talking PA systems here.

The energy used to make any speaker, it’s materials, and ship it etc matters far more. Indeed, using one inefficient speaker for decades is likely better than any energy savings from sensitivity. Or indeed if the factory it was made in used coal based electricity. The amp efficiency matters more. The amount of time the amp is left on matters more. Or that it uses in standby mode.

With class D ones wireless router likely has a far larger energy footprint in use. And those are typically left on 24/7…

Though the lifespan and ability to repair class d is a huge factor when considering lifecycle energy footprints.

To save the planet in terms of audio, Get the best stuff you can and keep it forever. May even be cheaper amortized over the years. Repair it when it breaks.

The thing is it would take 10 times more power to drive a 83 dB/2.83V speaker (this design) to the same level of loudness as a 93 dB/2.83V speaker. So i think you're underestimating a bit how inefficient speakers are in terms of power consumption. it's kind of insane.

One thing to add in mind is that sure one might buy a 1000W Class D from Hypex or whatnot but power consumption at idle / low load is usually positively correlated to the maximum power an amplifier can provide.

Also from an engineering stand point it makes sense to build a sensitivity focused design with multiple (purifi) woofers or compression drivers. it's trivial to get a flat LW with the Klippel NFS, so it makes more sense to push for the other parameters in the speaker design dilemma (sensitivity and size) more aggressively on future designs

 

[thewas]

It'll probably have better IMD for some tones actually. Woofer in kh310 needs more excursion due to sealed design and still plays up to 800 hz.

650 Hz actually is the crossover frequency but the biggest advantage is the lower IMD above at the more critical and audible mids.

 

[voodooless]

Also from an engineering stand point it makes sense to build a sensitivity focused design with multiple (purifi) woofers or compression drivers. it's trivial to get a flat LW with the Klippel NFS, so it makes more sense to push for the other parameters in the speaker design dilemma (sensitivity and size) more aggressively on future designs

Not so sure that is the case. If one were to make a more than 2-way system, and it still needs to be relatively small, I would just buy the cheaper Purify (with lower Xmax, or even the smaller ones), and add two Dayton Ultimax 8" woofers in a closed box. Down low. They will outperform the Purifies in SPL and probably in distortion as well (especially < 50 Hz). Sensitivity is bad obviously, but in an active system and class D amps available, who cares?

Now if you were to go bigger, things really change. Then it would make sense to go for the sensitivity goal, for small narrow enclosures it rarely is useful if you want some real low-end extension.

 

[PeteL]

Lol. No tangerine wave guide, instead we get fusilli screen. $240 each driver. Though I chuckled at the review by some dude who cannibalized the amp/dsp/etc from an ikea/Sonos speaker to use with it.

Right now The KEF 350 are listed fir $750 a pair. So the drivers alone would be 2/3 the cost of a finished pair. maybe one can DIY make a better measuring speaker that the 350 for the remaining $270 for hardware, finish, cabinets and crossover (if passive) and we don’t count labor). Those KEF engineers seem to earn their pay though.

All I am saying is that the Directiva is amazing and impressive and a feat at $1k and as others have noted bests many much more costly designs. The excellence speaks for itself.

No need to value engineer it. Indeed, not only is it good value, it makes me appreciate more the value presented by some manufactured speakers. The directiva demonstrates that DIY can now be about high-end quality and value.

But the Directiva is far from 1k... 1400$ in parts and does not include the DSP and amplification. Not include tooling, not include R&D, engineering, fabrication, nice finish, marketing (OK as DIY you can scratch Marketing, but none of the other stuff... ). Amir's evaluation of the real cost at 4-5k for a full finished speaker is in my opinion under evaluated, we are maybe looking at a considerably more expensive than 5k pair of bookshelf powered speakers.

 

[abdo123]

Not so sure that is the case. If one were to make a more than 2-way system, and it still needs to be relatively small, I would just buy the cheaper Purify (with lower Xmax, or even the smaller ones), and add two Dayton Ultimax 8" woofers in a closed box. Down low. They will outperform the Purifies in SPL and probably in distortion as well (especially < 50 Hz). Sensitivity is bad obviously, but in an active system and class D amps available, who cares?

Now if you were to go bigger, things really change. Then it would make sense to go for the sensitivity goal, for small narrow enclosures it rarely is useful if you want some real low-end extension.

I'm not sure I quite follow, wouldn't larger (yet along multiple) woofers increase the Air volume requirement of the cabinet considerably making it already significantly larger?

it seems like size increases always in your scenario, sensitivity is still meh in the ~85dB range and nothing is really improved except maybe max SPL but you sure as hell would need a 1000W amp to get there.

Even just adding more woofers to this current design till it's tweeter limited would get the sensitivity up to the 90 dB range without major changes in design approach.