Erin's review for the March Audio Sointuva

[DDF]

The significant frequency response drop outside the axial frequency response, for frequencies above 5kHz, leads to a drop in the sound power frequency response and must therefore be compensated by increasing the axial frequency response.

The direct response dominates timbre at high frequencies, much more than the "room". Toole's and Olive's studies reflect this by their recommendation to eq the room response in the bass, more so the on axis at higher frequencies. MLSSA's perceptual FFT window follows this logic as well.

I've designed many speakers since the 90s, and find that some on-off axis amplitude trading is necessary even around tweeter crossover (most DIY speakers miss this). IME and based on the research, trying this >5 kHz never really sounds neutral. I'd expect this to be bright on axis.

However, if the speaker is toed in so on axis is flat again, this sort of EQ can sound really good. I've done this on a few speakers and it helps get rid of that dead sound that highly directional tweeters may have.

Simultaneously bright and dead sounds like a contradiction, but its not.

 

[jhaider]

He's right though, the model does not favor narrow directivity speakers over wide directivity speakers or vice versa. it only punishes directivity mismatches.

Actually, I think the Olive model does “punish” wider dispersion speakers to a degree. The PIR of a wide dispersion speaker with flat on axis response will have less tilt than the model wants, even though IME in most rooms it won’t be subjectively brighter. Ironically enough some of the best engineered Harman speakers - 7-series monitors, with genuinely wide constant dispersion to 20kHz - have depressed scores as a result.

 

[Holmz]

For me it would be very interesting to send this speakers to the more subjectivist prominent reviewers. The outcome would be so interesting. What they hear, how they do descripe.
Please do this March Audio !

Well I have a trip to the US coming up, and the SoCal Audio convention was delayed.
And it is about 40 miles down the 405, and then down the 55, from where the kiddies ar at.
But that scheme would assume that there would be people that would want to have a listen and that Alan had a set available. Etc.

To clarify, I'm not dismissing subjective reviews. Merely, when you used the word "prominent", I understand you mean "famous" as the definition is such.
…

I do not usually get too impressed with many speakers, and when I find I like them, they usually have 5 digits in the price.

These would sound bright because there is no roll off in the frequency response at the listening position which normally results from an anechoically flat speaker and room interaction.
…

I don’t like overly bright speakers. (Look at my avatar FCS.)
These were not in the harshly bright group.
It was only about a 2 hour session, but it was nothing but good.

Lastly:
The only thing that has me chin-scratching is the step-function (impulse response). And I am sort of leaning toward an active XO/DS… and whether that would be better. Plus I am not a “make your own” passive crossover kind of guy.

 

[wwenze]

No surprise here I guess. Very well executed design. Though they are super hideous! The Purify looks like shit, the wood as well, and the tweeter color doesn’t match the rest at all.

The filter looks quite extensive. Quite some work went into this. Directivity wise, it is very smooth, but also a lot of narrowing at the top end. I think I’d like to see a slower slope.

"Don't let engineers make powerpoint presentations"

 

[Rick Sykora]

As a group, think we tend to be hypercritical. Have been guilty of this at times too....

With the room being such a key aspect of what we end up hearing, as long as the speaker can be eq'd, starting to think close is good enough. When I first joined ASR was skeptical that measurements would fully correlate well with what we hear. Along the way, started to think we might get there, but Amir's last few reviews have brought me full circle.

 

[jhaider]

No surprise here I guess. Very well executed design. Though they are super hideous! The Purify looks like shit, the wood as well, and the tweeter color doesn’t match the rest at all.

IMO questions about the finish aren’t that relevant. It’s just an option and one benefit of a MTO approach is you can get what you personally want. I’m more concerned by apparent lack of provision for a grille, as the worth of a loudspeaker without one approaches zero to me. Though maybe that is on the option sheet as well.

 

[DDF]

As a group, think we tend to be hypercritical. Have been guilty of this at times too....

With the room being such a key aspect of what we end up hearing, as long as the speaker can be eq'd, starting to think close is good enough. When I first joined ASR was skeptical that measurements would fully correlate well with what we hear. Along the way, started to think we might get there, but Amir's last few reviews have brought me full circle.

Nothing hypercritical [edit] about good engineering based on sound psychoacoustic principles.
Taking a speaker otherwise as good as this and insisting on EQ is ridiculous.
Toe it in, or fix the xover if its meant to be listened to on axis.

 

[DDF]

• The narrow radiation - while helping to keep the treble from sounding bright in-room - is the only underwhelming part about this speaker on paper to me. I generally prefer a speaker with a wider radiation pattern. However, when listening, I didn’t find myself having any issues with this. In fact, in many tracks, I felt the soundstage was wider than the data would have me believe. I’m not sure what the driving factor here is. Could it be the extremely smooth horizontal off-axis response? I noticed that on-axis, the soundstage didn’t seem that wide but when turning the speakers off-axis to tame the highs, the soundstage actually broadened more than I would have expected. Could this be due low crosstalk from speaker to the opposite side when listening in stereo?

It all comes down to the use case. I'm listening to one of my DIY speakers now using D26NC. A low distortion and flat tweeter but very narrow dispersion up top.

In a big room, it sounds great. This is why so many DIY speakers without controlled directivity get raves at DIY shows, the rooms they use are usually big. In my small office with close side walls, lateral reflections are more prominent and the narrow directivity makes it sound a bit dead up top. Tilting up the on axis doesn't fix it.

I purposely picked this tweeter for it narrow dispersion to create a more solid and well defined image in a small room. Works great for that but I miss the wider dispersion of a typical scan speak that make cymbals and hi-hats sound more real.

 

[Rick Sykora]

Hi Rick,

Unfortunately this is wishful thinking that we are all having. The reality is that cost of these drivers is not in the material costs.

It’s in the cost of them doing business; and Purifi or Scan-Speak or SB set their prices as they see fit. When Scan-Speak released their beryllium dome tweeter over a decade (!) ago it was one of their top selling tweeters. There’s little incentive to reduce prices.

If one wants a cheaper beryllium dome the SB29BAC is already available, albeit sans waveguide.

At least with SB wave-guided tweeter and Purifi mid-woofer, despite the high price, the high the performance is there; and the data-sheets have more detail than others. There’s no surprises and I know what I’m getting, as a system designer.

Unlike some other manufacturers where I have to spend hour$ doing my own preliminary studies, and worry about consistency in both supply and product quality.

If one wants improved economies of scale; then you have to move manufacturing offshore.
A manufacturer in USA, Canada, Australia, Denmark, Norway etc cannot compete with those in China or Indonesia or India, or even Taiwan.

This is our global economy; and each countries comparative advantage. There’s just no way I can see it changing.

Yes I can have a HP or Microsoft or Apple computer Made in USA; but how many people will pay double or triple the price for it?

I mean, look what happened after the Tymphany takeover of Vifa and Peerless.
Fast forward 10 years and no they have all but disappeared from retail / distributor channels. Unless you order 500+ units.
Same with Usher (Taiwan)

are we sure we want improved economies of scale?

Finally, it’s hard to face the facts, but in the grand scheme of audio, 6.5” midwoofers are now a niche market.

A 4” driver is now a “subwoofer” and automotive audio and phone audio businesses are even bigger than laptop computers, and the TV audio business, which in sum just eclipses the hifi and home theatre market.

It’s becoming like tonearm cartridges. A top quality niche product will have a high price. Unless there’s market competition there is just no incentive to reduce the price by a great factor.

I can certainly see this is a fine debut by Alan. I hope it is a home run hit, and Alan gets swamped with more orders than he can handle. It’s a good problem to have; rather than sit on a MOQ of 500 units…

Time will tell. But can tell you that few companies price their overhead into individual products. Overhead tends to get spread across products lines, groups or maybe the entire company.

The more likely scenario for a techology like these drivers is that they have priced to be a premium or something less. If they have a patent or some other advantage, this may last longer, but even without economy of scale, once a premium product loses that advantage, you better have the next big product advance ready or better be prepared to lower the price of the (now less premium) product.

EDIT:
Should note am talking price here, but mainly cost in earlier post. There is some likelihood that have improved product manufacturing over time (and/or volume has improved significantly) and reduced costs. This may allow me to sustain margin without having to lower the price of my older technology.

 

[YSC]

one more thing I think is great for this design is that it is narrow and shallow, which suits well in a desktop scenario! in a money no object case I would definitely want to try it as a ultamite desktop computer speaker

 

[Newman]

Assessed as a full-range in-room loudspeaker, it fails to deliver at that price.

For that price of unpowered speaker, would like to see:-

• deeper bass
• more SPL capability in the deep bass
• flatter axial treble output
• slightly less beamy treble
• even smoother directivity through the crossover

IMO the technician-reviewer's conclusions are a bit too soft and cosy. His conclusions would read reasonably if the speaker was being assessed as "a satellite speaker with external EQ correction". But as a full-range in-room speaker, used 'naked', the above dot points are missing.

A speaker at this price justifies some sort of testing of the claims made for the speaker. Would therefore like to see distortion and compression tested at 250W when tested within the long-term IEC protocol.

IMO one could develop a satellite speaker for $750/unit that matches this speaker in the satellite range, plus delivers the last 3 dot points. In which case, this speaker is overpriced as a satellite and inadequate as a full-range, and can't see why it should be recommended, at that price. Recommended as what?

 

[tktran303]

^^^

Mannnn…

As I’ve said many times previously.

There’s data. And there’s data.
And people’s interpretation of data.

Newman, did you even read the whole review?
Or are you trying to be provocative?

Do you need hand holding to open up different tabs and compare it to the Kii Three, Dutch and Dutch 8c, and JBL M2.
That Erin’s draws sonic comparisons to.
The Sointuva is not inexpensive, but more affordable and smaller that all three aforementioned.

The compression testing was done to 102dB.
If you can do the math, that’s nearly a full 200W into 4 ohms to hit that SPL; at 1m.

The 250W power handling you infer to, as published by Purifi, is IEC 268-5 18.2 and it applies to transducer units.

IEC 268-5 18.2:
“Long Term Power Handling – 10 times alternating between signal for one minute and pause for 2 min. Total test time is 28 min”

Quick side quest:

https://www.scan-speak.dk/datasheet/tech/Scan-Speak_Technote02_Powertest.pdf

It simply doesn’t apply in a speaker system.
There are many things that determine power handling, including box size and resonant tuning; but that’s beyond the scope of this discussion.

And Now I totally get manufacturers when nerds like me ask
“Hey Where are all the specs and graphs?”

And they say “well, people don’t understand all the data anyway”

not everyone is a trained engineer and can decider all the data.

I see now why we need marketing departments.

PS. Erin is a an actual rocket scientist / aerospace engineer IIRC. Not a technician.

 

[Rick Sykora]

Nothing hypocritical about good engineering based on sound psychoacoustic principles.
Taking a speaker otherwise as good as this and insisting on EQ is ridiculous.
Toe it in, or fix the xover if its meant to be listened to on axis.

You appear to have misread my statement. I said ”…we tend to be HYPERcritical”, NOT hypocritical.

So am not being hypocritical when I request that you might want to reconsider my words more critically please.

Thanks,

Rick

 

[Holmz]

... but Amir's last few reviews have brought me full circle.

So you can have your 2*pi and eat it too?

Assessed as a full-range in-room loudspeaker, it fails to deliver at that price.

For that price of unpowered speaker, would like to see:-

• deeper bass
• more SPL capability in the deep bass
• flatter axial treble output
• slightly less beamy treble
• even smoother directivity through the crossover

IMO the technician-reviewer's conclusions are a bit too soft and cosy. His conclusions would read reasonably if the speaker was being assessed as "a satellite speaker with external EQ correction". But as a full-range in-room speaker, used 'naked', the above dot points are missing.

A speaker at this price justifies some sort of testing of the claims made for the speaker. Would therefore like to see distortion and compression tested at 250W when tested within the long-term IEC protocol.

IMO one could develop a satellite speaker for $750/unit that matches this speaker in the satellite range, plus delivers the last 3 dot points. In which case, this speaker is overpriced as a satellite and inadequate as a full-range, and can't see why it should be recommended, at that price. Recommended as what?

In a more cozy fashion… when I heard them it was in a room (from memory) that was probably ~4.5 to 5 M wide, and ~4.5 deep with maybe 8-9 foot ceilings.
Over here on post #175 https://www.audiosciencereview.com/...march-audio-sointuva.26085/page-9#post-972485 was my impression from almost 2 months ago now.

We did not listen to any rap, or electronic music, but I noted that for the Doug MacLeod stuff there was no need for any lower, deeper, or more bass.
A lot of it was at 85 dB(A) with peaks up around 102-105 dB(A), and it was being pushed with a 125w/ch class-D Purifi I think.

So I am not sure that I would need too much more SPL.

@Newman do you have a list of music that you can recommend?
I might be able to get another listening session, and it will likely cost me a bottle of wine and a cheese plate, so as gain entry.
If that can happen, then I can try taking some better subjective notes.
And maybe some suggestions for mean SPL level?

I am not sure who else had heard these otheir than myself and Erin?
Since I am at an advantage geographically, then I am happy to try and get some more impressions.
(I’ll go back and read Erin;s subjective part again.)

Usually it is easy to do subjective listening, as I have something negative to focus on. But turned up high enough to not be able to have a conversation, and no evil distortion sounds, is always a good sign to my ear(s).

 

[Mnyb]

What’s the xover frequency and what kind of filters do we have here ?

 

[napilopez]

He's right though, the model does not favor narrow directivity speakers over wide directivity speakers or vice versa. it only punishes directivity mismatches.

Actually, I think the Olive model does “punish” wider dispersion speakers to a degree. The PIR of a wide dispersion speaker with flat on axis response will have less tilt than the model wants, even though IME in most rooms it won’t be subjectively brighter. Ironically enough some of the best engineered Harman speakers - 7-series monitors, with genuinely wide constant dispersion to 20kHz - have depressed scores as a result.

Just to add to the above, @jhaider is absolutely correct that the model does punish wide horizontal directivity speakers, but more specifically, what it does is punish behavior that trend toward constant directivity (if the on-axis is flat). It just so happens that the wider directivity a speaker is, the more it looks like constant directivity, so the ER and PIR end up having less tilt, and therefore the score may not be as high.

However it's important to note that olive acknowledges this as a limitation in the paper. Two relevant quotes:

The degree of tilt varies among curves for Test One and the larger sample. Test One includes mostly 2-way designs whereas the larger sample includes several 3-way and 4-way designs that tend to have wider dispersion (hence smaller negative target slopes) at mid and high frequencies. This suggests that the ideal target slope may depend on the loudspeaker’s directivity.

Emphasis mine. And then:

The speakers with the flattest sound power had rising frequency responses on-axis and/or reduced low frequency extension. Both are necessary compromises to achieve flat sound power for speakers that have a rising directivity at higher frequencies. Such speakers represent the vast majority of all speakers sold. A speaker with constant, flat directivity could theoretically satisfy the flat sound power criterion and still achieve high preference ratings, so long as it had a smooth on-axis response well-maintained off-axis. However, such speakers are not widely available."

To be clear, on the above paragraph "high preference ratings" means ratings in an actual blind test, not the predicted preference score.

Wide directivity aside, we also see this issue show up with some of Harman's own designs -- JBL monitors in particular, which tend to have a tilt and then be constant directivity from beyond a certain frequency; they often don't usually score as high as one might think. But that doesn't necessarily mean they aren't as good as some of the competition (not saying they are either, just one case where the score could be misleading).

In general, the model's 'ideal' directivity tends to align more closely with your typical Neumann/revel/genelec not-too-deep waveguides, assuming a perfectly flat on-axis. If the speaker is very wide, the score benefits from having a slight downward tilt on-axis. But that doesn't necessarily mean it'll sound better. Might be dependent on the room, content, and individual preference at that point.

 

[witwald]

• While the on-axis response shows a rising treble, I didn’t hear this as offensive.

This seems to imply that you did hear the rising treble. That indicates a non-natural response on axis, which in many long-term listening situations will end up being fatiguing.

• The narrow radiation - while helping to keep the treble from sounding bright in-room - ... when listening, I didn’t find myself having any issues with this. In fact, in many tracks, I felt the soundstage was wider than the data would have me believe.

This seems to correlate well with you not finding the rising treble to be a problem. Do you have a general preference for a loudspeaker that has a rising treble on your choice of test material that you use during the listening tests?

• Imaging also seemed quite good on this speaker and didn’t suffer when turning the speaker off-axis.

Did the imaging change at all? "Didn't suffer" leaves a lot for the reader to assume what was meant.

• In terms of output levels, due to the average sensitivity of about 83dB @ 2.83v/1m, this speaker needs power. I am using a Parasound HINT-6 which has about 220wpc @ 4ohm and was close to maxing it out.

What was the size of the listening room? How far from the loudspeakers was the listening position? These two factors would have a significant bearing on the power levels that you used during your listening evaluation.

The lowish sensitivity aligns well with the enclosure being relatively compact but with a good amount of bass extension.

 

[restorer-john]

as a ultamite desktop computer speaker

That's a good name for a TOTL small desktop speaker - "UltaMite"

 

[witwald]

The 16kHz won't make much of a difference, but 3dB around 10kHz will make stuff like high hats and sibilance in voice sounding harsh.
btw: there is also a crossover dip

Those certainly aren't ideal, but the boost at 10kHz and a higher would best be tamed. That will produce a better, more natural response for the listener. I'm not sure how the crossover dip would come across to most listeners. I think that a dip of that magnitude is much more preferable than a peak.

 

[restorer-john]

IMO questions about the finish aren’t that relevant. It’s just an option and one benefit of a MTO approach is you can get what you personally want.

I agree with this.

The options and finishes (at this stage) appear unlimited as long as the buyer pays for it. Like Wilson will paint your speakers to match your Lamborghini if you want.

I reckon a Union-Jack on the top like the Mini cooper's roof with a few white stripes on a BRG base colour would do the job. Might cost a few dollars extra, but it'd be fun.

If he'd sent a gloss white speaker, people would complain, same with gloss black. The wood is amazing, different and has provoked discussion- that's probably a good thing.