Ascilab A6B Speaker Review

[eboleyn]

Having built some DIY speakers with Purifi woofers in them, I'm very excited to see well-done commercially produced ones!

 

[PristineSound]

The real audible event happened at 101 dBSPL, not 96. At 96, it was just starting to be audible. In that comparison graph, I am commenting in general as I do with other review measurements.

Below the tuning frequency the woofer excursion moves twice than above that frequency. So if you play 20~30Hz bass note like sub-bass test songs loudly you can easily hear the bottoming distortion. It is widely occurred in ported or PR passive speakers. Sometimes this issue can broke the woofers. To avoid this problem, it’d be better to use 2nd order HP filter above 30Hz. It prevents unnecessary excursion(distortion).

As Amir said, in normal music there isn’t so much output below 40Hz(seeing typical music curve). Therefore with most of musics you won’t get that issue unless play them extremely loud.

Thanks for the clarification, looks like an excellent speaker. Just keep it <= 96SPL, which is plenty loud for the average person, especially if you want to preserve your hearing.

Can we expect the full towers Ascilab have even lower distortion?

 

[Maiky76]

Here is my take on the EQ.
Please report your findings, positive or negative!

For the score rational your journey starts here
Explanation for the sub score
The following EQs are “anechoic” EQs to get the speaker right before room integration.
If you able to implement these EQs you must add EQ at LF for room integration, that is usually not optional… see hints there.

The raw data with corrected ER and PIR:

Score no EQ: 6.6
With Sub: 8.3

Spinorama with no EQ

• Great design
• Great directivity
• Nice LF
• Not sure if the WG or the box is responsible for the ON vs LW discrepancies
• Could be smoother compared with other SOTA (Kef, Genelec...)
• Passive, is that an advantage or an inconvenience?

Directivity:
Better stay at tweeter height
Horizontally, better toe-in the speakers by 10/15deg and have the axis crossing in front of the listening location, might help dosing the upper range. explanation here

EQ design:
I have generated two EQs. The APO config files are attached.

• The first one, labelled, LW is targeted at making the LW flat
• The second, labelled Score, starts with the first one and adds the score as an optimization variable.
• The EQs are designed in the context of regular stereo use i.e. domestic environment, no warranty is provided for a near field use in a studio environment although the LW might be better suited for this purpose.
• One can model the EQ with Vituixcad by using the DSP "Generic" setting with 96000Hz sampling rate.

Score EQ LW: 6.8
with sub: 8.5

Score EQ Score: 7.1
with sub: 8.8

Ascilab A6B APO LW EQ 96000Hz
March042026-114658

Preamp: -2.20 dB

Filter 1: ON HPQ Fc 33.0 Hz Gain 0.00 dB Q 1.17
Filter 2: ON PK Fc 118.7 Hz Gain -0.70 dB Q 1.45
Filter 3: ON PK Fc 527.5 Hz Gain 1.51 dB Q 3.74
Filter 4: ON PK Fc 670.0 Hz Gain -1.16 dB Q 5.29
Filter 5: ON PK Fc 1210.0 Hz Gain -1.07 dB Q 5.94
Filter 6: ON PK Fc 5109.7 Hz Gain -1.48 dB Q 3.31
Filter 7: ON PK Fc 8685.2 Hz Gain -1.32 dB Q 3.57
Filter 8: ON PK Fc 13143.6 Hz Gain 2.09 dB Q 0.19

Ascilab A6B APO Score EQ 96000Hz
March042026-114658

Preamp: -2.00 dB

Filter 1: ON HPQ Fc 32.8 Hz Gain 0.00 dB Q 1.15
Filter 2: ON PK Fc 123.2 Hz Gain -0.85 dB Q 1.32
Filter 3: ON PK Fc 518.3 Hz Gain 1.35 dB Q 5.83
Filter 4: ON PK Fc 684.7 Hz Gain -0.85 dB Q 5.14
Filter 5: ON PK Fc 1225.4 Hz Gain -0.92 dB Q 5.99
Filter 6: ON PK Fc 5137.1 Hz Gain -1.37 dB Q 4.71
Filter 7: ON PK Fc 9309.3 Hz Gain -1.27 dB Q 4.77
Filter 8: ON PK Fc 17931.4 Hz Gain 0.75 dB Q 0.12

Spinorama EQ LW

Spinorama EQ Score

Zoom PIR-LW-ON

Regression - Tonal

Radar no EQ vs EQ score
No improvements?
Probably does not need an EQ

Bonus EPDR:
From scanned data so no 100% accurate but good enough...

The rest of the plots is attached.

 

[phoenixdogfan]

Below the tuning frequency the woofer excursion moves twice than above that frequency. So if you play 20~30Hz bass note like sub-bass test songs loudly you can easily hear the bottoming distortion. It is widely occurred in ported or PR passive speakers. Sometimes this issue can broke the woofers. To avoid this problem, it’d be better to use 2nd order HP filter above 30Hz. It prevents unnecessary excursion(distortion).

As Amir said, in normal music there isn’t so much output below 40Hz(seeing typical music curve). Therefore with most of musics you won’t get that issue unless play them extremely loud.

If such a filter is employed on the A6b's is that going to prevent the distortion Amir is measuring at his respective inflection points relative to volume? Or will the distortion still be there and just be mitigated by not having the extreme excursions and concomitantly severe distortions caused by having that woofer try to reproduce below 30hz frequencies? And could such a filter be added with a simple toggle switch on the back of the speaker in updated models?

 

[AsciLab]

If such a filter is employed on the A6b's is that going to prevent the distortion Amir is measuring at his respective inflection points relative to volume? Or will the distortion still be there and just be mitigated by not having the extreme excursions and concomitantly severe distortions caused by having that woofer try to reproduce below 30hz frequencies? And could such a filter be added with a simple toggle switch on the back of the speaker in updated models?

Below the tuning frequency, both woofer and PR cancel each other’s output while the excursion is rather higher. If you add the filter you won’t get the distortion by unnecessary movement of the woofer that actually doesn’t make SPL.

It’s really hard to add sub frequency highpass filter on passive speaker. Required parameter of inductor and capacitor must be ridiculously large. It is much easier to add it by DSP.

 

[jhaider]

Thanks for the clarification, looks like an excellent speaker. Just keep it <= 96SPL, which is plenty loud for the average person, especially if you want to preserve your hearing.

Given the woofer used, the clean output capability likely exceeds anything else of the same size. So if you’re happy with the output capability of pretty much any other 7” woofer speaker then you’ll be under the limits of this one.

I don’t know what if any other high-performance 7” woofer speakers (if any) @amirm has on hand (KH 150? M126Be? R3 Meta or Ref 1M?) but it will be interesting to see how the numbers on this distortion test start to stack up.

 

[kharan]

Given the woofer used, the clean output capability likely exceeds anything else of the same size. So if you’re happy with the output capability of pretty much any other 7” woofer speaker then you’ll be under the limits of this one.

I don’t know what if any other high-performance 7” woofer speakers (if any) @amirm has on hand (KH 150? M126Be? R3 Meta or Ref 1M?) but it will be interesting to see how the numbers on this distortion test start to stack up.

Bigger woofer for sure, but kh310 isn't much bigger in raw size.

Then there's orbit 11, which is a smaller speaker. But the directivity in the high mids low treble is unusual.

 

[amirm]

I don’t know what if any other high-performance 7” woofer speakers (if any) @amirm has on hand (KH 150? M126Be? R3 Meta or Ref 1M?) but it will be interesting to see how the numbers on this distortion test start to stack up.

Unfortunately all of those were loaners so nothing on hand. We will have to keep plotting to see a trend in the future. I have a few more speakers here to test.

 

[Everett T]

Does the tweeter borrow from the engineering of ceramic coated woofers of the Revel Performa BE line? I don't think I've seen a ceramic coated Aluminum tweeter before.

There are quite a few tweeters and woofers in the market place, nothing new.

 

[Rick63]

There are quite a few tweeters and woofers in the market place, nothing new.

Yep, Monitor Audio has been using their CCAM (Ceramic Coated Aluminum Magnesium) woofers and tweeters for quite a while.

 

[AndreaT]

Great review! Priced accordingly to performance. Is the crossover at 800 Hz (graph) rather than 1 KHz as in the spec sheet? Quite a lowish crossover point!

 

[olderman]

This is a review, listening tests and detailed measurements of the Ascilab A6B Stand-mount speaker with Purifi woofer and passive radiators. I purchased this from the company. The US cost is $1,955 each. Price varies in different countries due to tariff, local taxes, etc.
View attachment 515071
Please excuse the tapes. They are holding the magnetic rings in for shipping. I plan to sell these speakers after review so wanted to keep them in as new condition. On that front:

NOTE: My company, testedaudio.com, is the exclusive distributor for Ascilab products. While objective tests remain the same as for any other speaker, subjective remarks may have business bias.

The A6B manages to act serious and stately at the same time. The rounded corners and invisible fasteners give it an elegant look. And the solid weight, an impression of quality.

I am not a fan of the wrinkles the woofer surrounds of the Purifi woofers as far as looks. However, when the speaker is used dead on in normal lighting, they mostly disappear so were not an issue for me. The review picture however, highlights them to the max.

Speaker has dual Purifi passive radiators on opposing sides and are a nice indicator of when the speaker is being pushed by their excursion.

The bottom of the speaker is lined with some kind of neoprene which I much appreciate as it acts as some level of damping. To wit, I did not have to employ my elastic material for its measurements on top of Klippel Near-field scanner.

Ascilab A6B Speaker Measurements
As usual, we start with our all important, anechoic "CEA/CTA-2034" standardized frequency response and directivity measurements:
View attachment 515074
We have a very nice on-axis response (black). There is a bit of a drop in bass but then again, it extends way below to nearly 30 Hz with reasonable output!

The early window, and for that matter, sound power, track the on-axis extremely well indicating good directivity which we will examine more later.

We can see the bass extension comes from passive radiator tuning being on the low side:
View attachment 515075
The peak is around 40 Hz. As noted, resonances from the woofer is highly controlled. And given the fact that we don't have a port, there is not much to worry about with respect to enclosure resonances bleeding out. You pay a lot more for the radiator as opposed to just a port but this is the benefit it brings.

Early window sum looks very good, encouraging a room without much "treatment:"
View attachment 515076

With the resulting prediction of in-room response being what we look for:
View attachment 515077

Beamwidth is highly controlled, all the way to upper registers, courtesy of computer optimized waveguide:
View attachment 515078
The small penalty there is narrower beam width which translates into more focused imaging and less diffusion. Don't treat the side walls as to let that do its thing to widen the source image.

View attachment 515079

Vertical directivity is what it is for a 2-way speaker:
View attachment 515080

Stay at the tweeter axis.

Let me give you my usual distortion charts and then something new:
View attachment 515081
View attachment 515082

I have been on a mission to better correlate distortion measurements to audibility. Here, I could hear the start of very low frequency distortion at 96 dBSPL and heard the speaker nearly completely run out of excursion at 101 dBSPL. How can we better figure out from the graphs. Check this work in progress:
View attachment 515083
These are THD% levels charted below 200 Hz. That is where I heard the distortion. Above that, we are overdriving the speaker anyway relative to music.

We see a more or less plateau above 100 Hz indicating we have made the right choice regarding cut off at 200 Hz.

Below that we see different slopes in distortion. We see two distinct new distortion sources coming into play which I have marked as secondary and tertiary. I have placed a circle at the knee of those events. What this says is that you can play down to 42 Hz at 96 dBSPL whereas if you crank up to 101 dBSPL, you best stop at 65 Hz! This matches my hearing of the sweeps where I heard slight distortion at the very start of the 96 dBSPL sweeps. Whereas with 101 dBSPL, that lasted quite a lot longer and at much higher levels of audible distortion.

Going the other way, if you are playing at 81 dBSPL, you can essentially take advantage of the full response down to 30 Hz. We will see if this and the above prediction proves to be the case in listening test section.

Assuming above is correct, you don't need a subwoofer for 81 dBSPL and possibly 86 dBSPL. Above that, a subwoofer would be needed for those playback level and the frequency for that crossover is on the chart.

Finishing our measurements, we see some traces of resonances in waterfall graph:
View attachment 515084
Step response of the woofer is very smooth/clean:
View attachment 515085

Edit: impedance and phase:
View attachment 515092

Ascilab A6B Listening Tests
I paired the A6B with Topping LA90 amplifier which has modest amount of power. I did this to see how much power is needed given the fact that the sensitivity of A6B is a couple of dBs lower than average. I was pleasantly surprised that in near and mid-field listening, there was plenty of power available, albeit with one channel being driven.

First impression was quite a surprise: there was impressively clean and deep bass even in tracks that don't sport that spectrum! Research indicates 30% of our enjoyment comes from bass and there is no better example of this. The extra warmth, especially when it is this clean, balances tonality and creates butterflies in your stomach. I am not used to such a compact speaker being able to deliver such a bass.

I quickly skipped to my reference track for deep bass. This sized speakers either faintly reproduce the sub-bass, or distort it at slightly higher levels. The A6B in sharp contrast, reproduced the low energy like it was a full range speaker!!! It is one thing to see frequency response down to nearly 30 Hz, it is another to experience it with real music.

Excited, I cranked up the volume and bam! Massive distortion set it, just like the 101 dBSPL. You hear a static and see the dual radiators try to detach themselves as the fly out left and right! Crank it down a bit and you get a bit of roughness in bass. A bit more and you are back in business.

Note that the above test track has some of the lowest frequency energy track I have. On music with milder version of it, the A6B produced clean and authoritative bass that I could feel in my stomach (from 2 meters/6 to 7 feet away) and feel the air brushing against my face! I could get to mild distortion with cranking up the volume up again but not as bad as my reference track.

With general music tracks, no worries where there whatsoever with bass distortion and frankly any distortion. The high frequency notes would stand out in the midst of bass notes like nobody's business. Even tracks I have that don't sound very good on speakers (I use them for headphone testing), sounded nearly excellent here! The rest, wow, I kept smiling in disbelief! This is the most perfect speaker I have heard in this size. It just can't blast out its beautiful very low frequency beyond medium levels of loudness.

I am stunned....

Of course, at no time did I feel the need to EQ anything.

Nicely so, very little vibration was coupled to my desk where I had it -- unlike many speakers that uncover all kind of resonances in this setup (my far field listening space is down).

Imaging is rather focused as predicted. You hear the sound come from a circle close to the speaker.

As a side effect of my testing in near and mid-field, these speakers will also make an excellent studio monitor. I checked for horizontal directivity and it is excellent across 2 to 3 chair widths. Even after that, the drop off in high frequencies is rather small.

Conclusions
What a shame that I have started to sell these speakers and hence, my personal experience may not be trusted much. If you at all value my ethics, I am telling you, this is a speaker you have to experience. It will easily blow away any expectations you have.

Going into this review, I was kind of depressed. I an not into, "more expensive parts mean more performance" as some of you no doubt know. Those parts have pushed the price of this speaker rather high. What Ascilab has done with its tuning to deliver very low bass response that is ultra clean, is what justifies its cost. Yes, laws of physics can't be broken. It can't shake the house but if you are just a hair less power hungry than I am, you will be fully satisfied without having to mess with subwoofers and such.

As noted, I would give the A6Bs a serious consideration for mixing and mastering in smaller rooms. Pair it up with an ultra quiet amplifier like I did and you don't have none of the hiss issues of active monitors.

Please note that as of this writing, there is no production available of the A6B. Earliest time is May which means we will get them in June. I am going to put up my pair on Tested Audio but I have to say, I am so tempted to just keep them for myself! They are that good.

Needless to say, I am going to recommend the Ascilab A6B speakers. Objectively and subjectively, they have left me without pants.
-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Excellent review. I especially like the new distortion info. I hope you get your pants back.

 

[AsciLab]

Great review! Priced accordingly to performance. Is the crossover at 800 Hz (graph) rather than 1 KHz as in the spec sheet? Quite a lowish crossover point!

Nearfield measurement doesn’t reflect diffraction (baffle step) So the crossed point by NF measurement can’t be regarded as a real crossover point. You’d better see the null point in the vertical contour plot.

 

[amirm]

Great review! Priced accordingly to performance. Is the crossover at 800 Hz (graph) rather than 1 KHz as in the spec sheet? Quite a lowish crossover point!

Thanks. I just eyeball that alignment as the process is manual. I would go by the spec.

 

[phoenixdogfan]

Here is my take on the EQ.
Please report your findings, positive or negative!

For the score rational your journey starts here
Explanation for the sub score
The following EQs are “anechoic” EQs to get the speaker right before room integration.
If you able to implement these EQs you must add EQ at LF for room integration, that is usually not optional… see hints there.

The raw data with corrected ER and PIR:

Score no EQ: 6.6
With Sub: 8.3

Spinorama with no EQ

• Great design
• Great directivity
• Nice LF
• Not sure if the WG or the box is responsible for the ON vs LW discrepancies
• Could be smoother compared with other SOTA (Kef, Genelec...)
• Passive, is that an advantage or an inconvenience?

View attachment 515101
Directivity:
Better stay at tweeter height
Horizontally, better toe-in the speakers by 10/15deg and have the axis crossing in front of the listening location, might help dosing the upper range. explanation here
View attachment 515108



EQ design:
I have generated two EQs. The APO config files are attached.

• The first one, labelled, LW is targeted at making the LW flat
• The second, labelled Score, starts with the first one and adds the score as an optimization variable.
• The EQs are designed in the context of regular stereo use i.e. domestic environment, no warranty is provided for a near field use in a studio environment although the LW might be better suited for this purpose.
• One can model the EQ with Vituixcad by using the DSP "Generic" setting with 96000Hz sampling rate.

Score EQ LW: 6.8
with sub: 8.5

Score EQ Score: 7.1
with sub: 8.8

Ascilab A6B APO LW EQ 96000Hz
March042026-114658

Preamp: -2.20 dB

Filter 1: ON HPQ Fc 33.0 Hz Gain 0.00 dB Q 1.17
Filter 2: ON PK Fc 118.7 Hz Gain -0.70 dB Q 1.45
Filter 3: ON PK Fc 527.5 Hz Gain 1.51 dB Q 3.74
Filter 4: ON PK Fc 670.0 Hz Gain -1.16 dB Q 5.29
Filter 5: ON PK Fc 1210.0 Hz Gain -1.07 dB Q 5.94
Filter 6: ON PK Fc 5109.7 Hz Gain -1.48 dB Q 3.31
Filter 7: ON PK Fc 8685.2 Hz Gain -1.32 dB Q 3.57
Filter 8: ON PK Fc 13143.6 Hz Gain 2.09 dB Q 0.19

Ascilab A6B APO Score EQ 96000Hz
March042026-114658

Preamp: -2.00 dB

Filter 1: ON HPQ Fc 32.8 Hz Gain 0.00 dB Q 1.15
Filter 2: ON PK Fc 123.2 Hz Gain -0.85 dB Q 1.32
Filter 3: ON PK Fc 518.3 Hz Gain 1.35 dB Q 5.83
Filter 4: ON PK Fc 684.7 Hz Gain -0.85 dB Q 5.14
Filter 5: ON PK Fc 1225.4 Hz Gain -0.92 dB Q 5.99
Filter 6: ON PK Fc 5137.1 Hz Gain -1.37 dB Q 4.71
Filter 7: ON PK Fc 9309.3 Hz Gain -1.27 dB Q 4.77
Filter 8: ON PK Fc 17931.4 Hz Gain 0.75 dB Q 0.12

View attachment 515107
Spinorama EQ LW
View attachment 515102

Spinorama EQ Score
View attachment 515103

Zoom PIR-LW-ON
View attachment 515106

Regression - Tonal
View attachment 515105

Radar no EQ vs EQ score
No improvements?
Probably does not need an EQ
View attachment 515104
Bonus EPDR:
From scanned data so no 100% accurate but good enough...
View attachment 515121



The rest of the plots is attached.

Just to be clear here: When you write about the "Score EQ" it stipulates "The second, labelled Score, starts with the first one and adds the score as an optimization variable." You mean that you implement for first eight filters listed in the LW EQ and then add the four additional filters in the "Score" EQ as well, for a total of 12 filters? Or do you mean only the four filters listed in "Score" need to be implemented?

 

[Joe Smith]

They sound great! Definitely out of my price range, but it's great that there is a two-way solution that obviates the need for a subwoofer.

Those woofer surrounds are definitely different looking, but clearly, they do the job.

 

[Maiky76]

Just to be clear here: When you write about the "Score EQ" it stipulates "The second, labelled Score, starts with the first one and adds the score as an optimization variable." You mean that you implement for first eight filters listed in the LW EQ and then add the four additional filters in the "Score" EQ as well, for a total of 12 filters? Or do you mean only the four filters listed in "Score" need to be implemented?

The sentence describes the optimization process as follows:

Step 1: Design the LW EQ with a flat LW response as the target.
Step 2: Using the LW EQ as a starting point, re-optimize the EQ with the Predicted Preference Rating (“Score”) as an additional performance indicator in the global metric.
This results in two separate and independent EQs.

To achieve the intended response, the correct EQ must be used:

• For the LW response, use only the LW EQ.
• For the Score response, use only the Score EQ.

In the case of the A6B, each EQ consists of 8 biquads.

 

[ObjectiveSubjectivist]

Research indicates 30% of our enjoyment comes from bass

In my case it's 90%
But it must be the subbass that's pronounced, then I'm in heaven

 

[Sir Sanders Zingmore]

Note that the above test track has some of the lowest frequency energy track I have.

Thanks for the review!
Out of interest, what is the test track?

 

[dfuller]

I have to wonder if this would have been better served with ports than PRs. PRs are excursion limited, where ports are mostly limited by how much movement the air can do before it turns turbulent.

Great results at normal listening levels though.