Ascilab C8C Active Speaker Review

[thewas]

As all Ascilab loudspeakers fine performance, although on this model and price I am a bit let down by the rear radiation from which is quite changing in the mid region compared to other similar sized cardioids:

 

[ShadowFiend]

This is a great speaker but the calling it a cardioid speaker is a bit too liberal here. For my nitpicking eyes, I find it difficult to consider a region from 350-800Hz a cardioid pattern. It is more like a omnidirectional dispersion with some lobing more. Kii3 has the same problem but less severe while D&D 8C is much better.

 

[kharan]

View attachment 516363This is a great speaker but the calling it a cardioid speaker is a bit too liberal here. For my nitpicking eyes, I find it difficult to consider a region from 350-800Hz a cardioid pattern. It is more like a omnidirectional dispersion with some lobing more. Kii3 has the same problem but less severe while D&D 8C is much better.

Ascilab had a prototype (measures available on the Korean thread, also partially reposted on ASR) where the lobing was less pronounced, but the directivity also widened in the forward direction in the 350-800 Hz region. Presumably, Ascilab listened to both tunings of the side firing woofers and decided on this version based on how it sounded.

There is a lobe at 110° rearward and another at 180° rearward. Those are presumably less easy to hear and more easily treated via panels.

 

[ShadowFiend]

Then why can Kii 3do much better than C8C on this problem. And D&D 8C is actually far better while having DSP and configuration of driver position not far different from C8C?. A passiv cardioid configuration is also much better in that regards. If only D&D can use a Purifi PTT8.0 or even WO24P for it then the distortion below 300Hz of passiv cardioid will be much better.

One of the important selling point of cardioid speaker is that they can be placed near the front wall due to cancellation on the rear side. Here the big lob at 180° rearward is only -7dB compare to the front side in the fairly important and easily SBIR-proned midrange region. I fails to see the reason to call it a cardioid speaker. And if you talk about directivity widening then does the diretivity already widening in an area of 700-1100Hz?

 

[capslock]

Kudos to ASCILabs, they are out to eat D&D's lunch. And thank you to Amir for testing!

I am not quite sure how this works. The two midwoofers in front work as an d'Appolito or is one of the two rolled off? The side woofers have a delay up to the tweeter XO frequency to produce the cardioid response? But below the lower midwoofer XO of 150 Hz, they are just plain sealed woofers?

I'm not sure nearfield can tell the story without the phase information:

How can the side woofers help with maintaing the radiation pattern if they are rolled off at about 8 db/oct above 200 Hz? Those little midwoofers don't start to beam until the tweeter XO frequency. Also, why do the midwoofers droop by about 3 dB/oct above 200 Hz?

 

[AsciLab]

View attachment 516363This is a great speaker but the calling it a cardioid speaker is a bit too liberal here. For my nitpicking eyes, I find it difficult to consider a region from 350-800Hz a cardioid pattern. It is more like a omnidirectional dispersion with some lobing more. Kii3 has the same problem but less severe while D&D 8C is much better.

Ascilab had a prototype (measures available on the Korean thread, also partially reposted on ASR) where the lobing was less pronounced, but the directivity also widened in the forward direction in the 350-800 Hz region. Presumably, Ascilab listened to both tunings of the side firing woofers and decided on this version based on how it sounded.

There is a lobe at 110° rearward and another at 180° rearward. Those are presumably less easy to hear and more easily treated via panels.

Then why Kii 3 with DSP and driver position configuration is not far different from C8C can do much better than C8C on this problem. And D&D 8C is actually far better. And the less than perfect rear cancellation in the midrange region is a krypto to any active cardioid. A passiv cardioid configuration is so far much better in that regards. If only D&D can use a Purifi PTT8.0 or even WO24P for it then the distortion below 300Hz will be much better.

One of the important selling point of cardioid speaker is that they can be placed near the front wall due to cancellation on the rear side. Here the big lob at 180° rearward is only -7dB compare to the front side in the fairly important and easily SBIR-proned midrange region. I fails to see the reason to call it a cardioid speaker. And if you talk about directivity widening then does the diretivity already widening from 700-1100Hz?

The precision of a cardioid pattern, especially in the midrange, depends on the driver placement, the width of the baffle, and the degree of forward directivity.

As forward directivity is narrowed, lobbing can become more pronounced, and as the distance between drivers increases, lobbing also becomes stronger. A wider forward radiation pattern could have reduced the lobbing pattern, but we considered control of the forward radiation to be more important than achieving somewhat reduced side and rear radiation, and accepted that trade-off. This decision leads to the result of consistent beam width above 200Hz.

For the Kii Three, it has wider forward beam width than C8C, and though D&D has the finest directivity pattern among three of them cause it has closest placement between front driver and the side slots, but it compromised the distortion around mid-bass which is inevitable of passive cardioid design.

To analyze the actual behavior in more detail, the energy radiated from the sides to the rear is already 9 dB lower on average that means there will be only about -3dB dip when the reflection adding to the direct sound as an out of phase timing. Furthermore, this falls within roughly the 400–800 Hz range, which is relatively easy to absorb. By contrast, the 100–300 Hz range, which tends to have a much more significant negative impact in real rooms, clearly exhibits a cardioid pattern. For this reason, it would be difficult to say that it should not be called cardioid. To mention a specific example, Genelec claims that the W371A has a cardioid pattern up to 500 Hz, but in reality it crosses over around 150–250 Hz, which means the cardioid effect can only be seen below roughly 200 Hz at most. In that sense, such criticism is somewhat unfair.

If the drivers had been placed closer together, the lobbing would have been reduced, but it would have been difficult to achieve the level of low-frequency output we obtain with the current 8-inch subwoofers, and it also would have been harder to relieve the midrange from low-frequency duties. In fact, a prototype using only 6-inch woofers showed a cleaner lobbing pattern, but within a nearly similar size it left something to be desired in low-frequency output and required much more low-frequency contribution from the midrange.

In that respect, some lobbing in the midrange is in fact the result of a realistic compromise made in consideration of perceived performance, and the benefits gained from that compromise were substantial enough to justify tolerating the lobbing in that band.


Of course we also considering smaller model which have cleaner midrange directivity pattern. That might satisfy your requirement.

 

[Exprymer]

Being a FIR tuned speaker, it is curious that the 4 KHz boost still persisted.

 

[AsciLab]

Being a FIR tuned speaker, it is curious that the 4 KHz boost still persisted.

As I mentioned early it was came from the baffle diffraction.

3. A little bump at 4-5kHz is caused by baffle diffraction. To optimize that bump we should make the baffle 3D shape. It costs a lot more than now even its benefit must be 1dB or so. You can use EQ or 15deg toeing out.

 

[jbrown]

I would be VERY interested to see the impact of a 40Hz and 62Hz HPF on the distortion tests.
Is there any chance that removing that energy from the system will result in a significantly lower measured distortion at higher frequencies? If so, how much difference?

 

[Sokel]

I would be VERY interested to see the impact of a 40Hz and 62Hz HPF on the distortion tests.
Is there any chance that removing that energy from the system will result in a significantly lower measured distortion at higher frequencies? If so, how much difference?

No need to remove them as they go higher, sweep does just that, and if so, as it's H2 dominant it would affect only 80Hz and 124Hz and that's it.

IMD is a whole different story though, a multitone distortion test would be far more interesting when lots of drivers are involved.

 

[Chromatischism]

Default (measured) on-axis and power average slopes are not very compatible with directivity spectrum. Frequency range with close to constant directivity index should have negative on-axis slope to get suitable slope for PIR and SP. Suitable slope targets for conventional multi-ways are mentioned in 'Multiple Regression Model for Predicting Loudspeaker Preference...' by S. Olive. PIR slope of -1.0 i.e. -0.69 dB/oct. is good 'factory default' target also for concepts closing to constant directivity. Ideal CD has the same slope for on-axis and power averages such as PIR and SP. Slope of 0 dB/oct. would produce too thin sound as CD with low overall directivity such as DI=5...10 dB. Stronger overall directivity with much larger LF radiators are able to challenge this rule of thumb, but probably doesn't enable 0 dB/oct. Has not enabled in practice so far ime.

This EQ produces more suitable slopes with C8C. General tilt of ca. -0.37 dB/oct. and reducing LF power due to omni bass should help overall balance to far field. Near field is different case.
View attachment 516224
Result as spinorama.

View attachment 516216

Another problem is that directivity changes from omni (DI=0) to cardioid (DI=4.8) from 68 Hz to 200 Hz. That is kinda unavoidable with tiny cardioid speakers, but it happens too rapid and too high. Cardioid covering also mid and upper bass (40...160 Hz) is helpful in practice (without multi-sub or flush-mounted half space system).

The CD and flat response remind me of my S400 MKII, and those I did not find bright. And I'm very sensitive to bright speakers.

Though looking at Amir's measurements, the C8C looks brighter from 2 kHz on up. At least on the Buchardts, the off-axis steps down more, making for a smooth sound.

This would be a good case for dialing in a slope with Audyssey, Dirac, etc.

I really like set-it-and-forget-it speakers that don't require that level of fussing, but otherwise they look great.

It's still a toss up for me between the C8C and the ELX Towers, or to see how the new Revels perform.

 

[kharan]

The precision of a cardioid pattern, especially in the midrange, depends on the driver placement, the width of the baffle, and the degree of forward directivity.

As forward directivity is narrowed, lobbing can become more pronounced, and as the distance between drivers increases, lobbing also becomes stronger. A wider forward radiation pattern could have reduced the lobbing pattern, but we considered control of the forward radiation to be more important than achieving somewhat reduced side and rear radiation, and accepted that trade-off.

For the Kii Three, it has wider forward beam width than C8C, and though D&D has the finest directivity pattern among three of them cause it has closest placement between front driver and the side slots, but it compromised the distortion around mid-bass which is inevitable of passive cardioid design.

To analyze the actual behavior in more detail, the energy radiated from the sides to the rear is already 9 dB lower on average that means there will be only about -3dB dip when the reflection adding to the direct sound as an out of phase timing. Furthermore, this falls within roughly the 400–800 Hz range, which is relatively easy to absorb. By contrast, the 100–300 Hz range, which tends to have a much more significant negative impact in real rooms, clearly exhibits a cardioid pattern. For this reason, it would be difficult to say that it should not be called cardioid. To mention a specific example, Genelec claims that the W371A has a cardioid pattern up to 500 Hz, but in reality it crosses over around 150–250 Hz, which means the cardioid effect can only be seen below roughly 200 Hz at most. In that sense, such criticism is somewhat unfair.

If the drivers had been placed closer together, the lobbing would have been reduced, but it would have been difficult to achieve the level of low-frequency output we obtain with the current 8-inch subwoofers, and it also would have been harder to relieve the midrange from low-frequency duties. In fact, a prototype using only 6-inch woofers showed a cleaner lobbing pattern, but within a nearly similar size it left something to be desired in low-frequency output and required much more low-frequency contribution from the midrange.

In that respect, some lobbing in the midrange is in fact the result of a realistic compromise made in consideration of perceived performance, and the benefits gained from that compromise were substantial enough to justify tolerating the lobbing in that band.


Of course we also considering smaller model which have cleaner midrange directivity pattern. That might satisfy your requirement.

Thank you for answering in such detail, the various compromises in speaker design can be quite obscure for the non-specialist.

I think this should be useful for readers to figure out which trade off you make in which case.

And also, for fun we can compare to a non-cardioid design.

 

[Matias]

Amir's measurements have also been registered on spinorama.org
https://www.spinorama.org/speakers/AsciLab C8C/ASR/index_asr.html

View attachment 516311

The ranking is getting better and better lately, congrats for your entries!

Ranking table of speaker's measurements

www.spinorama.org

 

[capslock]

Then why Kii 3 with DSP and driver position configuration is not far different from C8C can do much better than C8C on this problem. And D&D 8C is actually far better. And the less than perfect rear cancellation in the midrange region is a krypto to any active cardioid. A passiv cardioid configuration is so far much better in that regards. If only D&D can use a Purifi PTT8.0 or even WO24P for it then the distortion below 300Hz will be much better.

One of the important selling point of cardioid speaker is that they can be placed near the front wall due to cancellation on the rear side. Here the big lob at 180° rearward is only -7dB compare to the front side in the fairly important and easily SBIR-proned midrange region. I fails to see the reason to call it a cardioid speaker. And if you talk about directivity widening then does the diretivity already widening from 700-1100Hz?

Could you please try to rephrase your post? I have trouble understanding what you want to say. I suspect it was lost in translation.

 

[kimmosto]

The CD and flat response remind me of my S400 MKII, and those I did not find bright. And I'm very sensitive to bright speakers.

S400mkII is quite totally different case. It has 3-4 dB more relative power at 200-1400 Hz than C8C so it's definitely less thin with the same on-axis balancing.

 

[dcolak]

I don't get this. Why try playing so low when everything falls apart above 90 dB? Why not simply use bigger driver, bigger box, more sensitive drivers with headroom for playing loud?

 

[kimmosto]

Maybe get a couple more subs first, then?

Separate subs are usually crossed at ca. 80 Hz. In that point polar pattern of these has already collapsed to average conventional box. Large integrated stereo cardioid bass units would help, but availability as commercial is question. No problem as DIY.

 

[IngMishael]

I don't get this. Why try playing so low when everything falls apart above 90 dB? Why not simply use bigger driver, bigger box, more sensitive drivers with headroom for playing loud?

falls apart?

 

[AsciLab]

I don't get this. Why try playing so low when everything falls apart above 90 dB? Why not simply use bigger driver, bigger box, more sensitive drivers with headroom for playing loud?

Ascilab C8C Active Speaker Review

My main take is that Ascilab has optimised the response for flat listening window and PIR at the expense of on-axis response. This may or may not be correct for users. Some guidance on this matter in the user guide may be beneficial. what type of guidance are you referring to?

www.audiosciencereview.com

You are right, But check this first. Do we really fall apart?

 

[dcolak]

Ascilab C8C Active Speaker Review

My main take is that Ascilab has optimised the response for flat listening window and PIR at the expense of on-axis response. This may or may not be correct for users. Some guidance on this matter in the user guide may be beneficial. what type of guidance are you referring to?

www.audiosciencereview.com

You are right, But check this first. Do we really fall apart?

I'd say we do. Having to cut it at 60Hz to have it play clean?