Sigberg Audio Manta Cardioid active speakers: Full measurements (Spinorama)

[MoreWatts]

Since every system needs subwoofers, kudos to Sigberg for actually addressing this with an actual "satellite plus subwoofer system."

Since the advent of subwoofers, bookshelves/standmount speakers should never bother to address frequencies <80 Hz, which small drivers can't handle very well at higher SPLs. Roll off quickly below 80-100 Hz and the 'satellite' all of a sudden can handle more power before distortion sets in.

I've never understood the obsession with the low frequency response of small speakers on this site. It doesn't matter if they have bass below ~80 Hz. Leave that to the subwoofer(s). Miller & Kreisel (M&K) promoted this >40 years ago.

Since the Sigbergs are not 'small' satellites, this system must sound absolutely explosive when cranked. Would love to hear some day.

 

[ryanosaur]

Since every system needs subwoofers, kudos to Sigberg for actually addressing this with an actual "satellite plus subwoofer system."

Since the advent of subwoofers, bookshelves/standmount speakers should never bother to address frequencies <80 Hz, which small drivers can't handle very well at higher SPLs. Roll off quickly below 80-100 Hz and the 'satellite' all of a sudden can handle more power before distortion sets in.

I've never understood the obsession with the low frequency response of small speakers on this site. It doesn't matter if they have bass below ~80 Hz. Leave that to the subwoofer(s). Miller & Kreisel (M&K) promoted this >40 years ago.

Since the Sigbergs are not 'small' satellites, this system must sound absolutely explosive when cranked. Would love to hear some day.

I would share my thoughts, but my pair of Mantas Thorbjorn was supposed to send me for longevity testing never arrived.

To be fair, it was just a suggestion in conversation one day, but...

Perhaps you can check your shipping department @sigbergaudio , please?

 

[sigbergaudio]

I would share my thoughts, but my pair of Mantas Thorbjorn was supposed to send me for longevity testing never arrived.

To be fair, it was just a suggestion in conversation one day, but...

Perhaps you can check your shipping department @sigbergaudio , please?

Hm.

 

[sigbergaudio]

Since every system needs subwoofers, kudos to Sigberg for actually addressing this with an actual "satellite plus subwoofer system."

Since the advent of subwoofers, bookshelves/standmount speakers should never bother to address frequencies <80 Hz, which small drivers can't handle very well at higher SPLs. Roll off quickly below 80-100 Hz and the 'satellite' all of a sudden can handle more power before distortion sets in.

I've never understood the obsession with the low frequency response of small speakers on this site. It doesn't matter if they have bass below ~80 Hz. Leave that to the subwoofer(s). Miller & Kreisel (M&K) promoted this >40 years ago.

Since the Sigbergs are not 'small' satellites, this system must sound absolutely explosive when cranked. Would love to hear some day.

Thank you, I agree with all of this of course.

And yes, cranking up this system is quite the experience. We've installed a pair of Mantas with dual subs in the control room of a recording studio this spring. They had existing in-wall main monitors with 18" drivers that they've now stopped using alltogether.

 

[poopy]

The Mantas and the two D10 subs create the best sound I have heard to date .
Keith

Muswell Hill, second pilgrimage destination after Lourdes. A place to see, touch and listen to the Mantas. Hallelujah!

 

[pierre]

Actually, they are only similarly priced if you forget to buy the necessary subwoofers.

Prices from Google today in Switzerland / Germany:

- Mantas + 2 big subs : 16k CHF (ref)
- KII 3: 15-16 k euros (ref)
- KII 3 + BXT : ~32k euros
- D&D 8C: 13k (ref)
- Genelec 8361A + 2 big subs: 14.5k CHF (Thomann)


So I would say Mantas or Genelec are cheaper (and/or have more output for the same cost). Another way to look at it is that KII and to some extent D&D are over priced

 

[holdingpants01]

Would you share full measurements at some point? There's no THD, separate drivers responses, decay, impulse response

 

[sigbergaudio]

Would you share full measurements at some point? There's no THD, separate drivers responses, decay, impulse response

Not putting a lot of effort into extended measurements now. THD is specified to be <0.5% THD @96dB/1m.

Here is a measurement of THD@100dB/1m (note that subwoofers will be overlapping at 100hz), red line is 1%:

Impulse response (measured at 2m distance in anechoic chamber):

 

[sigbergaudio]

Another interesting thing is on-axis response. The measured on-axis response for the Manta is not perfectly flat, I had half expected someone to point out the dip between 1-2khz by now. But let me do it for you. In isolation it looks a bit problematic, but is it really?

While there is a dip on-axis, there is excess energy off-axis in this same area. Here is on-axis vs 40 degrees:

So lifting this area to make it flat on-axis, doesn't necessarily equate to balanced in-room response, which is what we are ultimately after. A second point is that the Manta is designed for little to no toe-in (I recommend 0-15 degrees). That means on-axis isn't the listening axis, and the 10-30 degrees area is quite important.

Finally measuring is not enough, countless hours have been spent listening to ensure it actually sounds balanced in-room across different genres. It's also interesting to see how the speakers actually measure in-room.

Below are 500-20,000hz in two different rooms with no EQ. Notice how the 1-2khz area is now balanced. While it can be argued that a microphone records more of the reflective energy vs direct sound than our ears, balanced is also how it sounds.

 

[poopy]

Thanks for the clarification! In fairness, I did think about sending you a PM last night for the reasons behind this dip. So this choice is somewhat deliberate (as I was assuming) and not coming from any technical constraints. So why is it important for you to favour no or little toe-in? I mean you could have designed the speakers to have this balanced frequency response with large toe-in instead. Is it contributing to this huge 3d sounstage that the Mantas deliver?

 

[thewas]

Another interesting thing is on-axis response. The measured on-axis response for the Manta is not perfectly flat, I had half expected someone to point out the dip between 1-2khz by now.

It seems many ASR members are knowledgeable of loudspeaker engineering and tuning basics, namely that someone shouldn't fill for example on-axis dips which are results of diffraction effects and disappear off-axis but consider all angles and find a nice compromise between smooth direct (listening window) and reflected (early reflections and sound power) sound like you did.

 

[sigbergaudio]

Thanks for the clarification! In fairness, I did think about sending you a PM last night for the reasons behind this dip. So this choice is somewhat deliberate (as I was assuming) and not coming from any technical constraints. So why is it important for you to favour no or little toe-in? I mean you could have designed the speakers to have this balanced frequency response with large toe-in instead. Is it contributing to this huge 3d sounstage that the Mantas deliver?

There isn't technical constraints with regards to removing the dip by adjusting the filters, but there is excess energy off-axis between 1-2khz due to the baffle width that means it has to be done while considering the overall effect. This would have been even worse without the openings in the baffle on each side of the coax.

Looking at the graphs now I'm not entirely convinced we can't have our cake and eat it too, in the sense that it may be possible to balance this while still having "prettier" measurements too, possibly by adjusting the crossover point. Something to look into before Xmas perhaps. If you compare it to the SBS.1 measurements we can see that it isn't the coax itself that is the problem.

With regards to no toe-in, yes this combined with the coax gives the somewhat unique combination of a precise and yet very wide and large soundstage. The quest for the large and still natural sounding soundstage is also what makes me focus almost more on off-axis response than on-axis.

 

[Purité Audio]

Not particularly relevant to anyone else but the two in-room measurements I have made here certainly show no dip in that region, the whole target is a nice gentle downward slope.
A gentle toe in, last evening selected the p3 preset ( low -3dB high shelf centred around 2kHz )

https://cdn.shopify.com/s/files/1/0284/1427/9783/files/SIGBERG_AUDIO_MANTA_USER_MANUAL-2023-PRELIMINARY.pdf?v=1683128509

Keith

 

[Thomas_A]

Another interesting thing is on-axis response. The measured on-axis response for the Manta is not perfectly flat, I had half expected someone to point out the dip between 1-2khz by now. But let me do it for you. In isolation it looks a bit problematic, but is it really?

View attachment 322783


While there is a dip on-axis, there is excess energy off-axis in this same area. Here is on-axis vs 40 degrees:

View attachment 322779



So lifting this area to make it flat on-axis, doesn't necessarily equate to balanced in-room response, which is what we are ultimately after. A second point is that the Manta is designed for little to no toe-in (I recommend 0-15 degrees). That means on-axis isn't the listening axis, and the 10-30 degrees area is quite important.

Finally measuring is not enough, countless hours have been spent listening to ensure it actually sounds balanced in-room across different genres. It's also interesting to see how the speakers actually measure in-room.

Below are 500-20,000hz in two different rooms with no EQ. Notice how the 1-2khz area is now balanced. While it can be argued that a microphone records more of the reflective energy vs direct sound than our ears, balanced is also how it sounds.
View attachment 322781
View attachment 322782

I can see it is a clever design to deal with the stereo errors and rooms...and 0 degree axis is not always design axis.
Should sound very smooth in-room.

 

[hex168]

I wonder if a fully-padded front panel could remove the diffraction-related differences between on and off axis, and if it could be done in an aethetically appealing way. Some form of upholstered front panel?

 

[sigbergaudio]

I wonder if a fully-padded front panel could remove the diffraction-related differences between on and off axis, and if it could be done in an aethetically appealing way. Some form of upholstered front panel?

Something like that was indeed considered. But the ports in the baffle are quite effective too. Here are from early prototypes with and without the baffle ports (non-anechoic measurements). Notice how 30 and 60 degrees bunches up around 1500-1750hz and even crosses the on-axis in the first graph (plain baffle), as opposed to being more spread out and even in the second (baffle ports).

 

[sigbergaudio]

Tested a bit now, slight adjustments to crossover design and filters means we can probably get away with something like this. The overall balance should be very similar, but it looks a bit better. Soundwise I'm hard pressed to tell much difference as of yet, but will test this for a while and see (red is new). This is on-axis, non-anechoic nearfield.

 

[poopy]

Tested a bit now, slight adjustments to crossover design and filters means we can probably get away with something like this. The overall balance should be very similar, but it looks a bit better. Soundwise I'm hard pressed to tell much difference as of yet, but will test this for a while and see (red is new). This is on-axis, non-anechoic nearfield.

View attachment 323036

Can the Manta’ owners fiddle around with the filter settings by using the Hypex HFD software connected to the usb input on the amp plate?

 

[sigbergaudio]

Can the Manta’ owners fiddle around with the filter settings by using the Hypex HFD software connected to the usb input on the amp plate?

No, not directly. But they have access to a 9-band EQ with both PEQ and shelf filters. But that would be on top of the existing tuning.

Most people will very likely make the sound worse, so I would not feel confident having Sigberg Audio speakers in the wild that were in essence tuned by the customers themselves. It would be poor branding to allow customers to ruin the sound, and then show it off to their friends.

But if we update the tuning (as we have done with the SBS.1 once), you have the option of getting a remote software update. Also, to be clear the speaker is done. But when we do have digital filters, it's always tempting as a designer to keep improving.

This is a blessing as a curse, as it might as likely be an "improvement", so any iteration will be tested extensively. The change in SBS a year or so ago, was a shift to lower Q filters based on experience during the Manta development.

The change I suggested above was just a test to see if it could be smoothed out a bit on-axis, not sure if it will be actually implemented in the production model. I'm currently less than confident that will actually improve the sound, I'm just as concerned it will be worse. The 1-2khz area is very delicate in the sense that our ears are sensitive here - so it's better with slightly recessed sound than too much. So I try not to be too focused on how the graph looks, despite the fact people are very hung up on that these days.

 

[poopy]

Most people will very likely make the sound worse, so I would not feel confident having Sigberg Audio speakers in the wild that were in essence tuned by the customers themselves. It would be poor branding to allow customers to ruin the sound, and then show it off to their friends.

Good point, totally understandable.

But if we update the tuning (as we have done with the SBS.1 once), you have the option of getting a remote software update. Also, to be clear the speaker is done. But when we do have digital filters, it's always tempting as a designer to keep improving.

No doubt the Mantas is a finished product.

Improvement is with respect to something we identify as being a weakness. However, my understanding was that the design of the Mantas was intended to maximise stereo imaging/3D soundstage while trying to keep the frequency response as flat as possible with no toe-in (or little).

If the cost of getting a flatter on-axis frequency response leads to a deterioration of the stereo imaging, I don't see this as an improvement. A deterioration instead, given the motivation behind the design.