• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

GGNTKT Model M1

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
Is it to reduce hiss that you use also passive attenuation at the tweeter?

No, it's not post amping, it's before, so between DAC an amp. This way we don't reduce any resolution in the digital domain nor do we use any passive components within the amplified signal.
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
@roland{at}GGNTKT , could you take a look at my previous post (#70) by any chance?

Will these be 'built-to-order' or will there be the option of auditioning them in-home eventually? Seems like a very interesting concept and while I'm intrigued, in this price range I'd prefer to hear (and measure) the benefits compmared to a 'normal' speaker for myself, in my own room, before making a commitment. From the preliminary measurements it looks like it will be a very nice loudspeaker for sure, but I do feel the cardioid settings will be room and placement dependent, hence the slight reservation.

Yes, most are built-to-order unless we have a speaker in stock, that exactly matches your order (color, special extras, etc).

As for in-home trials it depends on your dealer. In Germany we will have trial kits ex factory soon, but not outside. If you are based in U.K. you might check with @Purité Audio as soon as he gets Model M1.
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
Alright, so today I want to give a sneak preview of our Klippel NFS session. I tried to format them the same way as @amirm does, so you can compare to his diagrams. But please keep in mind that these are preliminary, there will be a another session in September and those will be the final ones.

So let's get started with CES2034 ("Spinorama"):
CEA2034.png

The green line at the button is the cardioid-mode, so you can see the extended directivity down to 200 Hz.

Early reflections:
Early Reflections.png

Typical dip of most non-coaxial settings, because of the vertical driver alignment.

So let's compare cardioid and non-cardioid mode and see how it works out:
Horizontal in non-cardioid mode:
Beamwidth_horizontal_non-cardioid.png

Very smooth and constant horizontal directivity 700 - 15000 Hz (red curve).
140° wide coverage (-6 dB), so a very broad sweet-spot in the horizontal plane.

Horizontal in cardioid mode:
Beamwidth_horizontal_cardioid.png

Mostly the same, but you can see the extra directivity down to 200 Hz. So it's 200° @200-600 Hz and again 140° @700-15000 Hz.

Also interesting – grope delay:
GroupDelay.png

Low and constant >80 Hz, thanks to FIR filtering. Even below it's pretty flat because of the closed cabinet (curve <30 Hz invalid, because of bad SNR). That is what audiophile call a fast bass ;)
 

richard12511

Major Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
4,335
Likes
6,700
Alright, so today I want to give a sneak preview of our Klippel NFS session. I tried to format them the same way as @amirm does, so you can compare to his diagrams. But please keep in mind that these are preliminary, there will be a another session in September and those will be the final ones.

So let's get started with CES2034 ("Spinorama"):
View attachment 75832
The green line at the button is the cardioid-mode, so you can see the extended directivity down to 200 Hz.

Early reflections:
View attachment 75833
Typical dip of most non-coaxial settings, because of the vertical driver alignment.

So let's compare cardioid and non-cardioid mode and see how it works out:
Horizontal in non-cardioid mode:
View attachment 75835
Very smooth and constant horizontal directivity 700 - 15000 Hz (red curve).
140° wide coverage (-6 dB), so a very broad sweet-spot in the horizontal plane.

Horizontal in cardioid mode:
View attachment 75834
Mostly the same, but you can see the extra directivity down to 200 Hz. So it's 200° @200-600 Hz and again 140° @700-15000 Hz.

Also interesting – grope delay:
View attachment 75836
Low and constant >80 Hz, thanks to FIR filtering. Even below it's pretty flat because of the closed cabinet (curve <30 Hz invalid, because of bad SNR). That is what audiophile call a fast bass ;)

Wow these look excellent. Very close to Genelec Ones neutrality, lower cardioid dispersion, and much less crippled by output limitations, and at a great price. Spectacular! I may have to check these out.

Question: I'd go for the higher powered model, for which the website lists 121db(?) as max output. Where in the frequency range is it most limited?

Also super interested in the M2 for use as mains with M1 surrounds.
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
Nice, do you have distortion measurements?

Yes, here is a 95dB plot which is comparable with the 96dB @amirm usually measures (also the same 5% scale):

Harmonic distortion (relative).png


So THD mostly <0.5% @95dB, which is dominated with K2.
THD raises <100Hz because of the woofers mechanical stress (excursion = because of the closed cabinet we need some EQ here).

As for K2 it's a bit elevated because of the compression driver. The deal is you get extremely low K3, but slightly more K2. But the trick is also, that it doesn't get more at any level. So it stays <0.5% K2 even @110dB, whereas dome tweeter can not scale to these levels regarding distortions.

But let's zoom in and look at the 1kHz region which is the most sensitive for hearing:
Harmonic distortion (relative)_zoom.png


K3 stays <0.05% (blue line) even @95 dB, that's extremely low – it might even beat some amps.

But again as for the elevated K2: As I already said with our pascal amps, were are optimizing for hearing thresholds. Here is a nice plot that shows K2 threshold at different levels:
PhonGF_HideK2.png


https://www.hifi-selbstbau.de/index.php?option=com_content&view=article&id=239

As you can see the K2 threshold line as with 90 dB is 2% as for 1 kHz (yellow line) – so we have >400% headroom in the most sensitive area.
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
Question: I'd go for the higher powered model, for which the website lists 121db(?) as max output. Where in the frequency range is it most limited?

Also super interested in the M2 for use as mains with M1 surrounds.

Yes, Model M1 S delivers +2db more peak level, because of more amping power and the advanced limiters, which control peak performance. This way we can push each woofers to >240W for a short period of time, whereas the nominal limit is 60W (per woofer).

SPL limits are mechanical, so <100 Hz it's getting lower (see distortion plot above). We might have a plot that shows max. SPL @3 and 10% distortion limits.

Right, M2 for L-C-R and M1s for surround is certainly a nice setting, for both audiophile multichannel and home theatre.
 

phoenixdogfan

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
3,294
Likes
5,071
Location
Nashville
So what are we looking at for -3db in room response, Roland? And will it give that at 100db? And it it possible, since you have a Klippel, to do the in-room response curve?
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
So what are we looking at for -3db in room response, Roland? And will it give that at 100db? And it it possible, since you have a Klippel, to do the in-room response curve?

In room-response should be down to 40-45 Hz depending on levels (dynamisch EQ).
I will post the final measurement when they are done.
 

TimVG

Major Contributor
Joined
Sep 16, 2019
Messages
1,181
Likes
2,573
Not asked before: is there any sort of quality assurance in terms of unit to unit variation?
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
Not asked before: is there any sort of quality assurance in terms of unit to unit variation?

No for the speakers itself (mechanics), yes for the electronic (additional QA with protocols).

This is also because the drivers need a bit of break in, which takes place at customers home (10-100h depending on content and level). Although we have had good experience with SB Acoustics (LF drivers) in terms of consistent performance. There might be unit-to-unit variations left, that can easily be EQed – remember you have access to the DSP via USB or you might ask you dealer or acoustical consultant.
 

jae

Major Contributor
Joined
Dec 2, 2019
Messages
1,208
Likes
1,508
How do the purifi drivers compare to the sb ones?
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
So here are the results of our final Klippel session this week. We worked on some details and implemented a new feature (in software) which is called "dynamic limiter". It works similar like an EQ at the lowest end and extends bass at low levels (now 32/36 Hz -3/-6 dB), while protecting at high levels. It also works in the time domain, so that you can safely overload the drivers for a short period (electrical). These limiters are very advanced as they work in a “look ahead” way, without clipping or any other negative audible effects. It sound very nice ;-)

Spinorama in monopol (non-cardioid) mode – you can also see the new bass extension:
GGNTKT_M1_Spinorama_Monopol.png


Spinorama in cardioid mode – again, nice to see the extended directivity to 200 Hz because of the radiation pattern:
GGNTKT_M1_Spinorama_Cardioid.png


There will be a full stack documentation in a couple of weeks.
 
OP
q3cpma

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
So here are the results of our final Klippel session this week. We worked on some details and implemented a new feature (in software) which is called "dynamic limiter". It works similar like an EQ at the lowest end and extends bass at low levels (now 32/36 Hz -3/-6 dB), while protecting at high levels. It also works in the time domain, so that you can safely overload the drivers for a short period (electrical). These limiters are very advanced as they work in a “look ahead” way, without clipping or any other negative audible effects. It sound very nice ;-)

Spinorama in monopol (non-cardioid) mode – you can also see the new bass extension:View attachment 82507

Spinorama in cardioid mode – again, nice to see the extended directivity to 200 Hz because of the radiation pattern:
View attachment 82508

There will be a full stack documentation in a couple of weeks.
Very nice. Some questions: isn't 8 ms of delay a bit much for TV/games? Do you expect the cardioid effect of the M2 to reach even lower, or will it just provide more LF extension and/or SPL ?
 

maverickronin

Major Contributor
Forum Donor
Joined
Jul 19, 2018
Messages
2,527
Likes
3,308
Location
Midwest, USA
Very nice. Some questions: isn't 8 ms of delay a bit much for TV/games?

I'd think that should be good. It's on par with fast gaming monitors. Average consumer TVs and monitors usually have a ton more.

I think I missed where you got the 8ms from though.
 

roland{at}GGNTKT

Active Member
Audio Company
Forum Donor
Joined
May 4, 2020
Messages
117
Likes
666
Location
Germany
Very nice. Some questions: isn't 8 ms of delay a bit much for TV/games? Do you expect the cardioid effect of the M2 to reach even lower, or will it just provide more LF extension and/or SPL ?

Well, 8 ms is very fast and should work for TV use cases. I can't speak for games, since there might be circumstances where every ms counts (player even change their network card).

Yes, Model M2 will get quite a bit lower, because (a) of the wider baffle (more distance between the drivers) and (b) because of the separate chamber inside (Model M1 has all 3x drivers in one cabinet).
 
Top Bottom