Article: Kii THREE Loudspeaker Review

[sapporo]

According to HiFi News review the internal sampling frequency is 93.75 kHz.

[firedog]

1 hour ago, sapporo said:

According to HiFi News review the internal sampling frequency is 93.75 kHz.

True. Wrote Kii and asked them.

[firedog]

20 hours ago, sapporo said:

According to HiFi News review the internal sampling frequency is 93.75 kHz.

Okay, wrote to Kii and got this answer:

 

The DSP runs on a high internal samplerate, which is purposely different from any of the typical samplerates used for LPCM audio (multiples of 44,1kHz or 48kHz).

The deciding factor is not the samplerate used (I wouldn´t even know the exact number to be honest with you), but the SRC algorithm at work.

The SRC was written by Bruno specifically for the Kii THREE and surpasses anything readily available on the market by a huge margin.

The jitter reduction at the output of the DSP is quite spectacular

Hope this was helpful.

[Alexander Kubiak]

Hi Mitch.

Late  to this topic also.

I have a pair of Kii three’s and have been enjoying them immensely for music connected to the Kii Control and my PS Audio Directstream.

Would like to know the correct connections,if it is possible, for Tv and  movies.

Input is the Nvidia Shield through my Anthem AVM 60 .

Since HDMI is not on the control and optical output from the Anthem is nothing but static, what is a method.

[sapporo]

Make sure the output is LPCM stereo and not bitstream surround.

[baconbrain]

21 hours ago, firedog said:

Okay, wrote to Kii and got this answer:

 

The DSP runs on a high internal samplerate, which is purposely different from any of the typical samplerates used for LPCM audio (multiples of 44,1kHz or 48kHz).

The deciding factor is not the samplerate used (I wouldn´t even know the exact number to be honest with you), but the SRC algorithm at work.

The SRC was written by Bruno specifically for the Kii THREE and surpasses anything readily available on the market by a huge margin.

The jitter reduction at the output of the DSP is quite spectacular

Hope this was helpful.

Maybe it is just me,  but this is now getting a bit difficult to understand... Why utilize an SRC algorithm which uses a LPCM audio "non standard" sample rate as a baseline?

 

Does this truly mean that the Kii will not benefit from being fed a specific (or preferred)  sample rate?  

[firedog]

7 minutes ago, baconbrain said:

Maybe it is just me,  but this is now getting a bit difficult to understand... Why utilize an SRC algorithm which uses a LPCM audio "non standard" sample rate as a baseline?

 

Does this truly mean that the Kii will not benefit from being fed a specific (or preferred)  sample rate?  

Obviously because Kii think their proprietary SRC is the best sounding way to do it. You can't ever accuse Bruno P. of relying on the truisms of audiophilia.  
 

My personal experience is that Kii are correct - it sounds best when fed unaltered input. You can hear, for instance, a difference if you feed it DSD or PCM upsampled to DSD - that sounds like DSD. But I personally think it sounds best when just fed an unaltered stream. At least I haven't heard any benefit when experimenting with upsampling the feed to the Kii (both with Roon and HQP).
If you want the sound of DSD instead of PCM, you could upsample everything to DSD and feed it only DSD. 

[baconbrain]

35 minutes ago, firedog said:

Obviously because Kii think their proprietary SRC is the best sounding way to do it. You can't ever accuse Bruno P. of relying on the truisms of audiophilia.  
 

My personal experience is that Kii are correct - it sounds best when fed unaltered input. You can hear, for instance, a difference if you feed it DSD or PCM upsampled to DSD - that sounds like DSD. But I personally think it sounds best when just fed an unaltered stream. At least I haven't heard any benefit when experimenting with upsampling the feed to the Kii (both with Roon and HQP).
If you want the sound of DSD instead of PCM, you could upsample everything to DSD and feed it only DSD. 

Thanks, understood.

 

Still I find it hard to believe that one can develop a SRC algorithm which is so good that it does not degrade the signal the whole time that the conversion is running to a constant degree Independent of the inbound sample rate . 

 

[sapporo]

DSD is 64x or 128x 44,1kHz. 

44100,  48000 and  93750 divided by 147,   160 and 312.5 (respectively) give 300.

To establish internal sampling at  93750  the master central clock must be scaled down

to exactly 12Mhz (eg Analog Devices https://wiki.analog.com/resources/tools-software/sigmastudio/usingsigmastudio/samplingrateconsiderations#setting_non-standard_software_sample_rates )

Perhaps some DSP engineer can put it together and tell us what/if is going on.

[sapporo]

 

https://www.grimmaudio.com/site/assets/files/1088/speakers.pdf

 

5.2 Clock

The clock circuit is the same as that used in the CC1 except that the sampling rate is set to 93.75kHz instead of one of the more traditional audio rates. This is specifi- cally done to improve the performance of the SRC chip. An uncommon clock frequency reduces the odds that mix products between the incoming clock and the internal clock fall inside the PLL loop bandwidth of the SRC.

 

[baconbrain]

6 hours ago, sapporo said:

 

https://www.grimmaudio.com/site/assets/files/1088/speakers.pdf

 

5.2 Clock

The clock circuit is the same as that used in the CC1 except that the sampling rate is set to 93.75kHz instead of one of the more traditional audio rates. This is specifi- cally done to improve the performance of the SRC chip. An uncommon clock frequency reduces the odds that mix products between the incoming clock and the internal clock fall inside the PLL loop bandwidth of the SRC.

 

Interesting article, thanks for sharing!

[Reggy]

How much are these in USD $ ?

[Alexander Kubiak]

Including the Control and the stands, which are made specially for the speakers though difficult to assemble, with tax $18,000.00

[firedog]

59 minutes ago, Alexander Kubiak said:

Including the Control and the stands, which are made specially for the speakers though difficult to assemble, with tax $18,000.00

I’ve seen them discounted before, especially if you buy the whole package. 

The stands wouldn’t be hard to install if they came with instructions. Mine, at least, didn’t. The stands aren’t complicated to assemble, but it isn’t obvious just how everything fits together.

[Alexander Kubiak]

No instructions on my either. Put together then redo until correct.

Since I paid full retail I should had the dealer do the installation,  but here in Palo  Alto 18 grand is ‘chicken feed’ to the Audio dealers. 

[soupcon]

the kii three's are remarkable out of the box, but i wasn't prepared for the improvement that 175 lbs of lead shot made after putting it in the stands. in any case, these are game changers. 

[firedog]

12 hours ago, soupcon said:

the kii three's are remarkable out of the box, but i wasn't prepared for the improvement that 175 lbs of lead shot made after putting it in the stands. in any case, these are game changers. 

I filled mine with sand, which is what I had available. Does’t weigh 175, but it is plenty heavy. Those things don’t move or vibrate at all. 

[Ralf11]

an iPhone contains an accelerometer, which can be used as a "low-rent" vibrometer - no warranty as to suitability or ability to replace your laser interferometry setup 

[man10]

Wow, what a clear and comprehensive review! I am looking at them for a true corner deployment. Which looks even tougher than yours. In an equilateral deployment against a back wall the nearest surface is 60 degrees off axis, in a corner only 30 degrees.  Have you ever heard them or better measured them in a corner deployment?

[baconbrain]

7 hours ago, man10 said:

Wow, what a clear and comprehensive review! I am looking at them for a true corner deployment. Which looks even tougher than yours. In an equilateral deployment against a back wall the nearest surface is 60 degrees off axis, in a corner only 30 degrees.  Have you ever heard them or better measured them in a corner deployment?

Mine are located in the corners and I just love them. Once you have the contour settings dialed in, I believe they sound just as good in the corners as if they were free standing on the floor. You can find many impressions here: 

 

[Chiptech]

I really appreciated the Kii Three review. I wonder if you have worked with Linn Audio Space Optimization with their active speakers or other brands? I have their 530 system. Really like the sound in my room.

 

And, you might be interested in this article: www.vox.com/energy-and-environment/2018/6/5/17373314/electricity-technology-efficiency-software-waste-3dfs

"This technology could fundamentally change our relationship to electricity"

 

There is likely a audiophile application in their future. After they get done saving us from climate change.

[mitchco]

Thanks @Chiptech It took a while to find a doc on what Space Optimisation does: http://docs.linn.co.uk/wiki/images/8/84/SpaceOptimisation.pdf 

 

I have no doubt it makes a difference, as room modes, factoring in distances, absorption coefficients in an open or closed space is well known in small room acoustics. Both the Kii THREE and the D&D 8c use a similar technique for room boundary issues.

 

It would be interesting to see before and after measurements to verify the effectiveness... Have you seen any?

 

Thanks for the link to the electricity article. I will look at it in more detail when I have some time. Happy listening!

[Emlin]

BXT specs now on the Kii website:

 

https://www.kiiaudio.com/bxt.php

[Gasman]

A general comment/query about DSP if I may?

 

I do use DSP to do DRC and implement the x-overs for my active speakers and I feed the DSP unit an analogue signal from my DAC

 

So my thought is this - given that any DSP unit, whatever role it performs, necessarily requires and AD-DA process internally, does that (not) mean that one is listening to the DAC in the DSP, rather than the, potentially quite costly, upstream analogue source?

 

Can any DSP be truly transparent in respect of any analogue input?

 

?

[firedog]

3 hours ago, Gasman said:

A general comment/query about DSP if I may?

 

I do use DSP to do DRC and implement the x-overs for my active speakers and I feed the DSP unit an analogue signal from my DAC

 

So my thought is this - given that any DSP unit, whatever role it performs, necessarily requires and AD-DA process internally, does that (not) mean that one is listening to the DAC in the DSP, rather than the, potentially quite costly, upstream analogue source?

 

Can any DSP be truly transparent in respect of any analogue input?

 

?

Yes, you get AD/DA conversion. But the conversion of the Kii is really world class. Your analog source will sound like your analog source. As far as I'm concerned, it's transparent to source.
Ask some of the people feeding an expensive turntable to the Kiis and who are quite satisfied.