Interesting design philosophy from SoulNote!

GrisPato

GrisPato

Super Member
Just found this last week. Designer worked for NEC on the A10 which had an interesting design. Then with Marantz.
In many points he goes exactly against main stream. He has his own argumentation for why he does what he does.
I find it really interesting and think he is on to something and has his own way of achieving his goal.
I am really interested in someone's experience.

DESIGN PHILOSOPHY

www.soulnote.co.jp www.soulnote.co.jp
 
Register to hide this ad
Haha, didn't expect anyone to respond. Thanks.
By the way, there is an American distributor, looks he sells online.
For me the interesting things are that a renown engineer takes this approach. No NFB, although there is no schematic, I assume he is using some local NFB but not global. His theory is that trying to improve static measured distortion figures makes the sound more boring, less dynamic.
Also interesting is his view on resonance, not controlling it but letting it resonate. But each panel with it's own resonance.
Even the top cover is set loose on top so that the air is not captured.
He creates extremely low impedance ground path but finds that to heavy gauge speaker cables also limit dynamics.
Some things sound silly. I find it always interesting when some one dears to be deviant.
He doesn't claim he can proove his point of few scientifically (yet) and describes his work as experimental.
Worth the read for sure.
 
Last edited:
Alright. I found the link to this thread from a post you made in another thread and gave it the benefit of the doubt.

For starters, I think it is plainly wrong what he writes about the FFT. He states it ignores time but time is used in the algorithm to make the response.
In the LPF experiment section he states that a LPF at 100 kHz is audible but no source to back that up.
In the following ferrite core experiment it's the same. -People can hear it but no source of the claim.

I skimmed some of the rest but it appears to me to be solely a pseudo scientific sales pitch for his products. E.G. chassis affecting sound, trapped air inside affects sound somehow..
I didn't find blind tests to back up any of the many claims of "sensing" and audible changes. Maybe I missed them? It should seem like a wasted opportunity to not do them and appear more convincing.

IMO this is deep into snake oil territory.. :)
 
Thanks for reading it!
I don't have the knowledge to judge the FFT results being influenced by time or not.y novice intuition pointed to your comment about it.
The trapped air is a far reach for me to. The resonance approach I find interesting because it goes against accepted solutions. Made me think of the turntable, some are as heavy as possible ( or impossible) some are fragile.
To me snake oil is a step to far. This is a respected engineer and used some his ideas on some TOTL products from Marantz without a lot of marketing talk.
It feels you are heavily on the science proof side. I was to, but became a partial romantic that wants to believe in magic (not yet prooven facts) because I realise that science has to be corrected often enough because of new knowledge.
 
GrisPato said:
Thanks for reading it!
I don't have the knowledge to judge the FFT results being influenced by time or not.y novice intuition pointed to your comment about it.
The trapped air is a far reach for me to. The resonance approach I find interesting because it goes against accepted solutions. Made me think of the turntable, some are as heavy as possible ( or impossible) some are fragile.
To me snake oil is a step to far. This is a respected engineer and used some his ideas on some TOTL products from Marantz without a lot of marketing talk.
It feels you are heavily on the science proof side. I was to, but became a partial romantic that wants to believe in magic (not yet prooven facts) because I realise that science has to be corrected often enough because of new knowledge.
Click to expand...
Yeah, well maybe it is a tad too far to call it snake oil. It's a lot of romanticized prose though.
The FFT is using time indeed to transform. Because a signal has amplitudes stretched over time because it's comprised of different frequencies. Frequency is how many times a positive and negative amplitude is generated pr. sec. So inherently the FFT is using time as a direct factor in its conversion and breakdown of the signal to spectrum.

en.wikipedia.org

Fast Fourier transform - Wikipedia

en.wikipedia.org en.wikipedia.org
1724513331536.png

1724513001108.png
 
In relation to the 100 kHz LPF experiment from the link I've found a study I haven't read before. You might find it relevant. It basically confirms that a 22 kHz tone burst is inaudible.

www.nature.com

High-frequency sound components of high-resolution audio are not detected in auditory sensory memory - Scientific Reports

High-resolution digital audio is believed to produce a better listening experience than the standard quality audio, such as compact disks (CDs) and digital versatile disks (DVDs). One common belief is that high-resolution digital audio is superior due to the higher frequency (> 22 kHz) of its...
www.nature.com www.nature.com

1724570824239.png
1724570842609.png
 
Thanks for explaining the FFT. I once read about it and I think I understand the basic principle now.
What I do like about his story is that he acknowledged that it is a non prooven theory. That he is clear about he allows more noise and more distortion and why he does that.
Ofcourse his reliance on listening is a shortcoming, this should be done blind to have real value.
I am fully aware of the fact that although the sensory input can be the same, our interpretation will depend on our state of mind almost completely. In that I agree completely with you.
Funny thing is that you are the only one that has taken the efforts to read it, although they accuse you from being a now it all.
What means that even in reading your posts there are already biased without being aware of it.
 
Last edited:
GrisPato said:
Thanks for explaining the FFT. I once read about it and I think I understand the basic principle now.
What I do like about his story is that he acknowledged that it is a non prooven theory. That he is clear about he allows more noise and more distortion and why he does that.
Ofcourse his reliance on listening is a shortcoming, this should be done blind to have real value.
I am fully aware of the fact that although the sensory input can be the same, our interpretation will depend on our state of mind almost completely. In that I agree completely with you.
Funny thing is that you are the only one that has taken the efforts to read it, although they accuse you from being a now it all.
What means that even in reading your posts there are already biased without being aware of it.
Click to expand...
Distortion and noise is most of the times a greatly exaggerated measure. I bet you'd be surprised of how much distortion you can't hear with music.

Listening Test

www.klippel.de www.klippel.de


Realising my own personal threshold and how relatively poor that is compared to every component in a quality system made me quite humble. The endless discussions and claims of hearing differences between X and Y component can afterwards be seen as somewhat extraordinary in this light.
My previous results from a year ago was -24. EDIT: it was -30. I think I had the volume higher then. Or less tired. These are fresh. Albeit done quick and dirty..

1724590695616.png
1724590717237.png
 
Last edited:
GrisPato said:
Worth the read for sure.
Click to expand...

I should stopped when I have reached "Also, anyone who likes audio knows that sound changes depending on cables and racks.", but I didn't. Nor did I at one of the F1, car, cooking and sushi analogies. Or at the photo with cable risers. Or at rubber vibration isolation coloring the sound rubbery...

One would think I learned my lesson. Nope! I read the german version as well.

Oy vey!
 
Thanks for reading !
Although I expected people to respond that do feel the same as the author/ engineer.
And to be clear, i will not say he is wrong . . . or right.
Like I said before, I like that he is clear about it being scientifically unproven, experimental. The rubbery sound of the rubber resonance absorbance is/can be a classic example of bias, expecting the effect to be as the material.
I have read some articles written by a former Sansui engineer when they developed the AU-D907 LIMITED (Limited version of the AU-919, Japan only).
They tried aluminium but found it sounded like aluminium. I have no idea what to think about how that should sound, light, volumineus, no idea.
They went with copper plated to improve chassis ground impedance. They didn't mention if it sounded like copper.
I have no problem with designers and or audiophiles looking for sq improvement that they like, just dislike the presentation of it as a fact. If you tested it with your ears, brain and mood, it is solely your interpretation. You can share the experience, but preferably in that context.
Still hoping some members that have similar experiences as this engineer to comment.
 
Baxandall and others who matched the input signal parameters to the amplifier output signal parameters in the time domain made a big mistake. As a result, they measured only the products of nonlinear distortion (THD), which is easily done by standard devices, such as Audioprecision, and the distortion products in the time domain remained unnoticed.
 
kinda looks to me like another take on the specs vs sound argument. Its been done endlessly and even someone like myself who dislikes the subjective aspect of all this has to admit that numbers ain't everything. I dislike that fact purely because it adds a lot of complexity to determining what sounds good and what does not. You have to involve the human meat computer, not just some instruments on the test bench and tweak things to get the best numbers.

I remember sitting at a friend's place listening to turntable platter mats. How do you measure that? I sure don't know, but they absolutely sounded different.
 
gadget73 said:
kinda looks to me like another take on the specs vs sound argument. Its been done endlessly and even someone like myself who dislikes the subjective aspect of all this has to admit that numbers ain't everything. I dislike that fact purely because it adds a lot of complexity to determining what sounds good and what does not. You have to involve the human meat computer, not just some instruments on the test bench and tweak things to get the best numbers.

I remember sitting at a friend's place listening to turntable platter mats. How do you measure that? I sure don't know, but they absolutely sounded different.
Click to expand...
I agree, it is. What is interesting about this to me, that this is a legitimate engineer, experienced with designing and building audio components. I like that he sees it as an experiment and acknowledge that there is no theoretical proof (yet).
The subjective part sure makes it a lot more complicated. Do you trust your ears? Or better said, the interpretation of your mind of what your ears received.
The biggest issue for me is this.
You can only loose the original information from the moment the sound is received by a microphone or electrical by the wire attached to the instrument.
There can be only distortion added and changes to frequency amplitudes by filters (on purpose or accidentally).
So, if the equipment, with NFB, inferior components, etc, that is used to record the music distorts the signal in time and frequency, how can you correct that to the original data? As far as I can see that would be impossible.
Then in light of this designer, how do you reinstall the "soul" of the music when it was already killed by the recording equipment?
 
FiddlerBeeps said:
d having a strong lead singer recording the lead track before recording backup vocalists was best, "technically," but loses a great deal of the interaction between closely mic'd group of singers all performing in the same, sound-damped room. They build upon each other, and that's not something one can add in the mix.
Click to expand...

That doesn't surprise me. The interactions of multiple voices will produce stuff that you can't get just mixing it. Same if you have multiple instruments playing at the same time vs recording them individually and layering it up after the fact. Fair guess you could even see the difference if you used a spectrum analyzer to look at the two side by side.
 
FiddlerBeeps, if I understood correctly from your comments, you do not agree that the fronts (slopes) of sinusoids can rotate in amplitude-modulated signals.
Okay, let's take a signal with frequency modulation.
In this case, the amplitude of the signal is almost constant, but the frequency of the signal changes within certain limits. The wider the frequency deviation, the greater the rotation of the slopes (fronts) of the signal, the greater the SID distortion associated with rotation.
Here are examples of testing using such signals.
p.s.
it's strange that the pictures are not inserted
 
You must log in or register to reply here.
Back
Top Bottom