New 28-bit DAC coming out.

[antcollinet]

Yes, that's the part of the concept
From product description: "It is known from psycho-acoustic research that broadband noise can be inaudible when it is below a certain program level due to masking."

Nothing to do with masking. Once any signal (noise or otherwise - is below a certain level, it is inaudble - even in complete silence. Human ears are not infinitely sensitive.

The limit of human audio sensitivity is around 0dBA. So if SNR is 120dB - even if your volume is set so full scale music while playing is at 120dbA peaks (seriously, if you value your hearing you don't want to be doing this), then the noise is going to be inaudible even if the music is silent.

Now tell me again why we need 160dB of SNR?

 

[Sokel]

Nothing to do with masking. Once any signal (noise or otherwise - is below a certain level, it is inaudble - even in complete silence. Human ears are not infinitely sensitive.

The limit of human audio sensitivity is around 0dBA. So if SNR is 120dB - even if your volume is set so full scale music while playing is at 120dbA peaks (seriously, if you value your hearing you don't want to be doing this), then the noise is going to be inaudible even if the music is silent.

Now tell me again why we need 160dB of SNR?

I can imagine a digital x-over with bits to spare for any imaginable use to match with the shorts of a 8381A or a multiway big Wilson actively amplified.
(no,good enough is not good enough for me )

 

[Multicore]

Now tell me again why we need 160dB of SNR?

Depends on who "we" is. For users of finished consumer audio products I feel that CD format is sufficient but my Boss RC-505mkII uses FP32, which makes sense to me given the practically unlimited layering it can do and the uncertainties of performance and therefore of inputs. Now, when it comes to converting that to something to listen to I guess ranging/scaling is going to be required so I guess maybe the kind of tech incorporated in this device could be useful for applications like that. And I can't blame the inventor for doing a bit of geewizz marketing to get attention.

 

[antcollinet]

I can imagine a digital x-over with bits to spare for any imaginable use to match with the shorts of a 8381A or a multiway big Wilson actively amplified.
(no,good enough is not good enough for me )

Then it's not good enough. It can't be good enough and not good enough at the same time - you seem to have gone a bit meta there mate


Or put it another way - one of the biggest engineering challenges : determining "what is good enough"

For audio (reproduction) though, I'd have thought it was pretty simple:

It is good enough when the imperfections cannot be heard - at all - by anyone - ever. For digital audio, from my understanding, that is 24/48. Actually, probably 20/48, or even 18/48 - but those options don't exist.

 

[KSTR]

I wonder it there exist a product, that can be measured.

We will have to wait until December, when the first units are planned to be shipped, according to their website stub.

I'm an engineer. Good enough is all that is needed. Any more is a waste of resources.

Well, 640kB of memory was considered good enough. Analog landline telephone was considered good enough, analog TV, what have you... the list is endless.
I do agree, though, that "infinite resolution" multipath DACs are a little less important than multipath ADCs which are already in heavy use for decades (German company StageTech pioneered this with their "TrueMatch" technology, a 4-level multipath ADC design). But they are convenient and that is enough reason to exist.

 

[solderdude]

For the recording side this makes sense as that ensures it will be easier to record dynamic signals without having to worry about headroom nor noise.
This is not the case with reproducing a finished product (recording) ... unless you want immediate absolute silence in your car that just reproduced some low subbass at 150dB SPL (where you can make the long hairs of the obligatory girlfriend wave all around the car).
Advancing engineering is fun and interesting but for home usage and reproducing recorded material it isn't really relevant.

 

[KSTR]

Why do you think this product is only intended for playback the final CD (or whatever) quality at home?
This is universal lab/studio gear.

 

[phofman]

I can imagine the use case in generating signals for measurements. A nice deep level sweep without having to switch the output relay attenuator along the pass.

 

[popej]

Now tell me again why we need 160dB of SNR?

I hope you have noticed the smiley?
There could be other applications than home audio, for example measurements. Except reference to psychoacoustics make them improbable.

 

[holdingpants01]

I have an idea, add a gate or relay that physically switches off the output below certain input level, boom infinitely low noise! Just like OLED TVs have crazy contrast ratio in the specs just because black image means the pixel is turned off

 

[boXem]

That's indeed what I was calling 'smart'.
If one removes the real time part and the adc feedback for clickless switching (that's at least how I understand it), what is the advantage of the multipath over a single path with various analog gain/attenuations and associated digital gain?

Replying myself after having thought a bit more about the concept.
Multipath allows the smart stuff. It seems extremely difficult to synchronize digital gain changes with analog ones without interrupting the stream.
Paralleling suppresses the digital gain changes so there is no more need of synchronization.
Doesn't make the system less impressive.

 

[Scytales]

To my mind, caveat is : assessing dynamic range of such a system with standard AES17 test (measurement of actual dynamic range with a -60 dFS test signal to avoid a particular DAC to mute its output with no signal and add 60 dB to the measurement result) may not be that relevant because of the signal dependent noise floor modulation.

A -60 dBFS signal will not triggered the upper path, thus such a measurement method will never see the noise floor modulation above the threshold where the upper path is put in the signal path.

A THD+N sweep against input level may actually give a more relevant assessment of the performance of this system.

What do you think ?

 

[Sokel]

Judging by the rule of thumb about a measuring system (at least 10dB better specs than the DUT overall) there are all shorts of shortcomings to it.
It would need something reversed comparable.Is there such a system?

 

[KSTR]

Or just simply noise spectrum vs. DC level. OTOH, such a DAC might have a DC-blocking digital filter...

 

[KSTR]

Judging by the rule of thumb about a measuring system (at least 10dB better specs than the DUT overall) there are all shorts of shortcomings to it.
It would need something reversed comparable.Is there such a system?

StageTech's ADC. Or low noise preamp in front of the AP.

 

[Hayabusa]

130 dB SNR is somewhere between 21 (126) and 22 (132) bits. 28 bits would be 168 dB.
Or somebody is playing with numbers, or my maths are really rusted.

indeed, 160db plus is very unlikely too happen

 

[antcollinet]

Well, 640kB of memory was considered good enough

Until it wasn't

Vinyl was considered good enough until it wasn't.

Next year's product will never be good enough until it is better than last year's.

Good enough can and does change. For some people who don't' understand either the engineering or human auditory capabilities (the "everything matters" crew) 24/48 isn't good enough. Nor is 24/96, or even 24/384.

For someone who does understand those things, more will (should?) never be needed until humanity has evolved (or installed??) better ears.

 

[Sokel]

Until it wasn't

Vinyl was considered good enough until it wasn't.

Good enough can change. For some people who don't' understand either the engineering or human auditory capabilities (the "everything matters" crew) 24/48 isn't good enough. Nor is 24/96, or even 24/384.

For someone who does understand those things, more will never be needed until humanity has evolved (or installed??) better ears.

Need is not a metric belonging to a hobby.
For needs alone,yes,you're right.
A hobby is about fun and awesomeness and pushes the "what can be done" to it's limits. All hobbies,not just audio.
Covering a need is ok.But boring.And we people don't go forward by boring.

The end example,my grave will probably be 2 x 4 meters.But still I have the Khufu pyramid to compare it*

*Edit: or better still the Qin Emperor's one.

 

[antcollinet]

Need is not a metric belonging to a hobby.
For needs alone,yes,you're right.
A hobby is about fun and awesomeness and pushes the "what can be done" to it's limits. All hobbies,not just audio.
Covering a need is ok.But boring.And we people don't go forward by boring.

The end example,my grave will probably be 2 x 4 meters.But still I have the Khufu pyramid to compare it*

*Edit: or better still the Qin Emperor's one.

You seem to be saying:

"I know that the difference from going to more than 24/48 can't be heard by a single human being on the planet - but I want it anyway"

I find that weird: To me, it makes no sense. It has nothing to do with boredom or excitement. I'm never going to get enjoyment by changing something I know neither I nor anyone else can hear.

It is just something I can forget about to focus on things that *do* matter for this hobby. That is where I find "fun and awesomeness" Pushing things to the limit that actually make a difference.

 

[sergeauckland]

Need is not a metric belonging to a hobby.
For needs alone,yes,you're right.
A hobby is about fun and awesomeness and pushes the "what can be done" to it's limits. All hobbies,not just audio.
Covering a need is ok.But boring.And we people don't go forward by boring.

The end example,my grave will probably be 2 x 4 meters.But still I have the Khufu pyramid to compare it*

*Edit: or better still the Qin Emperor's one.

This makes no sense to me. Who cares is something is 'better' if the existing one is Good Enough? Good Enough means one doesn't need anything better, it's Good Enough!

None of my hobbies are about 'fun and awesomeness', they're about challenge and intellectual stimulus or the satisfaction of having done something difficult. Sometimes it's about doing something that to me is trivial, but that just raises the sum total of human happiness.

If that's considered boring and unawesome, it's Good Enough for me.

S.