Thoughts on a bunch of DACs (and why delta-sigma kinda sucks, just to get you to think about stuff)

[twinkle]

....Much obviously is in the circuit implementation, with the Schiit bit perfect programmed DSP digital filtering pulling whatever remaining however niggly DAC resolution (and linearity) shortcoming chestnuts out of the fire.  That's the plan, anyway.

 
Originally Posted by Jason Stoddard /img/forum/go_quote.gif
  That's one plan. Not our plan. 

Such a tease...

 
By the way this is AD1851R-J apparently:

 

[jacal01]

  There goes the easily winnable leg of that bet with jacal01 I proposed.

Yours wasn't a 'not' bet, yours was the AD5791 horse bet.  Were that it was that easy, huh?: "You AD1851; me everything else".
 

Such a tease...

 
That less than helpful sniping is what I get for snarking.  I'm getting a lot of that lately.

 

[StefanJK]

  Yours wasn't a 'not' bet, yours was the AD5791 horse bet.  Were that it was that easy, huh?  You AD1851; me everything else.
 
 
That less than helpful sniping is what I get for snarking.  I'm getting a lot of that lately.

 What?  I'm not betting on AD1851, I'm betting against AD 1851.  The easy part, in my view, is betting against any 16 bit dac chip.  You, if I read you correctly, are/were betting on some 16 bit dac chip.  The hard part is me betting on AD5791, could obviously be something else.  Just not a 16 bit dac chip.

 

[jacal01]

The 'you' in this case is me.  I should have added quotation marks.  Allow me to edit.
 
I doubt if I'll speculate any further, tho.  Why bother?  Guess we'll just have to wait for the warm fuzzies to arrive.
 
However, if it does end up being the AD5791, I'll pay you your $10, no problem.  And props besides.

 

[twinkle]

 Originally Posted by twinkle Such a tease...

  That less than helpful sniping is what I get for snarking.  I'm getting a lot of that lately.

Huh? 

   I'm calling Jason a tease because he's not telling us what their plans are for DAC implementation.

 
Maybe it's a problem with multiple quotes of different people? I'll try to use a clearer form then...

 

[jacal01]

Rampant misunderstandings all around, it seems. I was referring to Jason's post above, too.
 
I guess I was in an emotional dither.  But I'm feeling better now. 

 

[StefanJK]

Thanks.  I only care about props.   Seeing a Yggy in the house would be nice too...else I'll start being tempted to try to learn more by building my own dac from parts.  Might take a while if I don't go with plug-and-play solutions.  I know enought to understand that I don't know what's really going on and what really matters.  Actually implementing a dac would change that.

 

[Tachikoma]

  The 'you' in this case is me.  I should have added quotation marks.  Allow me to edit.
 
I doubt if I'll speculate any further, tho.  Why bother?  Guess we'll just have to wait for the warm fuzzies to arrive.
 
However, if it does end up being the AD5791, I'll pay you your $10, no problem.  And props besides.

Will I get a piece of that too?

 

[twinkle]

  Rampant misunderstandings all around, it seems. I was referring to Jason's post above, too.
 
I guess I was in an emotional dither.  But I'm feeling better now. 

Cool

!!

I think the fact that we can quote someone directly but we loose what they were quoting is what creates the misunderstandings.... we'd need quotes embedded in quotes... (but then redundancies pile up)

 

[jacal01]

Sure, no prob.  I'm not worried. 

 

[StefanJK]

 
Will I get a piece of that too?

Yes, Tachikoma has priority.

 

[jacal01]

Cool

!!

I think the fact that we can quote someone directly but we loose what they were quoting is what creates the misunderstandings.... we'd need quotes embedded in quotes... (but then redundancies pile up)

 
Me, I'm quote averse, if it's directly above mine.  I knew what you meant, just imprecise phrasing on my part.

 

[Baldr]

A Definition:
 
Bit Perfect – in a closed A/D system, a give analog level with a defined maximum and minimum is converted to a number. What is significant is what is the bit resolution and speed of the converter. In an 8 bit case, there are 256 possible numbers – a 16 bit case yields 65,528 possible numbers. That number of numbers doubles with each additional bit. If the A/D converter (case 8 bit) yields 256 numbers from 1 to 256 (or more accurately, 0 to 255) then there are no missing codes; the device works for coarse MRIs or weapons. The D/A converter in this perfect system then converts these numbers back to analog levels which all should be unique according to the decoded numbers. There should be no missing or duplicated levels; this is Bit Perfect. A goal for high end products. No sonic glare; unbelievable detail levels.
 
This applies to multibit A/D and D/A converters only. At the higher bit and speed levels required for audio resolution, this becomes expensive. Hence the development of “audio” parts (Sigma-Delta A/Ds and Delta-Sigma DACs). Even worse is DSD, which I have previously addressed. These are offered by all of the “audio” chip makers, complete with reference designs and “Howto” data sheets that make it possible for fourth graders to build them as class projects. They are cheap, and have resulted in digital audio technology that is nearly as universal as it is insipid. That's not to say that a builder can't add “designer” capacitors, over-designed analog sections or power supplies, fancy over-machined front panels, water-cooling, palletized delivery, jewels, etc., etc, ad nauseum. This sort of extravagance is perfect for the user who wants to invite people over to have his guests admire the piece first. Unfortunately, even though you have wrapped plastic around the vile-smelling “audio” parts, they still have the same performance stench.
 
A good analogy is a tire. You can have the best performing car in the world and easily kill yourself if you have poorly designed tires. Now, do you invite all of your friends over and say “Look at my tires”? Of course not! All you care about is their performance. But I digress..........(Good thing Jason is around to make sure the Schiit stuff looks absurdly good.)
 
An SOF (Schiit only feature) – The Schiit Footlong Mega Burrito Supersauce Digital Filter:
 
It is a digital filter/sample rate converter designed to convert all audio to 352.8 or 396KHz sample rates so that it may drive our DACs. You get it from us; it is our filter. It keeps all original samples; those samples contain rudimentary frequency and phase information which can be optimized not only in the time domain but in the frequency domain. We do precisely this in the Yggy with said filter; this is the reason that on good recordings through Yggy you can hear the hall, its dimensions, and the exact position of anyone coughing or farting in the room, the motions of guitars being hoisted in preparation of being played, sheet music pages being turned, etc. etc. This comes from our mega burrito filter. A friend of mine, Jonathan Horwich, sells analog master tapes in ½ track form – at least 15 IPS, and 30 (I believe) as well. On those analog masters, you can also hear the entire environment before the music starts – what is amazing there is that even if on accounts for hearing “down into” the analog noise, the S/N indicates a 14 bit performance at best for those tapes. 14 bit or not – those tapes, totally scratch my itch. If you want that, we got that and more in the Yggy.

 

[jacal01]

Thanks for the cogent explanation, Mike.  You didn't know that educating the unwashed masses was part of your job description, did you?
 
But I think you said 3 things:
 
1.  Putting lipstick on a pig is exactly that.
2.  The DAC chip you're using can and does process data sampling at 352.8/396 kHz.
3.  Input sample data at 16 bit/44.1 kHz still has all the sonic information so that you can hear and place all the farts in the room, and that in the interpolation oversampling it to 396 kHz, none of the original information is lost or distorted.
 
Did I about get it right/close?  Should we be looking at a multi-bit 352.8/396 kHz data processing DAC chip then?  Do you want guacamole with that burrito? 

  

 

[Baldr]

  Thanks for the cogent explanation, Mike.  You didn't know that educating the unwashed masses was part of your job description, did you?
 
But I think you said 3 things:
 
1.  Putting lipstick on a pig is exactly that.
2.  The DAC chip you're using can and does process data sampling at 352.8/396 kHz.
3.  Input sample data at 16 bit/44.1 kHz still has all the sonic information so that you can hear and place all the farts in the room, and that in the interpolation oversampling it to 396 kHz, none of the original information is lost or distorted.
 
Did I about get it right/close?  Should we be looking at a multi-bit 352.8/396 kHz data processing DAC chip then?  Do you want guacamole with that burrito? 

  

I would restate number 3 as follows:
Input sample data at 16 bit/44.1 kHz contains rudimentary sonic information.  A time domain and frequency domain optimized filter (as in the Yggy) enables one to hear and place all the farts in the room far more precisely.