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

[purrin]

  So you've now said repeatedly.  The technical details are what make the 'magic sauce'.  Mr. Moffett certainly has some strong opinions about it.  And purrin's 'techno babble' remark was aimed at DSD more specifically, not electronics discussion in general.

 
Also referring to the digital filter algorithm which I find much more interesting than guessing the R2R (or whatever it may be) implementation. It's the reason why I picked up a Theta Gen V. I'm kind of following the designer back in time. Mike's tales of yesteryear about filters with a divide by zero problem solved by a guy who drove a crappy car got me interested. I figure the digital filter going into the Yggy might somehow be related to one in the Gen V. It's hard to find literature these days on the Gen V, the Theta did use their own DSP for digital filter relying on motorola processors rather than rely on the usual suspects back then such as SM5803AP, PMD100, PDM200, etc.

 

[Byrnie]

AMB Gamma 2 as well. Not on the list: the AK120, which I loved and preferred over the AK240.

Just to clarify, you're referring to the original AK120 (meaning not the AK120II)? I ask because I heard warren and mercer preferred the original AK120 to the AK120II.

 

[purrin]

The original. 

 

[twinkle]

Originally Posted by jacal01 /img/forum/go_quote.gif
Quote:

 Originally Posted by twinkle You know you might be right though: that analog implementation looks pretty serious, even sans tubes.

 
Who are you addressing here, me or estreeter?

Sorry, I meant I agree with estreeter.
 
To clarify, I think the main anticipated strength of Mike's design is in the closed form algorithm.  While the choice of DAC chip and/or other converting hardware will affect ultimate resolution, I think it's in encoded filters that you can reach that "organic", bit-perfect, artefact-free, non-destructive reconstruction. And then the analog section implementation does the remaining 60% of the magic

 
Still good detective work on your part Jacal

 

[Clemmaster]

   
Mr. Moffett certainly has some strong opinions about it.

 
I assume, Mr. Mosfet and Mr. Mosfat do, too 

 

 

[Byrnie]

The original. 

Nice and they sell for a nice price too!

 

[twinkle]

   
Also referring to the digital filter algorithm which I find much more interesting than guessing the R2R (or whatever it may be) implementation. It's the reason why I picked up a Theta Gen V. I'm kind of following the designer back in time. Mike's tales of yesteryear about filters with a divide by zero problem solved by a guy who drove a crappy car got me interested. I figure the digital filter going into the Yggy might somehow be related to one in the Gen V. It's hard to find literature these days on the Gen V, the Theta did use their own DSP for digital filter relying on motorola processors rather than rely on the usual suspects back then such as SM5803AP, PMD100, PDM200, etc.

Ok, now that is much more useful than the nonsense in a previous post:

 I was curious about certain special sonic qualities of the Gen V and Mike told me [redacted] I grabbed the Gen V on the hunch there would be some similarities with the Yggy because I wanted an Yggy-like DAC before Q1 2015. It turns out [redacted]

 

[jacal01]

   
Also referring to the digital filter algorithm which I find much more interesting than guessing the R2R (or whatever it may be) implementation.

 
That's funny now considering your earlier posted investigation into the square processor chip in the middle of the Yggy board foto.
 
Quote:

 
Yup. Now you know what I was thinking. No luck though. As I've said, been trying to bait Jason and Mike. They don't budge.

 
Do you want me to now find the 1917 H.K. Erlang "Solution of some Problems in the Theory of Probabilities of Significance in Automatic Telephone Exchanges" paper, instead?  

 

To clarify, I think the main anticipated strength of Mike's design is in the closed form algorithm. While the choice of DAC chip and/or other converting hardware will affect ultimate resolution, I think it's in encoded filters that you can reach that "organic", bit-perfect, artifact-free, non-destructive reconstruction. And then the analog section implementation does the remaining 60% of the magic.

 
As did I in a previous post (#1720), ala my "epiphany".

 

[StefanJK]

   
Also referring to the digital filter algorithm which I find much more interesting than guessing the R2R (or whatever it may be) implementation.

But doesn't the filter algorithm you want to implement depend on the R2R implementation?  The filtering you want to do depends on the DAC architecture, with less noise coming out of the dac you can focus on 'nice to have' filter aspects.  I'd guess that the Yggy filter may not be great on other dacs.

 

[purrin]

Ok, now that is much more useful than the nonsense in a previous post:

 
I wanted people to figure it out themselves. It's no fun when someone points you straight to the answer. I'm still surprised that no one has asked me what I thought of the Yggy's sonics.
 

That's funny now considering your earlier posted investigation into the square processor chip in the middle of the Yggy board foto.

 
Well square chip doesn't mean anything, and it's hard to tell what's surrounding it in that fuzzy pic. For all we know, it could still be this, which happens to be small and square:
 

 
We can make out three huge blocks of iron though. Three transformers? Probably a good thing.

 

[BirdManOfCT]

   
I wanted people to figure it out themselves. It's no fun when someone points you straight to the answer. I'm still surprised that no one has asked me what I thought of the Yggy's sonics.
 
 
Well square chip doesn't mean anything, and it's hard to tell what's surrounding it in that fuzzy pic. For all we know, it could still be this, which happens to be small and square:
 

 
We can make out three huge blocks of iron though. Three transformers? Probably a good thing.

Hey, what do you think about the Yggy's sonics?

 

[Sorrodje]

 
Hey, what do you think about the Yggy's sonics?

 
+1
 
Hey Purrin what do you think about the Yggy's sonics ?
 
 

 

[jacal01]

  But doesn't the filter algorithm you want to implement depend on the R2R implementation?  The filtering you want to do depends on the DAC architecture, with less noise coming out of the dac you can focus on 'nice to have' filter aspects.  I'd guess that the Yggy filter may not be great on other dacs.

 
Well, from my perusal of the DAC datasheets, it appears that being able to interface a programmable external digital filter is a key element of its architecture along with the usual conversion fidelity specs.  That and left justified PCM I2S data input.

 
I'm guessing that the closed form "bit perfect" time domain data stream interpolation with 18K+ filter taps digital filter would be a great match for most DACs, as opposed to finite truncated function waveform approximations.

 

[MohawkUS]

What are the chances they have something in place similar to the Musical Surroundings MYDAC II? It uses a form of EQ before and after digital conversion so that the lower frequencies are converted in a similar way to the higher frequencies, supposedly allowing them to be converted more accurately. That could explain the talk of 16 bits really being limited to 12 bits, no?
http://www.theabsolutesound.com/articles/musical-surroundings-mydac-ii-dac/

 

[arnaud]

   
+1
 
Hey Purrin what do you think about the Yggy's sonics ?
 
 

+2  Do you still have access to a prototype Purrin? From reading early on, it seemed - unit not quite warmed up - you felt it was rather neutral sounding, not bass light but not what you'd call a warm sounding dac. Then, the resolution galore also pointed away from a sirupy / forgiving type of sound - is that still the feeling?