Latest Thread Images

Thoughts on a bunch of DACs (and why delta-sigma kinda sucks, just to get you to think about stuff)
Mar 6, 2015 at 10:33 PM Post #2,881 of 6,500
souprknowva said:
Mr Moffat, I'd be interested in what you thought about the dynamic performance of the DAC chip you're using, as brought up by sapientiam.
Click to expand...

 
It's all in the spec sheet. There are papers on the Internet on how to deal with the glitching using methods other than sample and hold amps.
 
Mar 6, 2015 at 11:49 PM Post #2,886 of 6,500
baldr said:
  More later but for now..................
 
AD5791BRUZ - Headline specs:  1 ppm 20-Bit, ±1 LSB INL.  We use 2 per channel (1 per phase) to get an honest 20 bit level of performance.  That is four per Yggy.  The BRUZ version is the higher specced model.
 
I know only 20 bits you say?  You can get a 24 delta sigma bit (advertised) DAC  chips for 3% of the cost of one 5791.  Check it out.  Go to Mouser or Digi-Key and see how much AD5791s cost.  Yup, you get just about $400 worth of DAC chips in every Yggy.  I have seen $10,000 dollar D/A converters with $22 bucks worth of dac chips inside.  The Yggy is by far away the best fu***ng parts cost deal going in the arena of high end DACs
 
Biggest problem was figuring out how to get it running without glitching - sample and hold amps sound like ass.
 
Also you have to drive it with DSP because every sample requires a fixed preamble.
 
For the above reasons, I don't expect a lot of competitors to be using it.  After all, even analog devices told me it was not designed for audio.  The best way to get me to do something is to tell me I can't or shouldn't do it.
Click to expand...

 
For the above reasons, I expect you're going to piss a lot of people off in the high-end industry. 
smile.gif

 
Mar 6, 2015 at 11:50 PM Post #2,887 of 6,500
20 bits is heaps though right? Especially for multibit. Iirc the best delta sigma can do 22 bits or something? I doubt the last 4 bits will be missed? Personally I can barely perceive truncation to 16 bits from 24, if I'm doing it right.
 
Mar 7, 2015 at 12:15 AM Post #2,888 of 6,500
purrin said:
TBH, I don't understand why some people think being an audiophile and concert goer are mutually exclusive things.
Click to expand...

 
Because I can only afford one?
 
Though for the kids out there, I highly recommend checking if your local symphony has student rates. Oftentimes a student rate or season subscription is a fraction of the "adult" cost. Sit yourself at row 10 in a concert hall and get a taste of real soundstage with instruments coming at you from all angles.
 
 
 

 
Mar 7, 2015 at 5:22 AM Post #2,889 of 6,500
souprknowva said:
Mr Moffat, I'd be interested in what you thought about the dynamic performance of the DAC chip you're using, as brought up by sapientiam.
Click to expand...

 
Rather than clog up this thread with arcane technical stuff, now Mike's confirmed the DAC chip I'm putting down some thoughts on it on my blog here : http://www.diyaudio.com/forums/blogs/abraxalito/1224-schitts-flagship-dac-uses-not-audio-dac-chip.html
 
Suffice to say here that a DAC's static performance isn't the whole picture, not by a long shot.
 
Mar 7, 2015 at 5:50 AM Post #2,890 of 6,500
sorrodje said:
   
what measurements is wrong and why ?  Just want to understand :wink:
 
How to understand this answer ? (quoted from http://www.head-fi.org/t/693798/thoughts-on-a-bunch-of-dacs-and-why-i-hate-chocolate-ice-cream/2820#post_11384596)
 
Click to expand...

 
I believe his answer that only a 24bit DAC could resolve his signal is wrong. The error stems from mis-reading of an FFT's noise 'floor' which is a very common mistake on audio forums. DACs with many fewer bits than 24 can resolve very low-level signals when viewed on an FFT because the FFT inherently performs averaging over many thousands of samples. Where the 'grass' appears on an FFT plot is not the audio band noise floor of the DAC, rather the grass shows at each point how much noise is in a single FFT bin. An FFT bin has a very narrow bandwidth (typically single digit Hz) whereas audio noise floors are specified over the whole audio bandwidth (say 22kHz).
 
Mar 7, 2015 at 7:05 AM Post #2,891 of 6,500
Sweet Jesus - I go to sleep, come back to this thread and it has exploded ! 
 
The devil in me wishes I could set something up at a meet 3 or 4 months post Yggy becoming available whereby the audience thought they were listening to the Yggy while DAC duties were actually being performed by a Hugo TT.  Whether the latter compares sonically or not is less important than the expectations many would take to such a meet and how they might react if they perceived any of the notorious s-d hash purrin now considers kryptonite to his R2R superpowers. Those naughty people at Chord Electronics just wont leave it alone  :wink:
 

 
Mar 7, 2015 at 7:41 AM Post #2,892 of 6,500
purrin said:
I have to return this item. This was made clear to me from the beginning. zdfx had the first go at it with v0.96. It's now my turn with v0.99.

What makes me different? Surely you have heard that I am the Messiah?
Click to expand...

Heard it now for sure. And since I'm relatively new here I gotta ask: are you still in your pre-ressurection period or did I miss the best years? :)

Btw, doing it for glory is still not "charity for the god of sound". But nevertheless, I still thank you for doing it.


Consider this: I lose of a lot money on my endeavors and website; and most vendors won't dare let me touch their stuff.

P.S.

I also have an axe to grind with "high-end" audio bull-****tery. Starts with a story about 14 years ago when I asked my hifi dealer why we never saw bad reviews in the rags. These sentiments are reinforced every year I attend THE SHOW. So yes, I am doing this largely for the good of sound. Otherwise I'd be starting a true HF competitor website and raking in the dough. It wouldn't be that hard for a person in my position, I could start saying everything is "great", and vendors would be lining up at my door.
Click to expand...

That's the problem with bad reviews, noone wants to talk to you anymore. But I for one definitely want to see those too. So, good luck and keep on going. And if you ever open some sort of donations fund to cover the costs for bad reviews, pls notify.

Also, I would like to nominate yours for The FAQ Of The Year. So much personality ... So much passion ... I wonder who's getting FAQed so hard
 
Mar 7, 2015 at 9:58 AM Post #2,893 of 6,500
prot said:
Btw, doing it for glory is still not "charity for the god of sound". But nevertheless, I still thank you for doing it.
That's the problem with bad reviews, noone wants to talk to you anymore. But I for one definitely want to see those too. 
Click to expand...

 
Not to veer to far from the discussion of dacs - How many people have sought letters of recommendation, or used as "job references" people who would give really honest reviews of their abilities and performance- and not just say good things about them........   Thought so.  Bad reviews do not get written - if reviews are written by people who are blatantly lying or who write stuff because they have been bribed-- different story.
 
Buy things from vendors who allow returns - judge for yourself - problem largely solved.
 
Mar 7, 2015 at 10:03 AM Post #2,894 of 6,500
fzman said:
   
Not to veer to far from the discussion of dacs - How many people have sought letters of recommendation, or used as "job references" people who would give really honest reviews of their abilities and performance- and not just say good things about them........   Thought so.  Bad reviews do not get written - if reviews are written by people who are blatantly lying or who write stuff because they have been bribed-- different story.
Click to expand...

 
Recommendations are generally assumed positive as a genre. That's why it's a "Recommendation" and not an "External Review of Candidate's Potential." 
 
Reviews are not assumed positive as a genre, except - apparently - in high end audio. 
 
Mar 7, 2015 at 10:28 AM Post #2,895 of 6,500
So does this mean the Yggy portion of the R2R ladder DAC is 14-bits?
 
DAC ARCHITECTURE
The architecture of the AD5791 consists of two matched DAC
sections. A simplified circuit diagram is shown in Figure 49.
The six MSBs of the 20-bit data-word are decoded to drive 63
switches, E0 to E62. Each of these switches connects one of 63
matched resistors to either the VREFP or VREFN voltage. The
remaining 14 bits of the data-word drive the S0 to S13 switched
of a 14-bit voltage mode R-2R ladder network. To ensure
performance to specification, the reference inputs must be force
sensed with external amplifiers.
 

 
You must log in or register to reply here.

Users who are viewing this thread

Total: 5 (members: 0, guests: 5)
Back
Top