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

Not the first time Jason has used the 'it's a plan. It's just not our plan' line, but it's gold nonetheless. 

Hello Professor Baldr ,
 
Sonic Glare , oh my , HP just passed and here you are providing a Lexicon for DAC reviewers , the torch has passed to you , dear man ! 
  I am delighted to read your description and promise of Good Things coming , I'm holding off further Audio investments till these Good Things are sitting on my Desk , delivering  the promise of Digital's unrealized potential . 
  You are setting-up a rather high level of anticipation , if your words = performance in the hands of your followers ; You will be Enshrined . 
  Audio Opium for the new classes of music reproducers , coming from the Headphone segment - no less , people will whisper your name ( "where did JS find him" ? ) , and "I want him to Autograph my T-Shirt , honey get the camera , I wanna picture of this !  
  I have listened to the products from the lads up in Watsonville , their stuff lacks Sonic Glare , I felt their Ladder performance was something like the Canon 1d images printed by Epson's High Dynamic Range Printing technology compared with a Prosumer Canon and a lessor Epson printer , Sharpness , color fidelity , detail , depth , realism all astonishing and within reach but pricy , very pricy indeed , requiring buying customers . 
  You won't have difficulty finding the buying customers , they will find you , your word of mouth is already strong .  
  I suspect you have a working example in your living room , active and being listened to , how else could you be inspired to coin "Sonic Glare" and I hope you find a way to describe Piano harmonics and Cymbals Crashing & Splashing along with all the other sounds that musical instruments create , our reviewers need you to enhance their "tired and worn-out" phraisiology .  Providing us , the Un-washed masses , an ability to describe what we hear will be a god-send , oh-dear , I'm calling you a god ! , let me wash my mouth out , shame on me !!  
  14 Bit depth , I'm in , hell I've been ok with the 7 or 8 from Vinyl ( back in the day ) .  Who has electronics and transducers that go beyond 20 Bit depth ? , I know OPPO claim that much but is it true and where is the musical proof ?  Genelec & Focal have Active Monitors that promise beyond 120 db , half field , for mega $$$$$ but even a simi-anichoic room will have a 30db noise floor , add 84db to that ( 14 bits ) and we've hit 120 db , so , in practice , 14 Bit depth will work just fine , I think . 
  Piano reaches far beyond 20 bits , drum kits can sail another 20db higher yet :  Compression stays with us , how can we ever realize listening with electronic devices that feature 30 Bit depth ?  
  I'm feeling that we will never be able to reproduce Live Performances , it's not in the Cards , I'm not hoping or trying to hope , it's not worth contemplating , let HP experience it from the Angel's Choirs but it's not to be here on Earth . I'll settle for a good performing DAC and then I'll discover just how good my other "chain" of components actually are . 
  Buckle your seat belt Professor , you're in for a wild year ! 
 
Tony in Michigan 

I've heard the stuff from Watsonville too.
Yup, their stuff lacks sonic glare.
Smooth and butter,
Yet not as resolving.
In the larger scheme of things,
Especially when compared to Yggy.
Beware though because high resolution
from fancy tech and algorithms won't make
bad recordings sound good.
Quite the contrary in fact.

baldr said:

  ...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.

Click to expand...

 
Seriously, I want to give you my money right now....

Gents,
 
I cannot get too engaged at this point; given my efforts to finish Yggy and get it to market.
 
Let me just say that being in the business of building audio reproduction equipment allows no, nada, ******-all control of whatever the recording engineers did or didn't do, what equipment they used, whether it was originally analog or not, what microphones were used, what and how it was digitized, how it was processed, etc., etc.
 
Nor does it solve who or what was done to various issues of identical recordings to make them sound different, etc. etc.
 
Maybe I assume too much, but I accept it as a given that there are recordings of a very wide spectrum, from god-awful to sublime. It has been so as long as I have been addicted to this hobby. It is a constraint we must live with if we are to be audiophiles.
 
In the old analog days, we used the best components we could afford to give us the best possible sound. Everybody in the hobby knew they could not fix bad recordings. I thought that was yet obvious today.
 
Now I almost offer (next 90 days or so) a D/A converter. It has a very special digital filter/sample rate converter that is only available from Schiit. It is neither magic nor faith based. It neither raises the dead nor makes bad recordings sound good. There is no smoke, mirrors, or doves spontaneously appearing. It is pure science, and it is amazing because the technology was contributed over a 70 year period, from the 1910's until the 1980's. It exists because I am stubborn and kept going, finding new geniuses when necessary in the quest of trying to make digital sound better than analog.
 
Digitally, it takes nothing away from the original information. Nothing, nada, ******-all. It then takes a weighted average of the original samples and adds frequency (read flat) and time (read image) extra info between the samples to convert the samples to 352.8/396KHz. All complete calculations – NO approximations. All info is a function of the original. Real math – hard science. Not psychology or social science. 2 + 2 = 4. Now and forever.
 
The result is a D/A converter that images like nothing I have ever digitally heard. The promise is that with better recordings (Cowboy Junkies, for example) you hear the entire environment. If you check it against photos of the original session (often available as part of the LP/CD documentation or online), you may be shocked.
 
That's what Yggy digitally does. Period! (Pardon the shouts) IT DOES NOT MAKE BAD RECORDINGS SOUND GOOD. If you let it warm up all the way, IT DOES NOT MAKE BAD RECORDINGS SOUND WORSE. If you are listening to a lot of bad recordings, you may try stamp collecting or another hobby. You do not have to believe in the tooth fairy, the easter bunny, or swing dead chickens around your head while dancing nude and covered with moose dung in the Alaskan tundra in February.  Flippin' science.
 
There is no way to fix a bad recording, for now and ever shall be. Amen
 
Now to get back to finishing it!!!

I can only add this to MIke's post re bad recordings - if you read any of Mark Waldrep's rants you will probably despair for the future of recorded music beyond jazz and classical. When an engineer is repeatedly 'bludgeoned' into making a recording louder by a suit from a record label, that's a cancer that needs to be cut out. All this talk of whether a DAC is capable of 14/16/20 bits resolution is great stuff till you realise the brickwalling an album is no barrier to chart success. I previously blamed this on the production efforts of Rick Rubin but I dont believe he acted alone.....
 
http://en.wikipedia.org/wiki/Californication_(album)
 
I find some of the results in the Dynamic Range Database quite odd but they appear to have nailed the original CD recording of RHCP's most commercially successful outing. 
 
http://dr.loudness-war.info/album/list?artist=Red+Hot+Chili+Peppers&album=Californication
 
For the pros and cons of the numbering system used to allocated the various colors, the discussion which followed this Stereophile piece makes some very good points re purchasing decisions made on the basis of a set of numbers vs actually listening to the music. 
 
http://www.stereophile.com/content/unofficial-dynamic-range-database

To be fair, I think a lot of pop recordings would sound worse if they were recorded with proper dynamic range. Just imagine an audiophile recording of Gangnam style, for instance.
 
Thankfully, music that does benefit from being recorded well, usually isn't compressed to hell.

Quote:

tachikoma said:

  I'm guessing...
http://www.analog.com/en/digital-to-analog-converters/da-converters/ad5791/products/product.html
 
Not an audio DAC, but is designed for military/aerospace.
 

Click to expand...

 
Quote:

stefanjk said:

  This looks like a decent guess to me.  How common are voltage output dac chips?  Mofat seems to have hinted strongly that he's using a 'volts', which I take to mean no i-v conversion stage and high voltage (which this chip does up to 33 V).  20 bit is 'good enough' for what we want.
 
This write-up appears to be relevant:   The 20-Bit DAC Is the Easiest Part of a 1-ppm-Accurate Precision Voltage Source:

Click to expand...

 

purrin said:

  The only problem with that one is it doesn't take standard digital i2s or even anything resembling audio in digital format. Look at the write mode timing diagram. Doesn't exactly take 44.1khz straight up and need to run through a few loops to load up all the bits in word.

Click to expand...

 
I'm starting to warm up to this DAC chip.  A lot.  Recently released to the wild (~2010), true 20 bit resolution, R-2R architecture, ≤ ±1 (±0.25 typ) LSB INL, up to 34.3 V output, it pushes all the right buttons.  It has an unbuffered voltage output to feed into the external buffer analog stage re. Mike's post.  The single as in "...single 20 bit..." most likely refers to single channel in this case.
 
As to purrin's input data format argument, the chip has a 3-wire serial interface (SYNC, SCLK, SDIN) that is compatible with standard SPI and "...most DSPs..." (datasheet page 19).  Sound familiar?  All the frequency and phase oversampling (~8x) is done in the DSP programmed megaburrito supersauce digital filter at the front end, so it makes sense that its output is what feeds into the DAC chip, and even more so as an AD DSP compatible output to this AD DAC input.
 
I bet Schiit is pretty smug that this new generation high resolution/linearity R-2R DAC chip is now available and in time to be complementary to the closed form frequency and time domain data stream optimization algorithms that they've protractedly developed.  A high precision multibit D/A conversion chip downstream stable mate to the bit perfect digital filter/sample rate converter for their SOTA DAC product.
 
Looks like Tachikoma's got this one, and I may be out $20 here.

ADI & the AD5791 DAC chip ,
  Everyone buckled in ? , the DAC world is in for another advance , cross the board , probably game changing , the little dragonfly type things may be about to share space in the misc. drawers with the un-used RCA interconnects and record cleaning brushes .  
  The ADI chip costs $38 in one thousand lots , hmm , what will an upgraded Bifrost end up costing ? and what will this do in terms of those $2,500 Portable units out there with Gordon Rankin designs in their Chipsets ? .  
  Oh , Mr. Baldr , there is sooooo much to talk about now !  
 
  Tony in Michigan 

I don't think that a USB input converter/DSP processor filter/R2R DAC IC chip sandwich board will ever be able to fit into a Dragonfly type flash drive sized plug-in configuration, but I think that a USB & SPDIF input portable DAC/amp with that setup, considering how little or no analog gain stage is required for headphone output, could be a viable future product.  

baldr said:

..., jewels, etc., etc, ad nauseum.

Click to expand...

Mikologism:
 
Ad nauseum: museum-quality nausea-inducing repetition.
 
 
(edited for clarity)

baldr said:

  Gents,
 
I cannot get too engaged at this point; given my efforts to finish Yggy and get it to market.
 
Let me just say that being in the business of building audio reproduction equipment allows no, nada, ******-all control of whatever the recording engineers did or didn't do, what equipment they used, whether it was originally analog or not, what microphones were used, what and how it was digitized, how it was processed, etc., etc.
 
Nor does it solve who or what was done to various issues of identical recordings to make them sound different, etc. etc.
 
Maybe I assume too much, but I accept it as a given that there are recordings of a very wide spectrum, from god-awful to sublime. It has been so as long as I have been addicted to this hobby. It is a constraint we must live with if we are to be audiophiles.
 
In the old analog days, we used the best components we could afford to give us the best possible sound. Everybody in the hobby knew they could not fix bad recordings. I thought that was yet obvious today.
 
Now I almost offer (next 90 days or so) a D/A converter. It has a very special digital filter/sample rate converter that is only available from Schiit. It is neither magic nor faith based. It neither raises the dead nor makes bad recordings sound good. There is no smoke, mirrors, or doves spontaneously appearing. It is pure science, and it is amazing because the technology was contributed over a 70 year period, from the 1910's until the 1980's. It exists because I am stubborn and kept going, finding new geniuses when necessary in the quest of trying to make digital sound better than analog.
 
Digitally, it takes nothing away from the original information. Nothing, nada, ******-all. It then takes a weighted average of the original samples and adds frequency (read flat) and time (read image) extra info between the samples to convert the samples to 352.8/396KHz. All complete calculations – NO approximations. All info is a function of the original. Real math – hard science. Not psychology or social science. 2 + 2 = 4. Now and forever.
 
The result is a D/A converter that images like nothing I have ever digitally heard. The promise is that with better recordings (Cowboy Junkies, for example) you hear the entire environment. If you check it against photos of the original session (often available as part of the LP/CD documentation or online), you may be shocked.
 
That's what Yggy digitally does. Period! (Pardon the shouts) IT DOES NOT MAKE BAD RECORDINGS SOUND GOOD. If you let it warm up all the way, IT DOES NOT MAKE BAD RECORDINGS SOUND WORSE. If you are listening to a lot of bad recordings, you may try stamp collecting or another hobby. You do not have to believe in the tooth fairy, the easter bunny, or swing dead chickens around your head while dancing nude and covered with moose dung in the Alaskan tundra in February.  Flippin' science.
 
There is no way to fix a bad recording, for now and ever shall be. Amen
 
Now to get back to finishing it!!!

Click to expand...

 
Are you doing anything to minimize the possible jitter added by this math, and are you addressing jitter at a system level? Maybe a stupid question, but it would be a big turnoff if I read about all these interesting filter ideas and under the hood was 10M rubidium clock 

Antelope Clock ,
 
why would that be a turn-off ? ,  the Cost of the Clock is 4 or 5 times the total cost of the Schiit device .  
Any thoughts on the Antelope Pure 2 ? 
 
Tony in Michigan 
 
ps.  I don't think Schiit is trying to Control an entire 64 Channel Stage Presentation of any A National Touring Group , they may have a Quartz Oven though .  So far , Schiit designers haven't hinted at their Clocking requirements 

rubidium is useless for audio applications, and I was making a jab at blah blah blah
 
They haven't mentioned clocking which is why I'm bringing it up

When did we stop using the 'Spoiler' method for quoting large posts and can we please revert to that ? The only thing that irks me more is when someone quotes a massive post containing a collection of images - it's just not good netiquette folks.