[kstuart]

  Author's precis: Mike Moffat will certainly cover this in his own blog, but he's still got a few decades of catching up with the present to do...so, until then, you have my perspective on the Bifrost Multibit.
 
[size=small]2015, Chapter 17: [/size]
The Multibit Revolution Gets Cheap
 
[...]
  Plus, going discrete in the Bifrost Multibit was completely different than our other DACs:
 

• The discrete stage in the Yggdrasil and GMB are both just buffers. The DACs used there have voltage output. So, they are very simple discrete stages, just four active devices per channel. However, as measurements clearly show, this simplicity does not compromise distortion performance (this is usually the penalty paid...simple discrete amps typically have high THD.) That's why when you see some "2-PPM wonder amp" it usually has about 80 active devices per channel. We can argue till the cows come home which sounds better.
• The discrete stage in the standard Bifrost and Gungnir is actually a small amp stage—not exactly a discrete op-amp, due to its very specific gain structure and open-loop bandwidth beyond the audio band--but it also takes a voltage output of the DAC, amplifies it a bit, and passes it on. No I/V necessary.
• The AD5547 is a multiplying DAC, which means it needs two very well-matched gain stages for exceptional accuracy, not just a single I/V converter. Which means it would need twice the circuitry. Fully discrete? As I mentioned, aha, no.

 
So…even with surface mount components, the math was simple. It all pointed to “nope, sorry, divide by zero, sums to nothin, system is unstable, ain’t gonna work around here.” Anyone looking for a comparison to a mythical discrete-output board for Bifrost Multibit will be disappointed…because we simply never tried to do it.*
 
*And no, there is no conspiracy where we’re holding back a better analog board that will magically appear in 6-8 months, when sales start slowing down in the spring.**
 
**Now, that doesn’t say we’ll never have a better analog board for Bifrost. Hell, it’s possible that even better multibit DACs will be available in the future. Or a new delta-sigma technology will force Mike to recant and go 100% pro-DSD. Though the third sentence is, I think, about as likely as a meteor hitting me on the head before I finish this current sentence.***
 
***And, sentence finished. There you go.
 
[...]

Jason - Thanks for that description of some of the thought processes behind the design decisions for the Multibit Bifrost.
 
There was one more detail of interest, which I am hoping you can shed some light on.  The specs page says:
 
Analog Stage: precision I/V converter and output buffer based on AD8512
 
Can you talk about the decision to use the AD8512 ?   Did you consider any other op-amps ?   Or were there some electrical reasons why that part would be the right one ?
 
Thanks.
 
PS  I am continuing to enjoy the Multibit Bifrost.  BTW, several of us in this thread are of the opinion that it gets "smoother" after being turned on for several days.  In fact, the one marketing problem with the Multibit BIfrost is that it is so transparent, that it is not dramatic or exciting.  With the average poorly recorded album, it does not "sound great", it just gets out of the way of the music.  It takes a spectacular recording like the Reiner Scheherazade to make people notice the excellent sound quality.

 

[Baldr]

The AD8512 was warranted to keep the <1LSB DNL, INL, and noise specs for the AD5547 as specified by Analog devices.

 

[reddog]

Mr.45longcolt,

A couple more years and Intel will have M.Moffat's DAC design shrunk down to the size of a Snowflake for the Phone. 

You're a Pioneer and a lucky one at that, you didn't buy into the big $ MSB or that other one made in England or the Antelope for $13,000. 

This stuff is holding it's value, Ebay will get you 80% back, no problemo, a very small investment in your own education, you're more experienced than nearly everyone else, for a very small tuition.  

Besides, AtomicBob, up in Washington State owns all three of the Multi's.  You two may be the only ones to tell that story. 

You did well.

Tony in Michigan

Well said sir. You are full of fantastic data and information. I would love to have a drink with you and pick your brain.

 

[tonykaz]

Mr.reddog,
 
Thank you for the kindness
 
But it's Moffat & Stoddard who's brains you should concentrate on, they drink too.  I gave up drink 2 Decades ago, Courts, Company, Wife, family endorsed after I consumed 5,000 fifths of booze, ouch. 
 
Anyway, I'm only a dot connector, I'm typically wrong.  I need to work with a group of smart people to be effective, most of my work gets tossed early on.  Every once in a while I'll get a good Base Hit and an RBI.  
 
Get on a hot team and you're going places.
 
Schiit seems a division wining 'hot' team, nowadays.
 
I'd like to see em winning a Pennant, then I'll be pumped and excited to see em going for the World Series. How far off is that?  
 
I like having a 'Team' to root for, I'd love to have a Schiit Baseball Cap! ( I may even have one of my guys in our Print Shop make some up but I don't want to get into a dust up with those Schiit Guys over Intellectual property rights issues ).
 
Tony in Michigan

 

[reddog]

Mr.reddog,

Thank you for the kindness

But it's Moffat & Stoddard who's brains you should concentrate on, they drink too.  I gave up drink 2 Decades ago, Courts, Company, Wife, family endorsed after I consumed 5,000 fifths of booze, ouch. 

Anyway, I'm only a dot connector, I'm typically wrong.  I need to work with a group of smart people to be effective, most of my work gets tossed early on.  Every once in a while I'll get a good Base Hit and an RBI.  

Get on a hot team and you're going places.

Schiit seems a division wining 'hot' team, nowadays.

I'd like to see em winning a Pennant, then I'll be pumped and excited to see em going for the World Series. How far off is that?  
I agree 100 % the Audio Wizards Moffat & Stoddard are on top of their Schiit and have created a fantastic company with great products. I own a great many of there products, and I highly recommend there products to others.

I like having a 'Team' to root for, I'd love to have a Schiit Baseball Cap! ( I may even have one of my guys in our Print Shop make some up but I don't want to get into a dust up with those Schiit Guys over Intellectual property rights issues ).

Tony in Michigan

 

[RazrLeaf]

  Hard Design Decisions
 
From there, Mike and Dave seemed to drop the development of a Multibit Bifrost. It made sense—we were getting deep into Yggdrasil production and ramping up Gungnir Multibit, so they had plenty to do on those fronts without any other distractions.
 
I didn’t realize how close we were to having a Bifrost Multibit until Mike asked, “You know, if Bifrost Multibit had an op-amp analog stage, how upset would you be?”
 
“Why?” I asked. “Are we really that space-limited? I know a real discrete I/V is a nightmare, but if we do parts on both sides of the board—“
 
“You need two gain stages per channel,” Mike said. “It’s a multiplying DAC.”
 
“Oh,” I said, letting that sink in.
 
“It comes down to the burrito or a discrete I/V,” Mike said. “And even then, I don’t know—“

Noob question incoming: What's a I/V? What's a discrete I/V? I know what an op-amp is, but not I/V.

 

[Tuco1965]

https://www.by-rutgers.nl/IV-converter.html

 

[Ableza]

  Noob question incoming: What's a I/V? What's a discrete I/V? I know what an op-amp is, but not I/V.

Current (I) to voltage (V) converter.  Necessary for a DAC,

 

[Jason Stoddard]

  Noob question incoming: What's a I/V? What's a discrete I/V? I know what an op-amp is, but not I/V.

Yep, as Tuco and Ableza have said, it's a current-to-voltage converter. I could do a whole (boring) chapter on current-to-voltage converters, because they are one application that has very strange requirements when compared to a typical pre-amp or "ideal amplifier" situation.
 
An "ideal amplifier," for typical audio uses, has infinite input impedance and zero output impedance. Op-amps attempt to approximate an ideal amplifier stage. However, an ideal current-to-voltage converter has zero input impedance and zero output impedance. This zero input impedance can be approximated at the high-impedance inverting input of an op-amp via feedback, or a specific, discrete stage with low input impedance can be designed (a discrete I/V converter), which will have its input impedance lowered again via feedback to approximate the zero ohms a current-output DAC finds ideal.
 
Bottom line, you can do some spectacular things with discrete I/Vs, but they are complex, parts-intensive, and take lots of space. So that was never in the cards for Bifrost. 

 

[Tuco1965]

Multibit is the Schiit Audio Event Horizon
You're never going back.

 

[madwolfa]

Multibit is the Schiit Audio Event Horizon
You're never going back.

 
The Schiithole.

 

[rtaylor76]

 
An "ideal amplifier," for typical audio uses, has infinite input impedance and zero output impedance. Op-amps attempt to approximate an ideal amplifier stage. However, an ideal current-to-voltage converter has zero input impedance and zero output impedance. This zero input impedance can be approximated at the high-impedance inverting input of an op-amp via feedback, or a specific, discrete stage with low input impedance can be designed (a discrete I/V converter), which will have its input impedance lowered again via feedback to approximate the zero ohms a current-output DAC finds ideal.

 
I understand this is the English language and the individual words and even some individual phrases, but when you put it all together I understand nothing.
 
I feel like Jackie Chan in Rush Hour.

 

[kstuart]

Bottom line, you can do some spectacular things with discrete I/Vs, but they are complex, parts-intensive, and take lots of space. So that was never in the cards for Bifrost. 

One more thing (said Columbo ) - how is the final low pass filtering done on the Multibit Bifrost ?   Is there a built-in LP filter on the AD8512 that is used, or is there some other circuitry on the board that does that ?

 

[Jason Stoddard]

  One more thing (said Columbo ) - how is the final low pass filtering done on the Multibit Bifrost ?   Is there a built-in LP filter on the AD8512 that is used, or is there some other circuitry on the board that does that ?

There's no built-in low-pass on the op-amps, so we've applied an appropriate external filter.
 
(By the way, if some of you have seen some posts disappear recently, it's because, you know, we don't need to discuss every other random product on the planet, nor are comments on literally every tiny subject absolutely necessary. I appreciate all of your contributions, but let's see if we can keep at least within sight of the road.)

 

[jfoxvol]

 
Yep, as Tuco and Ableza have said, it's a current-to-voltage converter. I could do a whole (boring) chapter on current-to-voltage converters, because they are one application that has very strange requirements when compared to a typical pre-amp or "ideal amplifier" situation.
 
An "ideal amplifier," for typical audio uses, has infinite input impedance and zero output impedance. Op-amps attempt to approximate an ideal amplifier stage. However, an ideal current-to-voltage converter has zero input impedance and zero output impedance. This zero input impedance can be approximated at the high-impedance inverting input of an op-amp via feedback, or a specific, discrete stage with low input impedance can be designed (a discrete I/V converter), which will have its input impedance lowered again via feedback to approximate the zero ohms a current-output DAC finds ideal.
 
Bottom line, you can do some spectacular things with discrete I/Vs, but they are complex, parts-intensive, and take lots of space. So that was never in the cards for Bifrost. 

Like my electronics prof in undergrad said, "You don't get ideal op-amps until you get to heaven.  Until then, we have to use real ones."