First post here since the surgery! Still wearing the medieval torture device on my neck. Can't see my feet, dick, or keyboard. Here goes:
So here we go with an Yggy design discussion. Strengths and limitations. Limitations?? Yup. Relax! It's upgradable, remember? I'll be going to go into a section by section tear-down of details, before I close with an op-ed on the futility, hubris, and presumption of attempting to sonically tune a D/A converter component.
So how about we begin with the power supply?? Customary in the industry is series regulators for all of the DC supplies. We do that in all of the DACs below the Yggy. Sorta very standard. Now, just to separate us from the others, we tend to break up the power supplies in more regulators than others, and even more regulators as we go up the line. For the Yggy, we use shunt regulators, which lie in parallel with the load. Far less efficient. Far better performing until you add current sources to drive them. Wow! Even less efficient and even lower noise and better regulation. This is the reason that Yggy provides the detail it does over Gumby, which has a more standard power supply. It is also he reason Yggy draws just about triple the power from the wall socket.
What could be better? Batteries inherently charged by the regulators. Not the lithium kind. Not just because such batteries randomly launch flaming balls across your listening area which are capable of causing third-degree burns in vapers and bring down 737 airplanes for large quantities of them self-immolating in unremarkable flight. Nah – these are for lightweight dilettante listeners willing to overlook real batteries with real low impedance – lead acid. Now there ya go!! Imagine the bragging rights you can have explaining to all of your other fellow audiophile friends that your power supply weighs over 100 lbs!! Imagine the thrill of your audio room with acid burn spots on the floor as well as the scent of an auto repair shop!! No? Then I think we will stick to the power supply we have as being best we can do within the constraint of practicality.
So how about the USB section? Well, for now you got me. Despite the fact that I have almost certainly designed more D/A converters than any other living person (dozens and dozens), I have only designed 8 USB sections. Three are in and one has been obsoleted current Schiit products, 3 suffer from various levels of excess suckage, and one was XMOS based, therefore drawing more current than your average 747. Rest assured I am constantly tinkering with better sounding USBs, and when available, (not anytime soon) will offer them at a price where current users will NOT get hosed.
Now we get to the digital section. This is the section that if you hear what it does (some cannot), you will NOT be able to replace it in ANY other DAC. The megaburrito filter is unique in the D to A world. It is designed for the reproduction of redbook material above all else. It has a higher bandwidth than any other in the frequency domain, and an ability to optimize time domain cues so that the reproduced music stages in an uncanny way, particularly wih speakers excelling in the time domain. Head-fi threads are filled with anecdotes of startling staging being heard, even around corners, fooling the users or their friends into thinking real people are there. This is what some cannot hear. Why? Dunno. Some people are tone deaf, and some are time domain deaf, or need to upgrade amps/headphones/speakers. I have looked at every other filter and am confident that none can offer the time domain performance that ours does. If you listen exclusively to techno/electronic or other computer generated music, this will be of little use to you. If you use your system to reproduce musical events, it will be vital.
How about the DACs. Multibit rules! We use the AD5791 which some Luddite on another forum called me out on because it is an industrial DAC, and therefore unsuited to audio. Well, industrial DACs are far more accurate than audio DACs, which are designed to be cheap. Audio DACs will not work for medical imaging (random gross errors) or weapons guidance (dead children rather than soldiers). This is the same as why industrial tubes sound better than 12A_7 audio cheapos. What about the guys who make their own resistor ladder DACs, some with Vishay resistors. Wow! This requires that the DACs be huge. Huge = slow (tubey) – therefore quite euphonic. The problem is that, say, in a 20 bit system, the lowest bit is a millionth of the biggest one. It is a huge engineering nightmare to work on a ratio level for a sample size of one, much less in production. My money is on the fact that no two sound the same. Maybe that's not a goal. Besides unless the DNL/INL specs are in (I have no idea how they can do that in a discrete resistor cluster-they certainly do not spec it), you still have the thyroid coming out when the patient has colon cancer as well as the dead kids with these in the air DACs.
Analog!! Best for last! Just a few analog comments. I never, but never try to voice my system with a D/A converter. I try to make it as neutral as possible. If you want to voice your system, that's what amps, headphones, and speakers are for. If you want to go crazy, get somebody's equalizer. Here's my rules:
1. Solid state gain = more overly detailed (perjorative - bright)
2. Solid state buffer = much less overly detailed (perjorative – much less bright)
3. Tube gain = more tubey euphonic (perjorative – dull)
4. Tube buffer = less tubey euphonic (perjorative – less dull)
And the winner is ---- number 2. Closest to zero. Voice it with the amp/cans/speakers you like.
Remember, most of the real $$$ high end DACs are built in very unlimited quantities for a very unlimited price. You can bet the maker is dead serious about every unit he makes and voices it to his system and taste. I am sure that is why there is such a wide variety of opinion.
I'm not sayin' the Yggy is everything to everyone. It is not. Voiced with the system you like, it is a ton of fun which gives you an amazing amount back for what you put in.