D to A converter purchase....


I'm debated on either a Wadia 321 or the Schiit Gungnir Multibit.    Does anyone have experience with these two products?    Thoughts?    They are both close in price.   
whiskeypirate
Yes, I realize the gain in signal-to-noise ratio when you combine the converter chips that way, but it's still an 18-bit part and will only decode 18 bits worth of data.
Yes, I realize the gain in signal-to-noise ratio when you combine the converter chips that way, but it's still an 18-bit part and will only decode 18 bits worth of data.

This thread discussion is beginning to read a lot like this one :)

https://forum.audiogon.com/discussions/schiit-yggdrasil-21-bit/post?highlight=2097152&postid=148...


"Read this from MoJo Music:
'When a Redbook PCM file is played on a native DSD delta sigma single-bit converter, the single-bit DAC chip has to convert the PCM to DSD in real-time. This is one of the major reasons people claim DSD sounds better than PCM, when in fact, it is just that the chip in most modern single-bit delta sigma DACs do a poor job of decoding PCM.'"

I have seen this quote touted before and I really have no dog in this fight, but Mojo is hardly an objective observer as it sells a $7600 DAC based upon AD1862 R-2R ladder DAC chips.  I'm sure it sounds fine, but it is not the only way to build a fine DAC.  There are also several outstanding DACS built around the newest delta sigma chips. 

It has become commonplace to bad mouth DS chips these days, possibly because older generations of them were not top performers.  Also, they are easier to implement and therefore  ubiquitous.  So they are built into DACs with a greater variety of SQ.
"I would buy a piece of Schiit ..." Sorry, but I couldn't help but laugh. :-)
This thread discussion is beginning to read a lot like this one :)

https://forum.audiogon.com/discussions/schiit-yggdrasil-21-bit/post?highlight=2097152&postid=148...

Yeah... ha!

However there is an important point to reiterate here. Using two DAC chips in a hardware balanced configuration doesn't increase the number of possible output levels (unlike what you implied above and as mentioned in the other thread) but it does double the output signal level relative to the noise level. When you then use that fact in the formula for signal-to-noise ratio, 20*log(S/N), you get a 6 dB increase, which is equivalent to one bit.

All this isn't to say that the Gungnir isn't a good DAC. I just think it's rather silly to claim you're "preserving the original samples" when it's clearly not the case for 24-bit source material.