Comparison of latest DAC chips


I own a Bluesound Node 2i which greatly improved sound after I added the Cat 6 cable.  I do not own a separate DAC but am told it would be the next step improvement.  I have done enough reading that it appears the two latest chips are the Sabre or ESS ES9038PRO and the AK4499.  The brands I have looked at are Sabaj d5($469) and a Topping D90($699).  I saw a great review on Audiocircle of the Sabaj D5 which is now a year old.  The Topping D90 is newer and I hear the build quality of the Topping as well as customer service are both better.   Other brands cost more and most don't use these new chips. 

Is there a difference in how these chips sound?  I would appreciate any comments. 
128x128daledeee1

Showing 11 responses by rossb

Not even analytical, a good "discrete" R2R dac is even more detailed, yet it’s airier with much more musical sense and far more body to the mids, and then there’s the unmatched bounce and slam to the music they have as well.
Sorry, but this is complete rubbish.

Some of the best dacs out there use ESS chips - Ayre, Brinkmann, Lumin, Mytek.

I've owned a "good" R2R dac - a Metrum Pavane - and the Ayre QX5, Mytek Manhattan II and Lumin X1 with their ESS chips walk all over it. And, yes, I've owned all of them.

The idea that R2R dacs have "unmatched bounce" or are more detailed is just ridiculous. The current generation of ESS chips are outstanding which is why they are used in many very high end dacs.
The above R2R detractors, obviously have never A/B'ed their ESS delta sigma dacs (listening to PCM) compared to a good discrete R2R dac
That is completely incorrect. As I said, I have owned a very good R2R dac and a number of ESS dacs and compared them. I only listen to PCM. The R2R dac was okay on its own terms but I much preferred dacs using the current generation of ESS chips.

You are over generalising about dac chips - suggesting that the chip is the only thing which defines the sound of the dac, when many other elements of the design also come into play - the filtering, power supply and analog output stage for a start. There are good R2R dacs and good ESS dacs. The chip is only one factor of many in the final sound of the dac.
You are also making ridiculous generalisations about "R2R detractors". How can you possibly know whether "R2R detractors" have compared R2R and ESS dacs. As, I said, I have done such a comparison and I am sure that many others have as well.

And no one is suggesting that there are no good R2R dacs. Just rejecting your preposterous statement that all R2R dacs are inherently better than all ESS dacs merely because of the dac chip architecture, which is patently absurd.
Didn’t say "all" sunshine, I said the better hi-end ones, now your generalizing, with that "preposterous statement".
So the better ones are ... better. Can't argue with that logic.
If you’re going to refer to correct English, at least spell it correctly. "Queen’s English", not "Queens English".

And, as I have said - twice now - I have had a good R2R dac and prefer my Lumin X1 with its ESS chip.
With the X1, where do you 'fall' with regards: what is (may be) responsible for your preference of it, since it is more than just a DAC?
I dont think any one element can be identified as the reason for my preference for the X1. It is just a very well designed and implemented unit. I know that the Lumin streamers are very good (I also have a U1 mini), the ESS 9038 dac chip is capable of sounding great (I was less impressed with the previous generation of ESS dacs), the X1 has a very high quality linear power supply, and the output stage uses the excellent Lundahl transformers. It is finished to a very high standard. All of these factors combine to produce great sound.

The sound the X1 produces is dense and colourful, perhaps slightly warm, as well as being fast and dynamic. This contrasts with the slightly thin and cool sound of the Chord DAVE/Blu 2 which I owned for a while, and which is not a sound I typically enjoy. The DAVE did have more apparent detail, but it was an illusion caused by the leanness of the sound and overemphasis of fine detail. The X1 has all the detail, but it has a fuller, more natural sound so that detail is not artificially emphasised.
The Metrum Pavane I owned at one point, which is an R2R dac, was also very good, and had a similar sense of tonal density and richness. But it sounded a little slow compared to the ESS based dacs (and the DAVE) and was, to my ears, lacking slightly in dynamics. The X1 seems to combine tonal richness and dynamics in a way that is close to ideal. I'm sure that there are better sounding dacs out there, but you would probably need to spend crazy money to improve on it and achieve that last 1% - maybe.
George, all of your quotes are about recording in DSD. They say literally nothing about DACs. They certainly don't support your conjecture that delta sigma chips "do a poor job of decoding PCM". You will need to come up with something else.
With bandwidth and sampling limitations, ringing and aliasing artifacts, all digital is only a facsimile of the original. R2R is no better than delta sigma in this respect, and has technical limitations of its own. 
The "recording people" are not saying what you think they are saying. 

I think you have read that Mojo Audio web page without fully understanding it. 
Your quotes are all about DSD vs PCM recordings. What has any of that got to do with delta-sigma or R2R?
George, you keep saying the same thing, and getting the same response. I'm sure most of us have done this. I used to own a "good" R2R dac. I was not amazed. I prefer my current delta sigma dac.