Next best exponential DAC quality level?

I recently had the opportunity to try out some of the budget DAC's in my system to see how they compared to my AN DAC5 Sp, which I consider to be an outstanding dac. The Dac 5 has had quite a few mods ( 768K Femtosecond input receiver ) as well as the AN silver oil caps and the super regs. I bought a Gustard x26 pro for my office system and had access to a D90 Topping w a linear supply.

Granted this was a HUGE spread in price and I felt the result was interesting.

Not to brag but my system is super high end and I have been financially blessed but the result is interesting and worth consideration.

Bottom line is both of the Sabre dac's sound very, very much alike in comparison to the DAC 5 and unfortunately are nowhere near the realm of the AN dac. I somewhat regret spending the bit more for the Gustard but it is much easier to hook cables to and is balanced like my system. I fed the Gustard in NOS and internal OS, 24/384 from HQ Player and 16/44 straight. I have 14 tb of mucic, mostly SACD rips and the SACD was where the Sabres began to show promise but never gave the holographic air and soundstage that the DAC 5 did.

PRAT was in a quantum league higher with the DAC 5.

Images are confined to the space between the speakers with the Sabre Dacs vs an immersive wide and deep stage on the AN.

The Gustard was the strongest bass but not really that defined, however on a bright system it may be the ticket.

The interesting thing was the 2 low priced dacs were extremely listenable and any flaws were just shortcomings vs objectionable flaws.

I would not in any way call the Audio Note sound euphoric, syrupy or harmonically distorted. It on the contrary is very fast, open, clean and ever so slightly lean. There is a separation of instrumental textures and dynamic contrasts that make the Dac's I mention seem very sedate or compressed. There is much better HF extension with the AN even with the vivid setting on the Gustard. The Gustard bass is simply not as defined or could be considered woolly. Most people do not have single speakers that do true deep bass with the quality of my CS5i's or Divas and I would guess that the Gustard was voiced for bass shy smaller speakers. I have had many uber components in my systems over the years mainly from working in high end audio shops in NYC in the heyday of high end there and experienced what separates mid fi from high end in spades. The you are there sound is still a very expensive proposition in digital regardless of new tech. What is amazing is there is absolutely nothing record breaking about the AD1865 chip at 18bit and a 768k sample limit but the implementation makes all the difference.

The bottom line in my posting was to elucidate that there is a huge economic chasm to get the "quantum leap" that the OP was referencing and in my experience it comes down to costly implementations.

Al lot of assumptions regarding image placement, soundstage and instrument separation being from extra harmonic distortion.

I want to add that I am a retired prototype engineer with 14 years in consumer and medical electronics. I have made many many amps, preamps, several turntables, 3 tonearms and a slew of speakers back when I worked in that industry so my experience level is quite extensive.

Added harmonic distortion adds a haze that thickens the soundstage which is the opposite of clearer separation. IMD can add edges that seem to define edges but it is almost always localized to coming directly from the speaker as it is not consonant with the original impulse.

When I added the Jung Super reg to the dac is when the deep soundstage really came to life as well as much clearer soundstage placement. It is extremely improbable that the Jung 5v regulator added any level of distortion as some suppose.

The ASR crowd seems to be fixated on the singular measurement that they like to rank components on and while instructive it is not the end all be all ranking of a truly great audio component.

If you look at the Linear Audio article on regulator tests you will find that the superlatives I mention are closely associated with superior power supply implementation. Great image depth requires a very fast settling and low noise supply with a low output Z. The ASR ranking has absolutely zero methodology to quantify those aspects. It’s a shame that some people are so closed minded but I suggest people have a look at the article as it is really enlightening and done by the best minds in the field. Linear Audio article

I believe you meant bandwidth determines settling time not frequency response, but it is not 100%.

Phase angle, the load reactance and feedback also determine settling time.

I was referring to the engineers who designed the regulators and Jung, Swenson and the other companies designs represent the best minds out there.

The regulators absolutely need to have the bandwidth in digital circuits and to conflate human hearing with digital circuits is sort of disingenuous.

Possibly you have had seen base model AN units that only have RC or CLC power supplies and SRPP output stages but the upper tier models are CLC filtering into feedback regulated tube pass elements units with plate transformer opt’s.

Yes with the 44:1 opt the output z is only 30 ohms but that is more than adequate for 600 ohm balanced lines.

Yes I agree that full ground plane implementations are mandatory in high speed digital circuits but that is basic design implementations on nearly every digital circuit.

I really do not buy the thesis that an open soundstage with depth and delicate delineation is a result of some harmonic distortion from the use of tubes. I had a Goldmund 12 dac that has it in spades. I sold that to a local audiophile and hear those qualities in his system every time I go over for a listen. The Goldmund is SS gear of very wide bandwidth. No phase issues to be the cause there or tube oriented THD.  I also have a Soekris dac that is somewhere between the Gustard and the AN. There is more to it than just dismissing things as tube euphoria.  Being the DAC’s I recognize these characteristics are R2R It’s possible that R2R DAC’s have these qualities but I believe it’s implementation.
 

It doesn’t make sense to get into the effectiveness of decoupling vs dedicated regulators or sophisticated implementations as there is no control scenario here and I am a firm believer that we don’t know all of the factors that effect SQ nor do we have established tests for complex audio waveforms that have correlation to what we hear yet.