SMSL VMV D3 Review (R2R DAC)

[jcao309]

How does it compare to D1?

 

[garbz]

Just because something better/ Cheaper exists does not make this thing worse.

Yes it does. A value proposition literally depends on what alternatives are available and at what price. When spending money (shopping) we don't talk in terms of absolute performance we literally talk in terms of value proposition, whatever that "value" is, be it performance, looks, or if it comes with a free coffee.

People still buy Mechanical watches.

Indeed. And you should always ask those people *why* they buy them. Hint: Not for performance. A $10 Casio is orders of magnitude better at keeping time than my Breitling. It's 2022. Accurate time is freely available by looking at your phone or asking someone, far better than even the finest mechanical watch can offer. The value proposition of a watch is entirely in it jewellery value, its looks. And that value proposition is still related to what else is on the market. If a better looking watch for the same price was available I'd have bought that instead and consider the one I own "worse".
Same about your car comparison. No one I know owns their car because they are trying to perfect the past. Everyone with a classic car acknowledges it's worse than a modern car by every measurable metric except for the single intangible one: emotional attachment.

If you want to buy audio gear just so it can sit pretty in your living room then more power to you. But you won't find many friends here.

So vinyl records are not Hi-fi or no longer hi-fi?

Yes. Haven't been by any reasonable definition for a long time.

 

[Jimbob54]

How does it compare to D1?

The D1 was an expensive DAC but (for the time 4 years ago) was close to the top of DACs Amir had measured for SINAD for $1300

This is an equally expensive DAC , arguably more expensive in relative terms than the D1 as greater value and performance is offered for far far less these days. The D3 doesnt even perform well against best in class R2R DACs- never mind a half competent DS DAC.

But then performance isnt what this DAC is about...

Review and Measurements of SMSL VMV D1 DAC

This is a review and detailed measurements of SMSL VMV D1 DAC. It is on a kind loan from a forum member. It retails for USD $1,299 including free shipping on Amazon. It is a super hefty unit, far more than you can imagine from its small enclosure: The box on the left which is permanently...

www.audiosciencereview.com

 

[Lambda]

A $10 Casio is orders of magnitude better at keeping time than my Breitling. It's 2022. Accurate time is freely available by looking at your phone or asking someone

And the same way your Phone can give you Accurate Audio reproduction...
If not a 10€ dongle can.

Yes it does. A value proposition literally depends on what alternatives are available and at what price. When spending money (shopping) we don't talk in terms of absolute performance we literally talk in terms of value proposition, whatever that "value" is, be it performance, looks, or if it comes with a free coffee.

Speak for yourself.
I never claimed i made my rating for this in terms of value proposition.
And i would say it is hard to make a scientific objective value rating.

I see the ASR rating as an absolute performance rating.

If something is good it is good and it gets rated this way no matter what it costs.

So why focus on the price if it is not so good.
but still within >99% will never hear difference listening to a normal CD.

If you want to buy audio gear just so it can sit pretty in your living room then more power to you. But you won't find many friends here.
Yes. Haven't been by any reasonable definition for a long time.

Then why are there Reviews of Phone Preamps here that alone cost more than Many DACs performing an order of magnitude better...

The value proposition of a watch is entirely in it jewellery value, its looks.

Why can't this be true for audio gear?

Not saying its for me, im the guy with the 10$ Casio... but for some sure is.

 

[AnalogSteph]

They are using a chip that is rated only slightly better than -96dB THD (which is fine for 16-bit audio!) and not a design of their own like Denafrips.

I'm also surprised at the decent filter performance. Chip used: SM5847af.

Do note that filter ultimate rejection is short of datasheet spec (-117 / -116 dB)... possibly the filter was overloaded, which would not bode well for 0dBFS+ handling.

Converter linearity is about what you expect for a meh grade PCM1704, I guess... this chip has the disadvantage of no external adjustments and has always been a bit underwhelming in the linearity department. A well-tweaked PCM1702 can pretty much get to 20 bits, but I guess that part has been out of production for even longer.

When the PCM1704 was released in 1998, the writing for high-end multibit DACs was definitely on the wall... just 4 months later, it would be surpassed by the AK4393 in all major specs, and buried by an avalanche of new high-performance delta-sigma parts the next year. By late 2000, it was hopelessly obsolete.

 

[Vhond]

How does it compare to D1?

And what about the D2??

 

[Jimbob54]

And what about the D2?

Given it uses highly rated modern dac chips (akm 4499) and smsl can build good dacs, I would imagine it measured pretty well. Probsbly as well as some of there $200 + units. The question would still be why anyone would want a $1700 dac when the same manuf sells essentially the same unit in a different case for far less.

 

[Vhond]

I am still waiting for measurements for this D2, not for answers about the price of this dac.

 

[Ilkless]

I think this is where the confusion/tension arises. I'm not so much about recommending "cheaper options" solely on SINAD performance but rather for the premium price range of $3000 - $4000, there are better designed options where the SINAD is less relevant and usability + design is many steps above the SMSL. When they tread on the premium segment, it's no longer about performance inside a solid billet aluminum case, but all the intangibles from the click of the volume knob to the feel of the remote to ease of operation to visibility of the display. I would look at the Auralic Altair G1 for a few hundred less before I'd consider the VMV D3. If SMSL wants to play with the big boy DACs in terms of usability, design and the "feel" of high end, it needs to look at Lumin and Aurender as well. Upstarts like HiFi Rose RS250 for $2500 is also gaining momentum and looks much more premium:

View attachment 183363

The days of wrapping your DAC in R2R magic for $3,500 is simply not enough when you are attracting the discriminating clientele. It looked like a $1000 DAC to me, that's why I expected it to compensate elsewhere (like amazing SINAD) but with nothing special in measurements, I'm just left wondering about its intangible value proposition - is it purely the placebo of R2R? So no it's not terrible if you refer to only its sound quality, but it's pretty terrible everywhere else.

If the mystique around the circuit topology to matter ... who'd buy it over a Schiit Yggdrasil or a Chord Hugo 2 or a TT2? And within just Chinese brands who'd buy it over Denafrips or Audio-gd? Scratching my head.

 

[amirm]

If the mystique around the circuit topology to matter ... who'd buy it over a Schiit Yggdrasil or a Chord Hugo 2 or a TT2? And within just Chinese brands who'd buy it over Denafrips or Audio-gd? Scratching my head.

The local market in Asia may buy this over those brands which they may not able to get anyway.

 

[Snoopy]

If the mystique around the circuit topology to matter ... who'd buy it over a Schiit Yggdrasil or a Chord Hugo 2 or a TT2? And within just Chinese brands who'd buy it over Denafrips or Audio-gd? Scratching my head.

With import fees and shipping the YGgdrasil would be only a 300 USD cheaper than the SMSL V3. But the SMSL would have warranty by a local retailer.

The Chord stuff is just butt ugly.

There will be always customers for something like this. But you would have to be nuts to spend over 3000 USD if you could get a musician Aquarius for the same amount.

 

[Ilkless]

With import fees and shipping the YGgdrasil would be only a 300 USD cheaper than the SMSL V3. But the SMSL would have warranty by a local retailer.

The Chord stuff is just butt ugly.

There will be always customers for something like this. But you would have to be nuts to spend over 3000 USD if you could get a musician Aquarius for the same amount.

Or Denafrips, Audio-GD or Holo Audio, who all made their names on R2R mysticism - that's just the domestic brands. People buy into myths with these DACs. And as far for how much myth you "get for your money" (e.g. Chord with its wacky FPGA and WTA stuff, or Yggdrasil with the closed form filter, and US/UK construction for both) - the SMSL is near the bottom because mystic overcomplicated underperforming DACs was never what they made their names on.

 

[Spocko]

The Chord stuff is just butt ugly.

Ok, I thought it was just me.

 

[Lupin]

But you would have to be nuts to spend over 3000 USD if you could get a musician Aquarius for the same amount.

But you have to be nuts to spend that much on a musician aquarius imho

 

[Rotary Transformer]

I thought SMSL would discontinue the VMV.

 

[Esotechnik]

Or Denafrips, Audio-GD or Holo Audio, who all made their names on R2R mysticism

You have not measured or compared white noise autocorrelation time and settling time for different types of DACs.
PCM1704 has settling time 200 ns.

 

[Ilkless]

You have not measured or compared white noise autocorrelation time and settling time for different types of DACs.
PCM1704 has settling time 200 ns.

Seems rather disingenuous to buy based on measured performance (even if unconventional measures that are irrelevant for competently-engineered DACs) in this segment. People buy based on their intuition and feelings, they see which DAC has the better backstory.

 

[Esotechnik]

(even if unconventional measures that are irrelevant for competently-engineered DACs) in this segment.

You do not know how to measure the time errors of sigma-delta DACs, because you do not have accurate in the time domain ADCs (SAR).

Old 18-bit converters better than all modern "24-bit" with real music:

Df-measurements of DA/AD loops from Gearslutz forum - Articles - SoundExpert

Artifact signatures of DA/AD converters in similarity space.. SoundExpert news and articles

soundexpert.org

"Most tested devices have df-levels around -40 dB and are grouped in two clusters according to similarity of their artifact signatures. Usually devices within a cluster use similar engineering/circuitry solutions.
There is a small group of devices with higher precision around -69 dB: Texas Instrument PCM4222 Evaluation Module, Focusrite Blue 245, Forssell MDAC-4, MSB Platinum IV Plus DAC, MSB Platinum Studio ADC, SPL Madison, ReQuest Audio The Beast DAC, Swissonic DA24 MkII"

It is subtle on the sine signal, but you can hear it on the music.

 

[Jimbob54]

You do not know how to measure the time errors of sigma-delta DACs, because you do not have accurate in the time domain ADCs (SAR).
View attachment 184292
Old 18-bit converters better than all modern "24-bit" with real music:

Df-measurements of DA/AD loops from Gearslutz forum - Articles - SoundExpert

Artifact signatures of DA/AD converters in similarity space.. SoundExpert news and articles

soundexpert.org

"Most tested devices have df-levels around -40 dB and are grouped in two clusters according to similarity of their artifact signatures. Usually devices within a cluster use similar engineering/circuitry solutions.
There is a small group of devices with higher precision around -69 dB: Texas Instrument PCM4222 Evaluation Module, Focusrite Blue 245, Forssell MDAC-4, MSB Platinum IV Plus DAC, MSB Platinum Studio ADC, SPL Madison, ReQuest Audio The Beast DAC, Swissonic DA24 MkII"

It is subtle on the sine signal, but you can hear it on the music.

I'm not competent to understand this, but I suspect this will generate an interesting but hopefully polite debate.

One question in my ignorance. If one cant measure DS accurately enough I the time domain, how did they measure this R2R at 200 ns?

 

[Esotechnik]

If one cant measure DS accurately enough I the time domain, how did they measure this R2R at 200 ns?

With an analog oscilloscope, and later it became possible through a SAR ADC. They do not spoil the statistical characteristics of the signal like sigma-delta converters.
Two 16-bit ADCs from Analog Devices (instrumental vs. entertainment "for audio") give these pulse characteristics:

One of the best (in terms of noise) modern ADC "for audio" ES9822 measures 1 = 0.5...0.8

Audio Precision 555 has 2 channel: low (Fs≤216 kHz) and high bandwidth (≤2.5 MHz, SAR).
But Amir never tried to digitize "natural" white noise, e.g. amplified signal from an open stabilizer (zener diode). There should be a big difference even with the same bit depth and frequency.