Thanks for going into such detail on this. It really does sound like a good DAC, and I’m wondering if I should join it too. I have the MCTH, so this should pair nicely!
Not one based on actual listening, no.
I found the D50’s presentation to be sterile, clinical and analytical. Technically competent, but not something I wanted to spend more time listening to (and certainly not the amount of time it takes to review things properly).
There seems to be some possible intellectual property issues…
Here are some initial measurements.
I was interested, but my enthusiasm is a bit dampened after reading these posts.
I’m not going to be an early adopter for this one.
I’m mostly concerned about consistency and quality control with a technology that’s got a checkered history.
At times a clean/analytical DAC helps to work around other components. Stacking a colored DAC on my tube amp may be too much color for me. Some of my headphones are already warm/dark too – so I currently pair them with a clean DAC.
I’d love to demo this before buying.
I’m sort of on the same boat there. I have the MCTH, and even though this stack looks amazing, I’m already quite happy with my D50, which is known to be quite transparent.
Inputs and Supported Formats/Resolutions
The RDAC offers three inputs - S/PDIF via COAX (RCA) and TOSlink, and USB via a micro USB socket. These are selectable, sequentially, via the “Input” button on the front of the unit.
The USB input IS galvanically isolated , and the USB input does require USB host power to function as a result. As is typical for isolated USB interfaces, USB power drives the source connection side, and power from the DAC itself drives the DAC side of the USB interface. This also means the unit will show up on your computer even if it has not been powered on.
This isolation seems to be pretty effective as using the COAX or TOSlink inputs did not yield a useful, audible, difference in quality in my setup. Similarly, trying a variety of USB-to-USB “purifiers” and/or DDCs in the chain did not yield a conclusive improvement. So, while you may experience different results, depending on your precise chain, the USB input is the way to go with this unit in my opinion. Though the option for TOSlink is particularly nice if you want to pair this with a TV/AVR/game console, or are having issues with ground loops.
In general, though, to get the most out of the RDAC you’ll want to use the USB input. This will give you the broadest range of bit-rate and format support. Officially, USB will let you play PCM up to 24/384 kHz and double-rate DSD (DSD128). With S/PDIF and a native DSD stream you can get to DSD128 and 24/192 kHz for PCM. S/PDIF using DoP will limit you to DSD64 (single-rate).
Interestingly, and I’ll stress that this is NOT something that is officially supported, nor guaranteed to work , I’ve been able to run the unit at 24/768 kHz and DSD256 (quad-rate) without any issues at all via USB.
As mentioned, the Massdrop x Airist Audio R-2R DAC is natively a multi-bit PCM converter. It can accept native DSD input as well as via DoP, but that DSD content will be converted to PCM before being fed to the actual DAC ladders.
Absent completely discrete PCM and DSD conversion stages, all DACs must do some kind of conversion from one of these formats to the other. For example, the PS Audio DirectStream models convert everything internally to DSD before doing anything else, whereas units like the RDAC or the Soekris units convert DSD to PCM.
What matters most, of course, is what the end result sounds like. Interestingly all of the positive traits I found in the unit are preserved with DSD source material – the same sense of clarity and “pristine” nature of the sound is fully in evidence, and the tone retains the hint of sweetness that I keep going on about.
In practice, this turns out to mean that the relative positioning of this unit vs. the others I compared it to, remains almost identical regardless of whether you’re feeding the RDAC PCM or DSD. I will say that the low-end of the iFi Micro iDSD Black Label cleans up a little when fed with native DSD content, but at best that brings it on par on that one aspect and overall, I still prefer the presentation from the Massdrop unit.
No Snap, Crackle nor Pop!
It is very common for DACs, even expensive ones, to exhibit nasty pops, clicks, and other noises when switching between DSD sample rates and between DSD and PCM content. So, it’s nice to find that the RDAC makes no utterances during such switches. The unit mutes momentarily during these changes. In general, this is transparent unless you’re skipping tracks. There was no intrusion into the replay of a normal playlist of mixed rate DSD and PCM files.
Tweaking DSD/PCM Conversion
You can pre-empt the internal conversion by using something like HQPlayer or Audirvana+ to convert your DSD files to PCM ahead of feeding them to the DAC, using any of a number of more sophisticated routines if you wish. The differences are often audible, if quite subtle, though as an on-going theme with approach I didn’t find any of the them to necessarily be preferable to just letting the DAC do its own thing.
Don’t be daft and convert to DSD from PCM ahead of sending data to the RDAC, however!
Power & Power Supplies
Power to the Massdrop x Airist Audio R-2R DAC is supplied via an external switching PSU rated for 3W @ 5v DC and uses a standard barrel connector. The external nature of the supply means that you can, if you are so inclined, substitute a suitably rated linear power supply instead. Note that all of the listening performed for the main part of this review was performed using the stock PSU.
Out of curiosity, once the main listening was done, I did try a couple of different LPS units here:
The Teddy Pardo “Teddy5/3” yielded a small reduction in the noise floor, a side effect of which meant very low-level detail was more easily discerned vs. the stock PSU. Though it’s worth bearing in mind that this particular PSU costs more than the RDAC itself, and the very minor change in performance probably isn’t worth it, nor is it likely to be routinely audible.
The iFi Power, which is technically underrated for this application (2.5A instead of 3A), did not seem to offer any audible improvement at all. Though this is also an SMPS, just with more filtering and “active noise cancellation”.
And my lab-grade supply - which gave a larger improvement than the Teddy Pardo unit, but at an even more ludicrous value proposition (several times the price of the RDAC).
My recommendation is, worry about your headphone and amplifier first and when you eventually have nothing else to adjust, maybe add an LPS to the RDAC!
The Massdrop x Airist Audio R-2R DAC is a 24-bit discrete R-2R (multi-bit, resistor/string ladder) sign-magnitude, oversampling, converter, using low phase-noise NDK clocks and a custom linear phase filter.
That’s an impressive configuration for a DAC at this level, particularly the implementation of sign-magnitude conversion. This dramatically improves linearity vs. a simple ladder (as much as two orders of magnitude at -60 dB). In listening to other DACs using simple ladders vs. a sign-magnitude implementation it has been my consistent finding that a sign-magnitude approach yields clearly better results - even when compared to much more expensive units.
Sign-magnitude converters use two separate ladders per channel to improve conversion linearity. With a single ladder, any non-linearity in the ladder winds up being relative to the full-scale signal. A simple R-2R ladder specified with 0.001% distortion actually has increasing distortion in inverse proportion to the signal level, reaching 1% by -60 dB. With a sign-magnitude implementation, as used in the RDAC, the specified distortion remains constant … so 0.01% at 0 dB stays 0.01% at -60 dB.
Critical listening was done with the RDAC being driven via it’s USB input (after determining this was the best way to drive it in my system) via Audirvana+ and Roon. Source material was primarily 16/44.1 FLAC format CD rips, with some native high-resolution content and multi-rate DSD albums to test those capabilities.
Primary listening was performed via an iFi Pro iCAN and a Woo Audio WA234 MONO Mk2, with some additional amplifiers included for pairing/system matching commentary. Headphones ranged from the HD650 to the Focal Utopia and Abyss AB-1266 Phi.
Direct comparisons were performed in a hardware assisted blind(ish) fashion. Not strictly a fully blind comparison, but a lot closer than simple sighted listening.
I’m jumping on it, I have the El Dac which is nice, but in this hobby DACs are definitely where I lack knowledge and I will defer to @Torq and @I_want_all_the_tacos pretty heavily at this stage based on previous encounters with Taco, and from reading the posts from Torq on this forum. I have similar tastes to Taco, and I want all of Torqs gear lol (wife would never let that happen )
That’s fair, I think. For now, I am at a stage where I have a DAC which I really like (D50, though it’s the only great DAC I’ve owned), and an amp (MCTH) and headphones (HD600) that I love, with more on the way (HD6XX and MD+).
I’ll probably save up now for a significant jump in the Transducer quality (HD800, ZMF Auteurs, etc) before looking to upgrade the source chain any further. This one sounds like it’s gonna be a sidegrade in fidelity / clarity, with some coloration (which I may or may not enjoy).
If I do go for a DAC after this, I’ll look to make a jump down the line to an ADI-2 or Soekris 1421, or better.
I started with a Topping TP30 mark2, Beyerdynamics DT 990, Fiio A5, and HD6XX. I rapidly ascended from there. JDS Labs The Element, then picked up the El DAC to pair, M1060, Fostex X00 purplehearts, HD700, HD800, M1060C, Focal Elex, Campfire Audio Comet…I’m waiting on Massdrop Jubilee, the Mobius, and Campfire Audio Cascade, MCTH will arrive in the next couple days, and so will my Bottlehead Crack with speedball kit, and I’ll be picking up this DAC tomorrow! looking back this all started just under two years ago…yeesh my poor wallet But that is what debt free and expendable cash can do for you… plus I tend to obsess over a thing until I just get it. Before all that I did similar things with gaming headphones/sets, which eventually led me to this path of audiophile bliss!
Amp Pairing - Complements & Synergies
One of the things that I’ve seen mentioned quite a bit in various write-ups of some of the existing “Massdrop x” series products, is the desire for a form-factor matching source to stack with them. Given how impressed I was with the Airist Audio R-2R DAC, I decided to buy a couple of these units and see how they performed when paired with the RDAC.
In this case, I purchased the Massdrop x Cavalli Tube Hybrid directly via the “Massdrop Made” ships-from-stock program. The Cavalli Liquid Carbon X is not currently available via this program, so I wound up buying one used from an existing owner.
Massdrop x Cavalli Tube Hybrid (MCTH)
It’s fair to say that this little amp quickly endeared itself to me. I’m definitely a tube-guy when it comes to headphone listening. I tried the MCTH with a variety of sources and headphones to get an idea of its overall performance. But, since this isn’t an amp-review, I’ll limit my comments to how it pairs with the Massdrop x Airist Audio R-2R DAC.
In short, this was the most enjoyable pairing* I found with the RDAC absent moving up to something like the iFi Pro iCAN. The slight warmth and bottom-end weight that I got from the MCTH (using the stock tube) complements the super-pristine signature of the RDAC almost perfectly, and the end result is a very balanced and enjoyable listen.
The deftness of the RDAC’s bass delivery isn’t lost here, yet the weight and mass of a performance definitely benefits. I do not consider the RDAC to be light in the bass; it’s pretty neutral, but perhaps some attention/focus is drawn elsewhere by the sweet mid-range and treble. Pairing with the MCTH adds just enough low-end emphasis to balance that out, without robbing the overall sound of that sweetness.
The sense of air and space I got from the RDAC is further enhanced when paired with the MCTH. Ambient cues in live venues/recordings seem more apparent, string pluck and tone becomes even more addictive, and the end result is extremely compelling.
With some tubes, it’s possible that the innate sweetness of the RDAC might prove to be a bit on the “sugary” side of things. That’s obviously down to personal preference. I found that, at least with the stock tube, that this particular pairing was greater than the sum of its parts.
Perhaps it’s a “natural synergy” (I find similar kinds of symbiotic effects pairing Yggdrasil with Ragnarok, for example), or maybe it’s the tube-effect, but either way I’m hard pressed to come up with a better, more enjoyable, source/amp combination for the price.
Well worth listening to as far as I’m concerned – and this particular combination will find a place by my big picture window for when I want to listen and still enjoy a pretty view.
Massdrop Cavalli Liquid Carbon X (LCX)
This was my first exposure to any version of the “Liquid Carbon” amp. It was quickly apparent that, unlike with the MCTH, this unit is best listened to via its balanced output. This applied regardless of whether the input was single-ended (as with an RDAC pairing) or balanced. So that’s how I used it.
There’s a small, but immediately discernible, increase in apparent tonal weight when listening through the LCX, coupled with a more notable elevation of bass-weight. Bass-speed came across as being “quicker” than with the MCTH, but perhaps with a slight shift in textural conveyance. This resulted in the open, super-clear, rendering I got natively from the RDAC coming across as being somewhat thicker overall.
As with the MCTH … this resulted in the slight focus I was putting on the RDACs mid-range and treble producing a more balanced and cohesive end-result.
Raw resolution didn’t suffer ( actual resolution might actually be higher), but I felt with music that contained a lot of light, rapid, deft treble content lost a little of its sense of space and sparkle. Similarly, the scale and space of a venue, in live recordings, came across as being smaller … as if the walls have gotten a bit closer.
Beyond those changes, the LCX clearly lets the mid-range/treble “sweetness” of the RDAC play through and still has a distinctly pristine sound vs. other similar DACs, just with a different spin on the weight of what’s being presented. Not better, not worse … just different.
Which of these two would I choose?
That’s a question which, for me , actually had a very easy answer … until this showed up (on loan for a few days) yesterday:
Beyond saying I find that the RDAC/EC ZDT Jr. is a lot of fun to listen to, details (and a comparison between the three amps) will have to wait until I get to do a bit more listening with it (and try some other tubes in the EC ZDT Jr), so for now I’ll answer the question in the context of just the MCTH and LCX :
If I didn’t want to deal with tubes and was pairing with headphones with a “lighter” or “thinner” sound, then the RDAC + LCX combination would be an excellent way to go. As things stand, for me, I maintain a slight preference for the RDAC + MCTH combination. It winds up being the best combination of preservation of the traits of the RDAC I enjoyed the most, and which made it such an enjoyable listen, as well as a tweakable (via tube rolling) way to fine-tune that.
Beyond that, I’ve found that all three of these amps, paired with the RDAC, and the HD6XX, HD800(S) and, interestingly enough, even the HD660S are really good with these combinations. And the RDAC + MCTH + HD6XX is a lovely setup for under $800.
And THIS is when I miss living in the U.S.
I have my sights on the Cascade as well, so looking forward to hearing your thoughts on that. As for this DAC, between the unanswered questions about stolen PCB design and the December 18th (!!) ship date, I think I will pass.
Damn, the drama is real on the Massdrop forum… Hopefully this isn’t a problem, because I would like to get my hands on this DAC. But the evidence is pretty damning =( the ladders look dang near identical.
Yup-- this doesn’t look great. I’m sure it’s a fine DAC, and I was tempted by TORQ’s review, but I’m staying away from this one.
I mean there is definitely differences, and I’m really hoping that they are enough to not cause problems, I’m hoping it is just that they took from the overall design and didn’t overly copy it. But it does look improved upon. If nothing else.
Someone definitely copied the DAC here - definitely not a coincidence here. They may have a better quality PCB, definitely smaller footprint passive components, CPLD vs FPGA, and looks like they added a couple LEDs. I wonder if they also were able to get a copy of the CPLD/FPGA design code?
I don’t know if Airist had any knowledge of the copy; most likely they hired a sub-contractor for the design who did the copying.
The “original” DAC designer, sosolar, has spoken with his lawyers and looks like they will be taking some sort of action according to his post on Head-Fi. Also, interesting to see that soekris is confirming that this is definitely a copy:
Unless that’s a Kurniawan bottle, I’m going for the Latour.
Seriously though, I really appreciate your thoughtful write-up on this DAC. Everything else aside, you’ve set a high-bar for equipment reviews. Thanks!
The original designer of the R-2R board posted that it looked like Airist used cheaper components so the performance isn’t as good as it could be.
You got another point, 6656 using here is not suitable for ladder because it produce 50uvrms noise, while my design(6654) can go deep to 1uvpp that make the sound purer, darker. This should be a costdown point for Airist.
nope, 6654/6656 share the same footprint. It is simply a costdown(probably from 3 USD to 2USD), but 350 times noisier.
As an aside, I think it’s fascinating that there are so many positive reviews for a component which seems to exhibit shortcomings on all sorts of technical criteria. Maybe frequency response is really all that matters (it has since been measured and shown to be reasonably flat) and the rest are just details.
Yeah, all the reviews are pretty glowing. I picked one up, based on @Torq review on here. I hope they get this all figured out and it still ships because I am looking forward to this drop. Also I worry less about components and what not if it performs well… sum of its parts and what not.