• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Measurements of RME ADI-2 DAC and Headphone Amp

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
4,940
Likes
19,692
Location
Paris
Level matched, yes. Blind, no. Maybe the Transporter is just better?

What was the output voltage for both?

Again, "better" doesn't mean anything in Audio world. IMHO, the "best" is nothing but the most transparent possible. Meaning: a unit which reach all the thresholds mentioned here: Audibility Thresholds of amp and DAC measurements

We don't know much about Transformer's real world performance. Honestly, I would be really surprised if it outperforms the D50 in a measurably audible way!
 

Majestyk

Member
Joined
Oct 12, 2018
Messages
75
Likes
30
Location
Vancouver
I don't have the answers to your questions. and "LEAPS" might have been exaggerated but I am not the bigger or more expensive is better type. I do not believe in burn-in nor do I believe in expensive power and audio cables.

What I heard with the Transporter (vs the D50) was deeper bass, better transparency, slightly better (very slightly) soundstage. This isn't the first time I've compared a device with the Transporter in which I wanted the Transporter to lose. I compared a Squeezebox Touch as well and I really wanted to keep the Touch (I preferred the smaller size and touch screen) but it wasn't even close. I will say, and this is only going by memory, that the D50 was closer to the Transporter than the Touch was.

Anyway, I would have loved to come here to this forum and tell everyone that my cheap D50 sounded as good or better than the Transporter but it wasn't to be. Like I stated above, one of these days I'll try another Topping with the AKM. (Although the Touch had the AKM as well)
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
What I heard with the Transporter (vs the D50) was deeper bass, better transparency, slightly better (very slightly) soundstage.
Thanks. We need to go further. What do you mean by "better transparency"? Less distortion? no coloration?
Are you sure you did the levelmatching well? How did you do?
 

entrophy

Member
Joined
Jul 14, 2019
Messages
18
Likes
20
Location
Europe
I have recently purchased the RME ADI-2 DAC FS, and would like to give my own impressions, especially a few points I think haven't been mentioned yet.

First, a confession: I was a Schiit user. I have been off the Schiit for 2 weeks now. I want to thank Amir for the encouragement that he, unknowingly, gave me through ASR. Like so many others, I was lured into Schiit because I liked how easy and inexpensive it was to start with Schiit. Spend a hundred on the DAC. Spend a hundred more on the headphone amp. And one by one I was sucked into it. Or, more precisely, suckered into it. Before I knew it, I had spent more than 400. At the same time I was into Schiit I became aware of Amir's criticism of Schiit, but dismissed it because, let's be honest, the takedowns of Amir at other places were really funny sometimes. Also, at the time I did not see which alternatives Amir could offer other than some questionable Chinese products. So in the end I had this setup: Modi 2 Uber (which I later replaced with the Modi 3), Magni 3, Sys, and it still didn't work alright. High sensitivity to hum and coil whine, ground loops, external power supplies that ran hot even when the Schiits were turned off mechanically, noise creeping in through USB, and sometimes the Modi 3 would just mute on my Mac and not produce any sound. The whole stack was connected to my active Yamaha monitors, which meant going from unbalanced cinch out to balanced in, which gave me other problems.

I pulled the plug, and replaced all of it with a single RME ADI-2 DAC FS. Now I run one machine over USB to the RME, the other through optical, and output through XLR to the Yamahas. Which brings me to my main points:

Typical active monitors - and most people with a DAC/amp on their desktop will probably use actives - have balanced inputs. The ADI-2 DAC has balanced outputs, which means cable length becomes a non-issue. Many competing products, and I use the term loosely, have unbalanced cinch out only.

The ADI-2 DAC has an excellent equalizer built-in. Frequency manipulation happens in the DAC at its internal resolution in near-real-time. Again, if I were to use a Schiit DAC or similar I'd have to get either a separate equalizer, or do it in the player software, with unknown precision and latency.

The ADI-2 DAC also gives me a built-in tool to check for bit-perfect playback. Just download the test files from RME, and play them out, and the ADI-2 DAC will auto-detect the samples and tell you on the display if the samples arrived in the DAC unaltered. I haven't seen anything like this from the competition, certainly not from Schiit where I'd be flying blind. By the way, the test file samples range from 16 to 32 bit.

The ADI-2 DAC has a very nice LCD display. What's even nicer is that it will display information like the currently set clock source, resolution, volume, and channel balance. Again this is something where you'd be flying blind with most other competitors.

So apart from the RME device performing much better than the competition, we also need to look at the actual value per dollar spent. Or, to reverse it, how much would I have to spend to get a comparable setup from other vendors, e.g. from Schiit?

Something like this (and I'm being very generous in assuming that performance would be comparable, which it is not): Schiit Gungnir ($850) plus Schiit Magni 3 ($100) plus Schiit Loki ($150) = $1100. And then I still don't have the display, the IEM out, and the profile-based customizability.

And even if I started from my current stack and expanded it, I'd end up as follows: Schiit Magni 3 ($100) plus Schiit Modi 3 ($100) plus Schiit Sys ($50) plus Schiit Loki ($150) = $400. And this stack will definitely not give me an approximation of the sound quality or flexibility or connectivity that I get out-of-the-box from the ADI-2 DAC.

In the end I think RME identified the weak point of much of its competition: assembling your own stack makes little sense for the majority of users who really want to end up with a setup where they have multiple inputs, headphone amp, balanced out, and user presets. RME correctly engineered it into a single device that you set up once, and then be done with it for many years because it just works.

I'd also like to thank Amir for not only pointing out, repeatedly, the deficiencies of the competition, but also for doing in-depth measurements of RME products and for pointing out a real alternative where I, as a former user of "hip" made-outside-China products, get real value that not only sounds good but also measures as it should at its price point.

A final word, and an offer: the ADI-2 DAC is not perfectly immune to USB noise and ground loops. If there's interest, I can write up how I solved those.
 

suttondesign

Addicted to Fun and Learning
Forum Donor
Joined
Feb 4, 2019
Messages
732
Likes
1,310
Location
Bellingham, WA
Typical active monitors - and most people with a DAC/amp on their desktop will probably use actives - have balanced inputs. The ADI-2 DAC has balanced outputs, which means cable length becomes a non-issue. Many competing products, and I use the term loosely, have unbalanced cinch out only.

The ADI-2 DAC has an excellent equalizer built-in. Frequency manipulation happens in the DAC at its internal resolution in near-real-time. Again, if I were to use a Schiit DAC or similar I'd have to get either a separate equalizer, or do it in the player software, with unknown precision and latency.

I will be switching over to balanced initially with the Dutch 8c, but I have two separate digital sources in my setup, so the RME is good to go even though the Dutch do an A-D again. I will ultimately get a switch for digital to the Dutch.

I LOVE the parametric in the RME which lets me deal with the huge room mode at 79hz in my listening room, also the balance control that helps with my hearing imbalance.
 

entrophy

Member
Joined
Jul 14, 2019
Messages
18
Likes
20
Location
Europe
Ok, so this is from a non-electrical engineer, and Amir and others can probably shed more light on the correct EE background here, but basically there are at least two types of unwanted noise that could hit the signal chain of an ADI-2 DAC setup: USB port noise and ground loops. The first usually rears its ugly head through chirping noise whenever you move the mouse or when there's SSD activity, i.e. when the ADI-2 DAC is connected to a computer that itself does not make this noise, or at loudness not perceivable. The second depends on the setup, in my case I have two computers connected to the same ADI-2 DAC: one is a MBP laptop that's mains powered, the other is a gaming PC. Whenever the GPU of the gaming PC is under load at high frame-rates, it produces coil-whine that, surprisingly, travels through ground wires into the laptop and from there to the USB port through the USB cable into the ADI-2 DAC into the XLR cables into the active monitors. Funny enough, even when the ADI-2 DAC is turned off but the Yamahas are turned on, the coil whine is audible through the Yamahas. Stopping the GPU from working, or pulling the USB plug from the ADI-2 DAC, will immediately stop the coil whine noise.

How to fix USB port noise: this one won't cost you money but patience. I know it has been discussed ad nauseam, but it really works: change the port where the ADI-2 DAC is connected. At least in my setup I have found that a USB dock is not really an issue. My MBP has 4 USB-C ports, two on the left and two on the right. On the left I had initially connected, from top to bottom: Apple Ethernet adapter (through TB3/TB2 adapters), ADI-2 DAC through Apple USB-C to USB-A adapter. Right side: USB-C to DP adapter, USB-C dock with MBP power supply feeding MBP through dock. I have changed it to this, top to bottom: USB-C dock with ADI-2 DAC, Apple Ethernet adapter. Right side: USB-C to DP adapter, USB-C power directly into port.

This setup completely eliminated any USB noise from the laptop hitting the ADI-2 DAC, even though the ADI-2 DAC is now connected to an evil USB dock. As always, if you are experiencing this issue you probably need to try out a fair number of permutations until you find one that is noise-free.

How to eliminate buzz/hum/coil whine leaking through ground wires: having identified that pulling the USB plug from the ADI-2 DAC will break the ground loop, and knowing that the ADI-2 DAC is not USB-powered, I purchased an iFi iDefender 3.0 for around $50. I'm not entirely sure how it works, and I haven't found any PCB images through Google, but it appears that this dongle checks the phase of the ground wires on both sides of the dongle, and if it detects a ground loop it will break it. This eliminated 99% of the coil whine traveling from the GPU in one machine to the laptop to the ADI-2 DAC to the Yamaha monitors. I can still hear some very faint buzzing whenever the GPU is active, so I have ordered line isolators for the monitors (Radial IC-1) to see if they get rid of it. I shall report back next week.
 

entrophy

Member
Joined
Jul 14, 2019
Messages
18
Likes
20
Location
Europe
Another point on RME, which might help explain why it performs as well as it does:

RME have a background in producing professional-level 19" devices for music and post studios. (I happen to have their flagship device, the Fireface UFX+, too.) That's the kind of level where you can't afford loose heatsinks or noise pollution from inadequate topologies. In a studio, a device is expected to work 24/7 for years, and it's supposed to give you fidelity way beyond consumer gear. In hindsight, it's clear that companies like Schiit are well-meaning, but coming from the consumer space they just don't have the strict regime and discipline and EE background to produce devices that perform as expected. I'm not saying they always perform worse than cheap consumer gear, but they don't perform better, either, and this is probably where it went wrong: from a hobbyist adventure reaching into high-end when they weren't equipped for the high-end.

Whereas a company like RME has always been at the high-end, and now they happen to use all that expertise to produce a DAC/amp that, cynical as it may sound at first, is quite affordable given the feature set and quality. As Tim Cook would say, hey, it's really a coffee every day for a year. ;)
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
[ground loop].. in my case I have two computers connected to the same ADI-2 DAC: one is a MBP laptop that's mains powered, the other is a gaming PC.
How do you connect two PCs to the DAC, it has only one USB input?
 

entrophy

Member
Joined
Jul 14, 2019
Messages
18
Likes
20
Location
Europe
Exactly. The ground loop is not happening through the S/PDIF but through one of the many other possibilities: computer PSU leaking into the power strip, video cable (both computers use the same display), or keyboard/mouse cable attached to display (the shared display has a USB hub builtin that also switches based on which video input is selected). Probably not the most common setup, but seeing how many users are struggling with hum/buzz in their active monitors I guess ground loops are not uncommon, one way or the other.
 
Last edited:

MC_RME

Addicted to Fun and Learning
Technical Expert
Audio Company
Joined
May 15, 2019
Messages
854
Likes
3,564
entrophy, thanks for your detailed information and view of this topic. And indeed I have some comments.

First the ADI-2 Pro and DAC are 100% immune against ground loops or USB noise. One has to take this statement straight as to what it says. In my definition any electronic applicance that does not make ground currents audible is immune against ground loops. That is the case for many devices these days, except amplifiers. They often make the ground current audible, and I bet it is the same in your case (the active monitors). How to check what I mean: in your setup use the headphone output. It should be 100% clean. As this is identical to what the ADIs receive and put out at XLR/RCA the signal there is 100% clean as well. As you wrote: noise continued even after removing power supply. No appliance is expected to BREAK grond loops. But a good unit should not react on them. This is accomplished by having milliohms of ground-resistance bewteen all I/O jacks. A current flowing through the ground of the unit will not modulate its audio signal. Of course internal designs like balanced signal paths also help.

There is one weak point, though: the SMPS (power supply). It does not provide full isolation against ground loop noises, even produces its own (leakage current). But you were able to get it working nonetheless, congrats. IMO it pays out to have a battery that can drive the ADI totally isolated from mains, just to find out if that is one of the problematic connections.
 

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
There is one weak point, though: the SMPS (power supply). It does not provide full isolation against ground loop noises, even produces its own (leakage current). But you were able to get it working nonetheless, congrats. IMO it pays out to have a battery that can drive the ADI totally isolated from mains, just to find out if that is one of the problematic connections.
Thanks for the info. Would you please elaborate a bit on this because I didn't quite get it. I assume most switch mode supply has double isolated output so the mains noise which is at very low frequency won't come through the coupling circuit in the supply. It does produce its own leakage current but at very high frequency so I don't think it's comparable to mains leakage. Also you may educate me a bit, probably I'm not educated enough to aware this problem. Thanks.
 

MC_RME

Addicted to Fun and Learning
Technical Expert
Audio Company
Joined
May 15, 2019
Messages
854
Likes
3,564
Search for user KSTR posts, he explained that in detail already.
 

entrophy

Member
Joined
Jul 14, 2019
Messages
18
Likes
20
Location
Europe
There is one weak point, though: the SMPS (power supply). It does not provide full isolation against ground loop noises, even produces its own (leakage current). But you were able to get it working nonetheless, congrats. IMO it pays out to have a battery that can drive the ADI totally isolated from mains, just to find out if that is one of the problematic connections.

Thanks for the detailed response, Matthias.

I had already done tests to make sure that it's indeed the USB port of the ADI-2 DAC, and not the power connection, that closes the ground loop (or rather, one of the ground loops). Verification: (1) No iFi iDefender dongle in the USB connection path. (2) Let the GPU do its coil whine thing while the ADI-2 DAC is turned off. The monitors buzz. (3) Remove the power connector from the ADI-2 DAC. The monitors buzz, and the noise is not changed by the removal. (4) Remove the USB connector from the ADI-2 DAC. The monitors stop buzzing. (5) Re-attach the USB connector to the ADI-2 DAC. The monitors buzz. (6) Insert the iFi iDefender into the USB signal path leading into the ADI-2 DAC. The monitors stop buzzing.

Please note that in the above setup the USB connection is not actually between the computer that has the coil whine and the ADI-2 DAC. For this test they are not directly connected at all. The coil whine is making a round-trip across other devices, and as shown in the test, it's the USB input of the ADI-2 DAC that's closing a ground loop, and assuming that the iFi iDefender is really a ground breaker, it appears that at least the USB port of the ADI-2 DAC that I have is not immune to ground loops.

Of course it's entirely possible that this is an isolated issue with my particular ADI-2 DAC, and I'd be more than happy to send it in to RME for inspection.
 
Last edited:

entrophy

Member
Joined
Jul 14, 2019
Messages
18
Likes
20
Location
Europe
I just ordered the Delock 62982 galvanic isolator (3kV optical isolation, 480mbps), it costs only half as much as the Intona device that RME recommends. Should arrive Wednesday, then I'll report back.
 

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
4,636
Likes
2,809
I just ordered the Delock 62982 galvanic isolator (3kV optical isolation, 480mbps), it costs only half as much as the Intona device that RME recommends. Should arrive Wednesday, then I'll report back.

Would be interested to know how ground is isolated.

I would guess the data lines are optically isolated and power (when not using optional external PSU) is using DC to DC converters.
 
Top Bottom