RME DPS-2 Linear Power Supply

[antcollinet]

I have a large experience in A/B and blind testing,

Forgive my scepticism: but,

when i say obvious ... it is obvious.

and we have a device which measures as audibly perfect with the stock power supply, it is a bit difficult to understand how it could be come obviously better with any other power supply. Once it is audibly perfect, any improvement is still 'only' audibly perfect - there can be no obvious audible improvement.

Unless your PSU is actually causing more distortion or noise, or FR variation in a way you prefer. Or your system suffers from a ground loop that the battery breaks - but then you'd be hearing obvious ground noise.


More likely then, especially given the difficulty of ABing a power supply change, and then within the time frame of echoic memory, that your comparison was insufficiently controlled, or tells remained.

 

[Purité Audio]

You have to have two identical units, obviously easier if you sell them, tried it with both Metric Halo and Weiss ordered the power supplies from the then ‘guru’ Paul Hynes ha d to wait months for them didn’t make an ounce of difference because both components were properly engineered.
‘Just my Imagination running away with me’
Keith

 

[Aquascop]

I am not trying to convince anyone or proove anything. I was just sharing my experience on the PS of the ADI-2, saying that is was worth the trial in comparaison of the PS solution now proposed by RME.

> You have to have two identical units
by the way, I have 3 of them, same generation (so same DAC component), in full 3-way active.

 

[MAB]

I am not trying to convince anyone or proove anything.

Just convinced yourself. That's very easy to do. Unfortunately, ears are poor measuring devices.

I was just sharing my experience on the PS of the ADI-2, saying that is was worth the trial in comparaison of the PS solution now proposed by RME.

It's an anecdote. Sighted bias. In 100 tries would not be able to determine differences without looking or being told which PS was in use. Nobody is immune:

Audio Listening Test That Fooled the Hollywood Sapphire Group

Some of the most highly trained audiophiles in the industry were asked to do a blind “A-B” listening comparison test of two audio circuits. Were they able to hear a difference, or were they fooled?

www.audioholics.com

And even the true believers can't tell a Futterman from a Pioneer receiver (see page 80 for the classic test):
https://americanradiohistory.com/Archive-HiFI-Stereo/80s/HiFi-Stereo-Review-1987-01.pdf

> You have to have two identical units
by the way, I have 3 of them, same generation (so same DAC component), in full 3-way active.

Three independent clocks in an active system... How you sync the clocks? Odd, RME makes interfaces with enough channels you can have good performance and have your three-way active not suffer from clock drift.

 

[Chrise36]

I am not trying to convince anyone or proove anything. I was just sharing my experience on the PS of the ADI-2, saying that is was worth the trial in comparaison of the PS solution now proposed by RME.

> You have to have two identical units
by the way, I have 3 of them, same generation (so same DAC component), in full 3-way active.

What is your system chain?

 

[Aquascop]

What is your system chain?

PC with foobar playing 6 ways FIR, to RME digiface USB (and master clock mutec), to 3 RME ADI-2 DAC, to 3 Dual Mono Amps, to Speakers

 

[Chrise36]

PC with foobar playing 6 ways FIR, to RME digiface USB (and master clock mutec), to 3 RME ADI-2 DAC, to 3 Dual Mono Amps, to Speakers

Any audible issues should come from ground loops or smps related noise or hum affecting they amps.Especially pc psu noise is hard to isolate with usb.

 

[MAB]

Any audible issues should come from ground loops or smps related noise or hum affecting they amps.Especially pc psu noise is hard to isolate with usb.

Yeah, multiple USB introduce lots of ground loop vulnerabilities. Battery can help that, assuming the rest of the ground is worked out.

Regarding the other aspect of this, RME explains why this is an extremely odd implementation. Assuming these are non-Pro, then no way to sync clocks, not sure what else to say. RME explains it best:

ADI-2 DAC multi-interface operation - 3xADI-2 DACs? 6ch? (Page 1) — ADI-2 & 2/4 Pro series, ADI-2 DAC series — RME User Forum

ADI-2 DAC multi-interface operation - 3xADI-2 DACs? 6ch? (Page 1) — ADI-2 & 2/4 Pro series, ADI-2 DAC series — RME User Forum —

www.forum.rme-audio.de

 

[Aquascop]

Yeah, multiple USB introduce lots of ground loop vulnerabilities. Battery can help that, assuming the rest of the ground is worked out.

Regarding the other aspect of this, RME explains why this is an extremely odd implementation.

okay, i think we are a little bit off topic, but let me explain more my approach :

Regarding USB : there is only one USB link from the PC to the digiface. no loops or impossible synchro.
As I am a bit of a die-hard, the PC-USB output is done via a matrix element H usb board, with external (battery, again) power supply in input.

Regarding clocks :
All 6 channels are then clocked together by the same unit, the digiface.
And the digiface is now clocked by a mutec master clock ( obvious, ... if i may, ... improvement)

then, 3 fibers (perfect electrical insulation) are linking the digiface to the 3 ADI-2 units.

as the ADI-2's are locked to the clock of their optical input, no troubles, there is only one master clock, the mutec unit in my case, and the 6 channels are locked on this clock.

it is more or less what 'ramses' says in the link above given by #MAB, and no-one tells that it is an odd-implementation. complex, for sure, but perfectly managable.

then, and this is where the ADI-2 is ideal for me, 2 ways of playing music :
- the simple/family way : i use the ADI-2 EQ capability to filter each channel bandwidth for its dedicated speaker (woofer/medium/tweeter), and correct 3 major amplitude points.
- the High End way, where i use Foobar and a 6 way FIR, with EQ OFF on the ADI-2's : as in this case, i can manage amplitude & phase corrections in the FIR, the result is just awesome, providing that you manage the measurement tools, and correction chain (REW & rephase). This point took me a few months to master it, honnestly.

 

[Salt]

Respect for implementing up to date technology. But to drive speakers with distortions times more than signal processing and amplification: is it worth it?

 

[Aquascop]

Respect for implementing up to date technology. But to drive speakers with distortions times more than signal processing and amplification: is it worth it?

not sure to understand your question, sorry ... but yes, it is definitely worth it. Lot of fun and learning in the deployment of the complete chain, but in the end, i have an audio quality at home that I wouldn't have been able to reach otherwise. to my opinin, future is to full active speakers, with embedded amplifiers and DSP.

 

[antcollinet]

And the digiface is now clocked by a mutec master clock ( obvious, ... if i may, ... improvement)

(my bold) More evidence that whatever testing you are doing is not controlling for bias.

 

[Aquascop]

(my bold) More evidence that whatever testing you are doing is not controlling for bias.

did you ever try reclocking ?

 

[richard12511]

To be honest I expected the manual to show differences in measured performance in XLR/RCA/HP out and ADC in. If there are improvements, it would be a selling point, and let the owner choose between price x marginal performance improvement.

Genuine question, what would be a "marginal performance improvement" in this scenario? I thought RME products were all well beyond perfect in terms of audible defects, so there's no performance improvement to be had there. Would it be a reliability improvement?

 

[antcollinet]

did you ever try reclocking ?

The only point to a master clock in home audio is to synchronise otherwise unsynchronised devices.

Your RME Digiface already uses a single internal clock so sync is not an issue, and it already has internal clock jitter below the level of audibility (see : Is Jitter Audible) both due to absolute level, and profile (spread spectrum rather than sinusoidal)

So your clock master (again) cannot make an audible improvement to jitter - and I'm ignoring here the risk of the master clock acquiring jitter in it's interface to the RME, and actually making things worse.

I'm also ignoring that the jitter on the optical input of the RME DACS will likely be much worse than whatever jitter is on the internal clock of the Digiface. Again not an issue since clock recovery/jitter rejection on any competent DAC is a solved issue.


So no, I've not tried re-clocking, because I know (from the science and engineering) it is pointless. As I pointed out above, the fact you hear obvious differences tells us only about the validity of your comparisons - and nothing about actual audibility of the use of a master clock.

 

[Aquascop]

So no, I've not tried re-clocking, because I know (from the science and engineering) it is pointless.

What I always repeat to my team (of highly skilled engineers in high tech) is that the main quality of an engineer is to keep 'til the end of its carreer, his capability to learn, to be astonished, and to be able to question certitudes, and habits that too often become standards and false references. Without this mindset, you are quickly left behind by guys who 'dare' other things.

This is what i always liked in Audio : challenging science with hearing tests.

some things are easily explainable, others not, but many things are worth a trial, just to question your habits.

 

[VintageFlanker]

And what is really surprising here, is that when i say obvious ... it is obvious. no need to spend hours comparing

No, you're right: a few minutes, if properly done, are enough. Property done. Doesn't seem like it's been the case in this scenario.

I was also quite astonished, and i shared my experiment on a french HiFi forum : I even sent my battery to one of the most skeptical expert of the group ... and 24h later, he proposed me to keep my battery and bought me a new one on amazon ...

If the place you are mentioning is the one I'm thinking about (Forum-hifi.fr), then you have to know that both the aforementioned and ASR are absolutely nothing alike. If I want to remain polite.

I am not trying to convince anyone or proove anything

Well... Seems like you just are. You would have a hard time trying to convince anyone around here by such statements as "it's so obvious". However, if you're interested by checking what your battery is doing to the ADI-2 DAC performance, then we can arrange a loan, that way I could check whatever impact this would have on perceived sound reproduction and come up with data. If you're confortable doing measurements yourself, everybody would be glad to see them !

This is what i always liked in Audio : challenging science with hearing tests.

Even better with controlled hearing tests. That way it would remain science-related all along.

What I always repeat to my team (of highly skilled engineers in high tech) is that the main quality of an engineer is to keep 'til the end of its carreer, his capability to learn, to be astonished, and to be able to question certitudes, and habits that too often become standards and false references. Without this mindset, you are quickly left behind by guys who 'dare' other things.

Two threads where you could share your "innovative" thoughts:

"Things that cannot be measured"

Something the subjectivist crowd often brings up. "There are things we cannot measure but the human ear/brain can hear it." We retort, an analyzer can hear much better than any human can. Which is the truth. But thinking about the question I did being to wonder... Could it be possible for there...

www.audiosciencereview.com

Master Thread: Are measurements Everything or Nothing?

Seems like so many review threads get challenged with: 1. Measurements are not everything. 2. You all never listen. 3. I trust my ears, not graphs. 4. I don't listen to graphs. I listen to music. 5. You all must not listen to music at all. 6. Why don't you all buy the best SINAD gear? 7...

www.audiosciencereview.com

 

[antcollinet]

What I always repeat to my team (of highly skilled engineers in high tech) is that the main quality of an engineer is to keep 'til the end of its carreer, his capability to learn, to be astonished, and to be able to question certitudes, and habits that too often become standards and false references. Without this mindset, you are quickly left behind by guys who 'dare' other things.

This is what i always liked in Audio : challenging science with hearing tests.

some things are easily explainable, others not, but many things are worth a trial, just to question your habits.

Sure - keep an open mind. Just not so open your brains fall out.

And, as an engineer, if you perceive something that the engineering says is not there, it is not enough just to say - "oh well, obviously we don't understand the engineering" It is necessary to investigate and understand why. Otherwise there is a very good chance you are just fooling yourself - and probably others.

 

[MAB]

What I always repeat to my team (of highly skilled engineers in high tech) is that the main quality of an engineer is to keep 'til the end of its carreer, his capability to learn, to be astonished, and to be able to question certitudes, and habits that too often become standards and false references. Without this mindset, you are quickly left behind by guys who 'dare' other things.

This is what i always liked in Audio : challenging science with hearing tests.

some things are easily explainable, others not, but many things are worth a trial, just to question your habits.

I am a physicist. I have a team. We are highly skilled. So do many here. Some are just highly skilled! You are not alone here on ASR in that respect. Appeal to authority will not work.

While I am continually surprised by nature and science, the science of human hearing and perception is a field itself, and many engineers (and physicists) have who can make and evaluate technically excellent devices have no idea how this perception works. The simplest thing I can say is our ears are for predictions, not for measurements. Aside from the technical reviews on devices on ASR, there are plenty of posts from experts across multiple disciplines where you can read up on how the human ear and brain work. And, since it's an online forum, lots of clutter in the way too!

BTW, welcome to ASR, hope you stick around. Likely we have more in common that not. I appreciate you are looking for best performance.

 

[audiojj]

If RME would like to expand further into the consumer market I can think of better ways of doing it that this. For example a Qudelix competitor leveraging their Remote software would awesome, as would a power amplifier to complement the ADI 2.

This on the other hand look like a solution to a problem RME has already solved with built in switching power supplies.

What is expected of all owners of systems with all-in-one monitors (Genelec, Neumann, etc.) is a digital preamp (with all possible digital inputs and outputs) without a DA converter and without headphone output. But with an AD converter, equalization, loudness and RME quality. And why not a streamer. A bit like MiniDSP's SHD Studio.