Interested to see if anyone with capability could be interested in re-measure the RME DAC above and their new ultra low noise linear power supply DPS-2? Thinking about getting one. Most my efforts have been cancelling various noise, especially power supplies and that have proven very efficient. Already my current linear psu serving both the NAA output USB controller card from channel 1 and the RME from channel 2 made better. The DAC power stability through the entire device, should be of great importance to further up the signal-to-noise-relation?
RME claim <2 uV of ripple which is a very good value and further as far as I understand they feed back and correct irl if there are interference.
Do you have any measurements to back that up?
Have you any RME DPS-2 measurements to present?
Such measurement would be pointless.
As it is the RME ADI-2 FS has a total of noise and distortion 114dB lower than the signal level:
Any noise will be well below the threshold of audibility, so attempting to lower it further is just measurebating.
4 Likes
It is even some dB lower when using Direct DSD as shown i measurements by Jussi Lakko, Signalyst, and using his external DSP software @ DSD256 and some hefty modulators and filters. But there is always a challange to continue tweak the systems all parts to not add interference and acheive lower and lower level of audible parts of the sound that never was part of the musik, identified once removed. I think I will try this unit and see what will be the end result when listening to very gentle pieces of classical music or acoustic jazz. It would be interesting to see if those tiny, tiny note differences would be measurable, but not a condition for the decision.
Thanks for your reply 
Yet ASR likes to boast SINAD charts where people think much higher figures than 114 dB are important.
But you could always just ask Mathias Carstens of RME, on ASR forum, why he has designed such product. He’s pretty active there. He would be better person to explain.
I could, but would he give me a truly honest answer?
SINAD of well-designed audio equipment is well beyond both the real dynamic range of recorded music and the range of human hearing.
Having read all of the literature, it would seem it’s a product aimed at audiophiles who are convinced that LPS is better than SMPS. Does it make the DAC measure any better? Probably a couple of dB better SINAD on a product that already measures impeccably, Does it make the DAC sound any better? Highly unlikely, given the already very low noise and distortion well beyond the limits of hearing.
From the manual of the ADI-2 FS:
In other words, it will achieve its technical specs with just about any old power supply. However, they’re happy to sell you a PSU which costs as much as the DAC itself, marginally improves its technical performance, doesn’t improve its audible performance, but lets audiophiles smugly claim that they have the latest, greatest LPS connected to their DAC and also seeks to perpetuate the LPS>SMPS myth.
It’s a little sad that a company with RME’s reputation and pedigree has gone down this road…
1 Like
I don’t have any ADI-2 DAC’s, just two ADI-2 Pro’s. I don’t have the new RME’s PSU, but I can detect some differences with diferent PSUs on the ADI-2 Pro’s ADC performance. I personally use it mostly as DSD256 ADC. For example the medical grade SMPS brick I regularly use performs a bit better than the cheap stock PSU that the unit comes with.
2 Likes
It seems I just have to try it and judge for myself. As I said above, I usually detect a difference between LPS and SMPS. Now any LPS is not the solution, it better provide a transformer with static shield between primary and secondaries, magnetic outer shield, 105 degrees el caps providing low ESR and sufficient energy bank (here some new super cap applications are interesting), schottky diodes, frequent decoupling using ultra fast film caps, non-inductive, and zobel net to prevent resonances. Then the regulator will have a far better base to work from, as there are limits to what rippel rejection the regulator are capable of. There seem to be some decision made among technicians/designers that any noise or irregularities above the audible frequency range is not really essential. Noise in any form create an action/reaction in circuitry. Worst case is electronic parts subjected to noise beyond its bandwidth, trying to modulate it.
My personal opinion, as it has evolved during the years, is that clean PSU’s are essential for any device maximal performance and “good enough” as decided by brands/manufacturers rarely are. Even if there are claims that certain SINAD levels are to be considered inaudible, still things happen in terms of SQ. I have a strong belief in the hard to trace intermodulation distortion on system level. Things interact, that is not supposed to and is hard to foresee while measuring on single device in lab. And the entire network of devices interconnected is a great antenna.
So, all PSU’s performance matter. So do all interconnects impedances and shields that connects every devices grounds. Ground management make wonders to SQ. So also breaking galvanic interconnections. Not only WAN but also my LAN is fiber optics today, isolating the three clusters of devices from eachother. I do not do this to show of the number of LPS as an anxious audiophile, that is completely irrelevent to me. I focus on the total level of superimposed noise irrespective of wether it is analog or digital circuitry. Electric noise spreads as stray and eddy. And, It adds to eachother and becomes audible even if just one device at the time does not.
1 Like