Measurements of RME ADI-2 DAC and Headphone Amp

[Gus141]

Technically, in the actual signal flow, both are exactly equivalent. ...

Thanks for the clear explanation. Regarding high power mode, is there any advantage to forcing Hi-Power=On (w/ Auto-Ref=Off) for Phones when my headphones really don’t need high power and will be operating around -40 to -30 dB volume level in Hi-Power Mode?

I ask because I had a headphone amp (SoundBlaster X7) where its low-power mode was a fixed attenuation of the high-power output and had some reduced dynamic range; in that case Low-power did not sound as good as just using High-power and turnign down the volume.

With the ADI-2 DAC, I worried about too much digital-volume attenuation, so I chose Lo-Power mode to keep the volume level higher.

Note: I don’t use Line Out, just Phones and IEM outs.

 

[Schackmannen]

Thanks for the clear explanation. Regarding high power mode, is there any advantage to forcing Hi-Power=On (w/ Auto-Ref=Off) for Phones when my headphones really don’t need high power and will be operating around -40 to -50 dB volume level in Hi-Power Mode?

I ask because I had a headphone amp (SoundBlaster X7) where its low-power mode was a fixed attenuation of the high-power output and had some reduced dynamic range; in that case Low-power did not sound as good as just using High-power and turnign down the volume.

With the ADI-2 DAC, I worried about too much digital-volume attenuation, so I chose Lo-Power mode to keep the volume level higher.

Note: I don’t use Line Out, just Phones and IEM outs.

There is no advantage to forcing Hi-Power mode to be on all the time, in fact it will be theoretically performing worse than the Lo-Power mode at the same output since it has more noise and therefore lower SNR/Dynamic range. But I doubt you'll hear any noise with Hi-Power mode on anyways so if that is the case there will be no audible difference between the two modes. Personally I would just use auto-ref level on the Phones output and not think about it again.

 

[KSTR]

With the ADI-2 DAC, I worried about too much digital-volume attenuation, so I chose Lo-Power mode to keep the volume level higher.

A sane decision and I would have recommended the same. As long as you don't really need high-power mode (could be an issue with very dynamic classical recordings, with very low volumes at times), you can't go wrong by forcing low-power mode. It will operate the DAC chip in its optimum range.

 

[MC_RME]

The hardware is in the same state in both cases (same THD, SNR etc). The scale needs to be different when you have AutoRef On, because if you up the volume and the hardware changes into High Power mode - then you need 15 dB on top to be displayed. It's easy to understand with the 2 steps of the Phones out, but can drive one crazy with the 4 steps of the line output (not by using it, but by trying to understand it)

Edit: Oh wow, so many replies, thanks guys!

 

[neutronsix]

Good morning to all of you. And an infinite thank you for all your valuable technical information. I can't find where and if, you've already written about it. I ask you if, like other drivers, here too.
I have the possibility to reset the number of errors. I have tried in many ways how to reinstall the driver, increase the buffer, etc. But the number doesn't change. Can't be done?

 

[MC_RME]

This line shows 0 errors of the CRC 5 and CRC 16 kind. Got it?

 

[neutronsix]

@MC_RME. Thanks for the lightning response. If I understand correctly the complete string then it is "crc5/16 err". Until recently, I only knew the CRC32 code...

 

[Svperstar]

Is there a new revision based off the 4499 coming out?

 

[Veri]

Is there a new revision based off the 4499 coming out?

No, it really just got updated from 4990>4993.

 

[Alan S]

RME has updated the ADI-2 DAC twice.

 

[jsts]

Hello, folks!

I will try briefly.

MacPro, Mojave 10.14.6. Front USB port, RME ADI-2 DAC FS (v2). Safari, Youtube.

At 192 kilohertz, the sound of the video is output without problems.

At 384 kilohertz ONLY when starting / rewinding / stopping the video, distortions (cracking, clicks) are heard.

At 768 kilohertz, distortion (crackling, clicking) is heard constantly.

In case of distortion, the following messages are displayed in the log:


HALS_OverloadMessage.cpp: 165: perform: Audio IO Overload inputs: '' outputs: 'AppleUSBAudioEngine: RME: ADI-2 DAC (211) 1,2' cause: 'Unknown' prewarming: no recovering: no


Why does overload occur at 384-768 kHz?

 

[Veri]

You realise though, that setting the OSX oversampler to 384/768 kHz is actually inferior to the steeper 'Fast' oversampling filters in the RME? There's really very little point setting your MIDI settings so high. Your DAC is better at this operation than your computer's default.

Still, it should work. Best to contact RME.

 

[MC_RME]

Hello, folks!

I will try briefly.

MacPro, Mojave 10.14.6. Front USB port, RME ADI-2 DAC FS (v2). Safari, Youtube.

At 192 kilohertz, the sound of the video is output without problems.

At 384 kilohertz ONLY when starting / rewinding / stopping the video, distortions (cracking, clicks) are heard.

At 768 kilohertz, distortion (crackling, clicking) is heard constantly.

In case of distortion, the following messages are displayed in the log:

HALS_OverloadMessage.cpp: 165: perform: Audio IO Overload inputs: '' outputs: 'AppleUSBAudioEngine: RME: ADI-2 DAC (211) 1,2' cause: 'Unknown' prewarming: no recovering: no

Why does overload occur at 384-768 kHz?

Safari and Youtube are no reference when working in such exotic sample rates. Please note that under Mac the ADIs are Class Compliant - they run without RME drivers, so we can not influence the behavior of the OS. I also expect that dedicated players (JRiver, Audrivana, HQPlayer) will work at such sample rates on your computer. They do here with a Mac mini from 2012...

 

[jsts]

Oh, I see.

Yeah, iTunes, Audirvana+, Jriver and the like work great.

Thank you!

 

[ezra_s]

@MC_RME. Thanks for the lightning response. If I understand correctly the complete string then it is "crc5/16 err". Until recently, I only knew the CRC32 code...

don't worry I was on the same boat Funny thing is when I bought it I knew that but got confused later on, not sure why. Glad it is clarified.

 

[ShiZo]

God, I want to know the measured performance of the newest one so bad! Too bad it's my only dac . Mostly for the multitone test. The other measurements are a little bit more predictable. But with the measurements of the a90, the iem performance is in competition.

 

[TimF]

Most of my audio is played from a computer running JRiver Media suite, and now and then an actual CD played via a cd transport. Question: for small sound adjustments are the JRiver DSP Studio tools (Equalizer and Parametric Equalizer) good and sufficent? That is, if I want to upgrade my DAC is something like a Topping D70 or D90, or the recent Gustard DAC's, adequate in combination with JRiver and therefore there is no need to buy the RME ADI-2 for its EQ functions? Any opinions out there?

 

[Patrick1958]

I use PEQ and convolution in JRiver and have no complaints. IMO very adequate implementation of DSP in JRiver.
From what i read on the RME there are only 5 bands in the PEQ, whereas with JRiver you can use as many bands as you like/need.

 

[maverickronin]

Most of my audio is played from a computer running JRiver Media suite, and now and then an actual CD played via a cd transport. Question: for small sound adjustments are the JRiver DSP Studio tools (Equalizer and Parametric Equalizer) good and sufficent? That is, if I want to upgrade my DAC is something like a Topping D70 or D90, or the recent Gustard DAC's, adequate in combination with JRiver and therefore there is no need to buy the RME ADI-2 for its EQ functions? Any opinions out there?

If you're going to use it for music played through something with a good plugin structure like JRiver than you may not need the extra DSP features.

I have an ADI-2 DAC and I really love that I can apply the EQ to sound from other devices like game consoles and have have system wide crossfeed from my couputer without using latency and stutter prone virtual loopbacks though a DAW. If you usage case is straight music playback though JRiver then none of that really applies to you

OTOH, if you prefer to listen at low levels, the RME's adjustable loudness curve DSP is very useful though and I'm not away of any other DSP system with the same feature.

 

[DivineCurrent]

I had something weird happen to me today. For a few months now, I've only been using the Optical input with the ADI-2 DAC, with zero issues. I switched my desktop to USB today after adding another source, and I was greeted with stuttering music listening. After a few minutes of troubleshooting, I tried using a different USB port, and suddenly there was no stuttering! I later found out that my computer motherboard has both USB 3.1 Gen 1 and Gen 2 ports. Only the blue Gen 1 ports produced consistent clean audio, and the two Gen 2 ports (which happen to be colored red on my motherboard) both resulted in occasional stuttering audio every 15 or so seconds. This is very strange behavior, and I can't for the life of me understand why it's happening. I wonder if somehow my motherboard's USB ports have issues with some USB 2.0 devices, even after I recently updated the bios. Have any of you experienced this issue with USB 3.1 Gen 2 ports? Even though I solved the issue, I'm still curious. I'm using an ASUS ROG Strix X470-F Gaming Motherboard.
Edit: Now that I think about it, I remember a similar stuttering issue happening with the Khadas Tone Board with this same motherboard, which is also USB 2.0.