• 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

JIW

Senior Member
Joined
Nov 11, 2019
Messages
378
Likes
551
Location
Germany
@ShiZo did you disable 'De-Emphasis' (or set it to 'auto')?

This is what it does if enabled:
iu

Source (dead): https://www.rme-audio.de/images/techinfo/emphasis.gif
Emphasis-Pre-De.gif

Source: http://www.sengpielaudio.com/calculator-timeconstant.htm

According to the manual for the ADI-2 Pro FS on p. 72:
The ADI-2 Pro can also perform both pre- and de-emphasis outside the DAC with just a single band of its Parametric EQ. The emphasis filter is based on a simple first order RC filter with time constants of 50 μs and 15 μs. The frequency response curve looks like a low-Q treble boost with its +3 dB point at 3183 Hz, and the upper shelving point at 10610 Hz. At 20 kHz gain hits +9.49 dB.

For an inverted filter curve select band 5 with type shelf active, set Q to 0.5, Frequency to 5.2 kHz and Gain to -9.5 dB. Similarly, a pre-emphasis is done with the same settings but Gain to +9.5 dB.
 

RobertSc

Member
Joined
Oct 16, 2019
Messages
9
Likes
2
Had many, many DAC’s over the years. The RME ADI-2 DAC is by far the best: Transparent, High quality, Excellent support, Feature rich aso. However it’s not for everyone. Mastering the user interface can at times be a bit frustrating. Once done it’s a highly rewarding piece of equipment!

I'd agree.. The interface is irritating but you get used to it once you understand the way to do things. I have the new version RME ADI-2 dac (with new AK Dac chip) and after about 2 weeks burn-in it's sounding superb. Detailed, revealing, powerful DSP when you need it. I think paring will make a big difference. I have it running through the new SMSL SP200 THX amp and to an Audiolab 8200p power amp when I want to listen through speakers.
 

RobertSc

Member
Joined
Oct 16, 2019
Messages
9
Likes
2
What is the output of RME going to? An amp? Internal headphone amp? If external device, what did you set the RME level to?

I run the RME through the thx sp200 first (and outputting to the Audiolab power amp). Just because it's convenient to have 2 sources and better volume control. I set the sp200 to low-gain at 10/11 o'clock volume and use the remote for the RME output control. This goes to about -10 to 14db (without autolevels and loudness off) before hitting headroom on both hd660s and speakers - which seems to be perfect. No distortion on the RME levels.
 

ElNino

Addicted to Fun and Learning
Joined
Sep 26, 2019
Messages
557
Likes
722
Does anyone understand how the ADI-2 handles EQ headroom? The manual states:
The ADI-2 DAC has an internal headroom of 24 dB. Extreme boosts with overlapping filters could cause an internal overload. Such an overload will be visible as it is displayed by the level meter below the EQ, as well as the channel’s level meter. Reducing the output volume will prevent any clipping as long as the headroom of 24 dB is not exceeded.

I can't make heads or tails of this. Surely they're not reducing the gain digitally by 24dB to allow 24dB of EQ boost. (They don't seem to be, given the SOTA measurements.) Do they adjust digital gain by -24dB as soon as you switch on EQ? Do they adjust digital gain by the amount of the boost and then cap it at a maximum reduction of -24dB?

The manual is so good otherwise (exemplary, really), but incomprehensible on this point.
 

RobertSc

Member
Joined
Oct 16, 2019
Messages
9
Likes
2
Hello,

First, I would like to thank you all for this indeed interesting, helpful, most competent Website/Forum/Thread.
I am following it since a few weeks trying to make up my mind on a new decent DAC to play with my new tube AMP (Line Magnetic 845) on my PIEGA speakers from Audirvana on a Mac mini.
It should replace my older Micromega and IFI IDSD MICRO - both tested here.

So, it became the RME ADI 2 DAC V.2 - which BTW just arrived today (BTW after 5 days transportation for 300 KM!!) and is already playing fine.
Don't expect any technical/mesurement background from me, besides these graphs look very impressive.
Thanks for these reliable technical and measurable arguments and fact checks here.

However, I could give you some first totally subjective listening impressions first hand.
The (new) RME is currently playing as pre-amp into the Line Magnetic as power amp.
First you need to know, it is not transparent at all, at least not the box, my wife noticed it right away.
Sound-wise, the clarity and rapidity is indeed very impressive, transparent as the king new clothes.
Voices/instruments (piano/violin/cimbals/trumpets/organs...) became so much more credible.

I also had wabbling bass problem previously, which astonishing enough seems to have totally disappeared without even any EQ yet...
It is a pleasure to listen, at least with most of the records.

I was worried first about loosing any sound quality while playing MQA files from Tidal.
I am not missing anything at all.

So where is the downside?
Someone, reported some harshness before with a specific title.
I could add an other track here. Listening to the "Gardian of the Galaxy" becomes such a pain and I agree it must be the recording.
And no, less treble did not help it.
I am just discovering, so will have to find out set up/filters/EQ positions to help with such kind older rock recordings. Any suggestion are welcome.
Otherwhise, with any decent recording the RME is such a refreshing pleasure to listen to.

At this stage and after 3 hours listening, I don't regret the buy at all.

Cheers,

Indeed, and it will sound amazing after 2 weeks burn-in. :))
 
  • Like
Reactions: CMB

CMB

Active Member
Joined
Nov 26, 2019
Messages
260
Likes
510
Interesting that you mention a burning in time. At risk that people here will threw stones at me, especially as RME said that there is NO burning in time, I thought same thing today. I have the v2 since one week now. First day the sound were, lets say very « cristalline » and I escaped in NOS filter, upsampling, high cut, heavy EQ. Day after day and step by step, I had to come back from these settings and lowered EQ settings, cancelled the high cut, even changed filter to sharp or SD LD. I could not be happier of it sound, which impresses me daily more. Something here is definately adapting over time.

The other good thing from this, is that I know the interface by heart now
 

VintageFlanker

Major Contributor
Forum Donor
Joined
Sep 20, 2018
Messages
4,939
Likes
19,689
Location
Paris
First day the sound were, lets say very « cristalline » and I escaped in NOS filter, upsampling, high cut, heavy EQ. Day after day and step by step, I had to come back from these settings and lowered EQ settings, cancelled the high cut, even changed filter to sharp or SD LD. I could not be happier of it sound, which impresses me daily more. Something here is definately adapting over time.
No doubt about that. Something being your mind. That's how perceived burn-in works most of the time. The gear doesn't change one bit. But your brain simply get used to it!;)
 

JIW

Senior Member
Joined
Nov 11, 2019
Messages
378
Likes
551
Location
Germany
Does anyone understand how the ADI-2 handles EQ headroom? The manual states:


I can't make heads or tails of this. Surely they're not reducing the gain digitally by 24dB to allow 24dB of EQ boost. (They don't seem to be, given the SOTA measurements.) Do they adjust digital gain by -24dB as soon as you switch on EQ? Do they adjust digital gain by the amount of the boost and then cap it at a maximum reduction of -24dB?

The manual is so good otherwise (exemplary, really), but incomprehensible on this point.

No, the EQ'ed (DSP'ed) signal still has to satisfy being at most 0 dB FS at the DA-chip, i.e. for a 0 dB FS signal at the most boosted frequency, the volume needs to be lowered at least by the amount of the boost, e.g. if 100 Hz is the most boosted frequency with 12 dB, the volume would need to be set to at least -12 dB to prevent clipping the DAC (internal overload) when playing a 100 Hz signal at 0 dB FS. If the DAC is clipped, the post-FX/post-DSP meters will indicate this by showing 'OVR' in red.
 

ElNino

Addicted to Fun and Learning
Joined
Sep 26, 2019
Messages
557
Likes
722
No, the EQ'ed (DSP'ed) signal still has to satisfy being at most 0 dB FS at the DA-chip, i.e. for a 0 dB FS signal at the most boosted frequency, the volume needs to be lowered at least by the amount of the boost, e.g. if 100 Hz is the most boosted frequency with 12 dB, the volume would need to be set to at least -12 dB to prevent clipping the DAC (internal overload) when playing a 100 Hz signal at 0 dB FS. If the DAC is clipped, the post-FX/post-DSP meters will indicate this by showing 'OVR' in red.

Thanks -- that's what I'd expect would be the case, but that also directly contradicts what the manual says.
 

JIW

Senior Member
Joined
Nov 11, 2019
Messages
378
Likes
551
Location
Germany
Thanks -- that's what I'd expect would be the case, but that also directly contradicts what the manual says.

No, it does not. The DSP can use a different format as the in- and out-going signal.

If it uses 32 bits, the noise floor of which is about -193 dB FS, a 24 dB attenuation would still leave SNR at about 179 dB which is almost 60 dB below the SNR of the DAC causing only about a 0.00006 dB increase in noise compared to no 24 dB attenuation.
 

ElNino

Addicted to Fun and Learning
Joined
Sep 26, 2019
Messages
557
Likes
722
No, it does not. The DSP can use a different format as the in- and out-going signal.

If it uses 32 bits, the noise floor of which is about -193 dB FS, a 24 dB attenuation would still leave SNR at about 179 dB which is almost 60 dB below the SNR of the DAC causing only about a 0.00006 dB increase in noise compared to no 24 dB attenuation.

Sorry, I'm not following -- can you rephrase what you're trying to say and how it relates to your comment above that "No, the EQ'ed (DSP'ed) signal still has to satisfy being at most 0 dB FS at the DA-chip, i.e. for a 0 dB FS signal at the most boosted frequency, the volume needs to be lowered at least by the amount of the boost...."

My current working theory is that the manual is just incorrect.
 

JIW

Senior Member
Joined
Nov 11, 2019
Messages
378
Likes
551
Location
Germany
Sorry, I'm not following -- can you rephrase what you're trying to say and how it relates to your comment above that "No, the EQ'ed (DSP'ed) signal still has to satisfy being at most 0 dB FS at the DA-chip, i.e. for a 0 dB FS signal at the most boosted frequency, the volume needs to be lowered at least by the amount of the boost...."

My current working theory is that the manual is just incorrect.

The signal is lowered by 24 dB in the beginning of the DSP-chain, i.e. 0 dB FS is lowered to -24 dB FS. At the end of the DSP-chain the signal is increased by 24 dB, i.e. -24 dB FS is increased to 0 dB FS. Further, the way I read the manual, the volume control seems to be last in processing line even after the level increase.

Thus if after processing but before the level increase the signal exceeds -24 dB FS, the following 24 dB increase would lead to a signal exceeding 0 dB FS and thus overload the DAC if it is not lowered sufficiently using the volume control such that it does not exceed 0 dB FS after the DSP.
However, if the signal after processing but before the level increase exceeds 0 dB FS, it will overload the DSP and reducing the volume cannot remedy this.

In order to not degrade the signal by lowering the level, more bits may be added such that e.g. a 16 or 24 bit signal becomes a 32 bit signal. In the manual on p. 64 it is mentioned that the volume control uses 42 bits.
 

ShiZo

Addicted to Fun and Learning
Forum Donor
Joined
Sep 7, 2018
Messages
835
Likes
555
I still don't know >.<


It de-emphasis better left off? I thought it fixes certain music files that have emphasis added to it. @amirm, do you know anything about the emphasis setting? Sorry for being dense :(.
 

Bliman

Senior Member
Joined
Nov 9, 2018
Messages
419
Likes
150
Location
Belgium
One thing that is for sure is that the EQ feature is a godsend.
I am getting help from the RME forum and my sound is getting better and better by the measurements with the microphone and REW.
In my eyes it is a must-have. And more essential than a little bit better specs. And also important is someone who can help with these curves and what to change in your room.
The RME gives everything anyone wants. And I cannot recommend it highly enough.
I say buy the RME, buy a umik microphone (or another) and get some help from some forums and you are well on your way to audio greatness.
 

LTig

Master Contributor
Forum Donor
Joined
Feb 27, 2019
Messages
5,760
Likes
9,442
Location
Europe
Interesting that you mention a burning in time. At risk that people here will threw stones at me, especially as RME said that there is NO burning in time, I thought same thing today. I have the v2 since one week now. First day the sound were, lets say very « cristalline » and I escaped in NOS filter, upsampling, high cut, heavy EQ. Day after day and step by step, I had to come back from these settings and lowered EQ settings, cancelled the high cut, even changed filter to sharp or SD LD. I could not be happier of it sound, which impresses me daily more. Something here is definately adapting over time.
That "something" is you. No need to throw stones yet;)
 

ShiZo

Addicted to Fun and Learning
Forum Donor
Joined
Sep 7, 2018
Messages
835
Likes
555
The interface is one of the best things about this device. No stone is left unturned.

Once you get the hang of the interface all others seem inadequate.
 

MRC01

Major Contributor
Joined
Feb 5, 2019
Messages
3,422
Likes
4,029
Location
Pacific Northwest
I still don't know >.<


It de-emphasis better left off? I thought it fixes certain music files that have emphasis added to it. @amirm, do you know anything about the emphasis setting? Sorry for being dense :(.
Emphasis is a little-known aspect of the Redbook CD standard. It was rarely used back in the day and never used anymore. It was like Dolby B for digital audio: boost the treble before encoding to digital (to improve digital resolution of the treble), then attenuate it on playback.

The CD standard has a data bit flag that recordings using emphasis are supposed to set, to tell the DAC to enable it automatically. But sometimes this isn't always set, or the DAC doesn't always recognize it. So it's useful to be able to apply it manually. However, recordings using this are so rare you might never encounter one. You can leave it off and forget about it. The effect is not subtle. If you do have a recording that uses this, it will sound so artificially bright & nasty (when uncompensated) you will know without a doubt.
 
Top Bottom