Audio measurement gear

[Rja4000]

By the way, I think that the DAC is the limiting factor in the loop, as far as low-level IMD is concerned.
Or, at least, both DAC and ADC are in the same league.

I tried another ADC, using a pretty performant Focusrite Preamp I own, so I could boost the analog signal before ADC conversion:
(Note that this measurement is not in "Phone Balanced mode" output, but in normal Balanced output.)
The Focusrite performs a bit better, but less than if the ADC was the limiting factor, in my opinion.

A -60dBFS signal is sent to RME's XLR analog outputs.
Focusrite's MIC input is fed from one of them, while the RME Analog in is fed by the other.

The RME in loopback (blue) is left input, return level is @-60dBFS
While the Focusrite (red) is right input, return level @ 0dBFS, after an analog boost of +60dB.
The Focusrite ADC is used for the right channel and forwarded back to the RME via AES/EBU.

The noise on the red channel is obviously the noise of the RME DAC, boosted by +60dB.

Impressive.

 

[SIY]

Here is Amir's result for the ADI-2 DAC (copied from his review)
Thats -51dB or so for the same -60dBFS.
How I could I measure 15dB better ???

First two things to check (assuming you've run calibration)- are you using the same test signal (especially the ratio) and the same reference for the dB calculation?

 

[Rja4000]

First two things to check (assuming you've run calibration)- are you using the same test signal (especially the ratio) and the same reference for the dB calculation?

I didn't run calibration. How do you do it with Virtins MI Pro?

 

[wynpalmer]

I agree.

Just gave Virtins Multi Instrument a try:
View attachment 30286
OK, -100dB doesn't make sense...

But that's still -66.8dB for IMD SMPTE @ -60dBFS
That's coherent with REW above.

And that's what I don't understand:

Here is Amir's result for the ADI-2 DAC (copied from his review)
Thats -51dB or so for the same -60dBFS.
How I could I measure 15dB better ???

View attachment 30287

I also tried Virtins. I couldn't see any significant difference from the REW result in the single test that I did.
I've continued to refine my test environment as I mostly operate single ended.
I added an isolation transformer to break the RME ground loop.

The measurement was performed on the same home made MC preamp with 63dB of gain, but with -0.9dBv of output.
The 60Hz level I checked using other spectrum analysis programs and a stand alone analyzer that I borrowed and the result is consistent.
I'll attempt to measure the RME using the same tests as the original evaluation and see what I get.

 

[Rja4000]

First two things to check (assuming you've run calibration)- are you using the same test signal (especially the ratio) and the same reference for the dB calculation?

As for ratio, that's standard SMPTE 12dB less for 7kHz (1:4) and reference is the 7kHz signal (Well, I guess so. That's pretty much clear in SMPTE IMD measurement definition)

 

[SIY]

As for ratio, that's standard SMPTE 12dB less for 7kHz (1:4) and reference is the 7kHz signal (Well, I guess so. That's pretty much clear in SMPTE IMD measurement definition)

The APx analyzers allow SMPTE 1:1, SMPTE 4:1, and SMPTE 10:1, so it's worth checking.

 

[Rja4000]

The APx analyzers allow SMPTE 1:1, SMPTE 4:1, and SMPTE 10:1, so it's worth checking.

I use the RME ADI-2 Pro fs, and that's 4:1 on my side.
@amirm
What are you using?

 

[Blumlein 88]

I use the RME ADI-2 Pro fs, and that's 4:1 on my side.
@amirm
What are you using?

On the graphs in his DAC review, you'll see along the Y axis SMPTE/DIN ratio (db).

I'm pretty sure it is 4:1 ratio.

Here is a page from AP. It indicates the ratio can be varied, but I believe 4:1 is being used by Amir.

https://www.ap.com/technical-library/more-about-imd/

SMPTE (DIN)
SMPTE IMD is a technique for measuring IMD (intermodulation distortion) according to the SMPTE RP120-1983 standard. The DIN intermodulation distortion technique uses a similar method.

The stimulus is a strong low-frequency interfering signal (f1) combined with a weaker high frequency signal of interest (f2). f1 is usually 60 Hz and f2 is usually 7 kHz, at a ratio of f1_f2=4:1. The stimulus signal is the sum of the two sine waves. In a distorting DUT, this stimulus results in an AM (amplitude modulated) waveform, with f2 as the “carrier” and f1 as the modulation.

In analysis, f1 is removed, and the residual is bandpass filtered and then demodulated to reveal the AM modulation products. The rms level of the modulation products is measured and expressed as a ratio to the rms level of f2. The SMPTE IMD measurement includes noise within the passband, and is insensitive to FM (frequency modulation) distortion.

The APx500 implementation of SMPTE IMD provides the capability to vary the stimulus frequencies and to choose a 1:1 stimulus ratio.

 

[March Audio]

What is the bandwidth that Amir's test is run over compared to other people's tests?

 

[wynpalmer]

By the way, I think that the DAC is the limiting factor in the loop, as far as low-level IMD is concerned.
Or, at least, both DAC and ADC are in the same league.

I tried another ADC, using a pretty performant Focusrite Preamp I own, so I could boost the analog signal before ADC conversion:
(Note that this measurement is not in "Phone Balanced mode" output, but in normal Balanced output.)
The Focusrite performs a bit better, but less than if the ADC was the limiting factor, in my opinion.

A -60dBFS signal is sent to RME's XLR analog outputs.
Focusrite's MIC input is fed from one of them, while the RME Analog in is fed by the other.

The RME in loopback (blue) is left input, return level is @-60dBFS
While the Focusrite (red) is right input, return level @ 0dBFS, after an analog boost of +60dB.
The Focusrite ADC is used for the right channel and forwarded back to the RME via AES/EBU.

The noise on the red channel is obviously the noise of the RME DAC, boosted by +60dB.

Impressive.

View attachment 30288

For what it's worth...
From the ADI-2 PRO FS manual, page 80, "total harmonic distortion measurements", RME claim that the ADC is. c. 3dB better than the DAC with a SINAD of 113.1 for the ADC and 110.1 for the DAC, measured at 1kHz, -1dBFS. I don't see any IMD measurements, but it seems reasonable to assume that the DAC is indeed inferior to the ADC.

 

[KSTR]

My findings with Adi-2 Pro FS also were that the DAC distorts more than the ADC when running the same level. Once the ADC runs some 10dB lower than the DAC its contribution is minimal, looking at the changes of harmonics wrt level and phase (the phase option of REW is really cool).

 

[Rja4000]

For what it's worth...
From the ADI-2 PRO FS manual, page 80, "total harmonic distortion measurements", RME claim that the ADC is. c. 3dB better than the DAC with a SINAD of 113.1 for the ADC and 110.1 for the DAC, measured at 1kHz, -1dBFS. I don't see any IMD measurements, but it seems reasonable to assume that the DAC is indeed inferior to the ADC.

I don't have that in my PDF version of the manual (User’s Guide ADI-2 Pro FS – v 2.3)
Where did you get yours ?

 

[KSTR]

^ same question here. In the plots on that page (in versions 2.3 and 2.4) the distortion is actually higher for the ADC, and note that the DAC plot is an almost two years older measurement of the non-FS version (which should have no effect, though). And of course we don't know which analyzer was used other than that it must have been a normal soundcard working with the HpW software, and apparantly without notch filter.
EDIT: found it, the 3dB SINAD difference appears to come from the values on the specs page (pg.64), looking at the THD @ -1dBFS values and S/N ratios, obtaining a SINAD number).

 

[Schackmannen]

Amir has measured both the ADC and DAC part of the ADI-2 Pro (although non FS-version) and the ADC part perfomed a lot better than the DAC:

If the DAC part of the ADI-2 Pro FS is the same as the ADI-2 DAC (which i believe it is?) then the differences are smaller but still noticable.

 

[wynpalmer]

I don't have that in my PDF version of the manual (User’s Guide ADI-2 Pro FS – v 2.3)
Where did you get yours ?

Version 2.0 hard copy of the users guide for the PRO FS version,. Both ADC and DAC graphs are actually labelled for the ADI-2 PRO.
This is slightly better than the page 64 specs.
I pulled up the plots on version 2.4 of the users guide.
It looks like the third harmonic has degraded by about 4dB compared to the original ADC plot, which results in a significantly degraded SINAD as that harmonic dominates the THD number.
The DAC plot is still the original ADI-2 PRO version that was present on version 2.0.
Should I be demanding my money back and exchanging the fs for the original?

 

[wynpalmer]

Amir has measured both the ADC and DAC part of the ADI-2 Pro (although non FS-version) and the ADC part perfomed a lot better than the DAC:
View attachment 30409View attachment 30414
If the DAC part of the ADI-2 Pro FS is the same as the ADI-2 DAC (which i believe it is?) then the differences are smaller but still noticable.

RME's own measurements have the ADC performing several dBs worse and the DAC several dBs better.
How curious...

 

[Rja4000]

Amir has measured both the ADC and DAC part of the ADI-2 Pro (although non FS-version) and the ADC part perfomed a lot better than the DAC:
(...)
If the DAC part of the ADI-2 Pro FS is the same as the ADI-2 DAC (which i believe it is?) then the differences are smaller but still noticable.

Awesome ! I missed that comparison.

As I understodd, the Pro Fs and ADI-2 DAC use the same chip AK4490, but the pro has 2 of those.
In "Phone balanced, both are used, one for left channel and one for right channel.
This should improve the SNR by 3dB, in theory.

 

[Schackmannen]

RME's own measurements have the ADC performing several dBs worse and the DAC several dBs better.
How curious...

Are you sure? I had a quick look at their manual and they spec the ADC to -113 THD+N @ -1 dBFS and the DAC to -108 THD+N @ -1 dBFS.

Edit: Looks like you're right, their measurements in the 2016 manual shows the numbers i found but in the current manual on their website it shows the ADC and DAC being very similar with both having a THD+N at around -111 dB @ -1 dBFS. Not sure why they would improve the performance of the DAC but make the ADC part worse?

 

[KSTR]

RME's own measurements have the ADC performing several dBs worse and the DAC several dBs better.
How curious...

As I wrote, it's not clear how they measured it (that is, with what gear). Could be a Münchhausen type of measuring itself...

EDIT: a possible way to measure it halfway correctly by itself could be (THD sans noise):
- ADC : Feed -20dBFS to DAC at +24dBu and set ADC to +4dBu, so ADC distortion will dominate (but DAC noise will dominate, too)
- DAC : Feed 0dBFS to DAC at +4dBu and set ADC to +24dBu, so DAC distortion will dominate (but ADC noise will dominate, too)

 

[Rja4000]

Are you sure? I had a quick look at their manual and they spec the ADC to -113 THD+N @ -1 dBFS and the DAC to -108 THD+N @ -1 dBFS.

Copied from the RME ADI-2 Pro Fs Manual:
"User’s Guide ADI-2 Pro FS – v 2.3", page 65

33.1 Analog Inputs
XLR
• THD @ -1 dBFS: -113 dB, 0.00022 %
• THD+N @ -1 dBFS: -110.6 dB, 0.00029 %

33.2 Analog Outputs
1/2 XLR
• THD @ -1 dBFS: -112 dB, 0.00025 %
• THD+N @ -1 dBFS: -110 dB, 0.00032 %
• THD @ -3 dBFS: -116 dB, 0.00016 %