1 (edited by ddps 2020-03-05 00:30:33)

Topic: Intermodulation distortion / Xiph.org

This has long been an interesting page to me - https://people.xiph.org/~xiphmont/demo/neil-young.html

I now have 3 DACs available in my system: An Auralic Altair (original); a McIntosh MA8900 with the DA2 module; and the new version of the RME ADI-2 DAC.

I don't make much use of the Altair's DAC (I use it mostly as a streamer), but I use the other two as I see fit.

If you download the "Intermod Test" files at the URL above, I find it curious that the ADI-2 exhibits a higher degree of audible distortion than the internal McIntosh DA2 module (which, BTW, employs an ESS Sabre 9028 Pro). I'm surprised at this. The ADI-2 is otherwise a much preferable-sounding product to me. Does @MC have any idea of why these files present an audible challenge to the ADI-2? I have level matched the inputs to a reasonable degree, and the ADI-2's IM warbling is more noticeable in a way that is out of proportion to its core signal.

(Oh, FWIW, the choice of impulse filter doesn't have a large effect in this comparison.)

Thank you!

2

Re: Intermodulation distortion / Xiph.org

Set the ADI's volume lower than 0 dB.

Regards
Matthias Carstens
RME

Re: Intermodulation distortion / Xiph.org

MC wrote:

Set the ADI's volume lower than 0 dB.

Yes, thanks, but it's set to -1 already.

4 (edited by KaiS 2020-03-05 07:29:41)

Re: Intermodulation distortion / Xiph.org

When I play these tests through my ADI-2 Pro on a Sennheiser HD6XX, it stays dead quiet;
https://people.xiph.org/~xiphmont/demo/neil-young.html

Even if I set the "(original version of above clip) (16 bit )" really loud (ADI vol: -20dBr), all the test files show up on the level meter, but not the faintest bit of the mentioned IM distortions is audible with the 4 test clips, nothing.

This would mean no problem with RME ADI-2 Pro itself.


But maybe if you do feed the tests into another headphones amplifier.
I tried:

- Schiit Magni III - dead quiet, nothing.
Didn't test up to clipping level as I don't want to break the headphones, but went really loud on the HD6XX.

- STAX SRM-1 MK2 Pro (transistor) / SR-009: dead quiet up to clipping level with test 1, 2, 4. Test 3/192kHz 2dB below clipping some audible output.

- STAX SRM-T1 (tube) / SR 009: dead quiet up to clipping level with test 1, 2, 4. Test 3/192kHz 3dB below clipping some audible output.

- SE OTL Tube Amp, only with the 3rd, 192kHz test there is an audible result, but interestingly not with the other tests, although this amp has measurable distortions, it's built to have them.
Again couldn't go full up, but far louder than I would ever listen even short term.


Conclusion:
For RME ADI-2 Pro the problem doesn't exist.
Other amps can have an issue.

5

Re: Intermodulation distortion / Xiph.org

Correct. I use these test files from time to time since years, and never had a problem unless clipping occurs. Clipping requires EQ, Treble  gain or volume settings higher than 0 dB. Otherwise the output is dead silent despite the level meters at near full blast (just checked with a DAC).

Regards
Matthias Carstens
RME

6 (edited by ddps 2020-03-05 14:11:26)

Re: Intermodulation distortion / Xiph.org

MC wrote:

Correct. I use these test files from time to time since years, and never had a problem unless clipping occurs. Clipping requires EQ, Treble  gain or volume settings higher than 0 dB. Otherwise the output is dead silent despite the level meters at near full blast (just checked with a DAC).

Thank you - my question, then, is: how am I supposed to reconcile that I can hear these? Does it mean that possibly something is wrong with my ADI-2? I can hear them through the ADI-2's headphone amp as well, although the HF response of my phones isn't the same as my speakers, so the output is diminished somewhat in comparison.

FWIW, I have verified the signal chain to the DAC is good through the use of RME's test files. And, I have not adjusted the EQ at all on any of the inputs.

Thanks again.

7

Re: Intermodulation distortion / Xiph.org

Maybe your playback system is at fault. These files have invalid headers. Are you sure they are played at 96 kHz and 192 kHz respectively? You used the 16 and 24 bit test files at 96 and 192 kHz with the same player? It is also astonishing that your other DAC with ES9028Pro has the same problem - that points to a different cause than the DACs.

Regards
Matthias Carstens
RME

8 (edited by ddps 2020-03-05 15:01:37)

Re: Intermodulation distortion / Xiph.org

MC wrote:

Maybe your playback system is at fault. These files have invalid headers. Are you sure they are played at 96 kHz and 192 kHz respectively? You used the 16 and 24 bit test files at 96 and 192 kHz with the same player? It is also astonishing that your other DAC with ES9028Pro has the same problem - that points to a different cause than the DACs.

The 9028 really is comparatively soft. I have to push volume up higher to hear the files with that one.

Yes I have verified the playback rate.

Again I can hear the issues in the RME directly through the headphone amp. And the signal chain has been verified using RME’s test files.

9 (edited by ddps 2020-03-05 21:49:52)

Re: Intermodulation distortion / Xiph.org

Also FWIW 20+ years ago I was diagnosed with a hearing defect where I could hear high frequencies abnormally well. But I don’t presume to have that acuity anymore, and the intermodulation distortion would be in the normal human audible range (which it appears to be). But I’m just offering my hearing as a possibility. However, I don't think that my hearing is what is causing this to be audible, if you know what I mean....

Re: Intermodulation distortion / Xiph.org

Well, I tested this with a friend, who could not hear what I was hearing. It is possible that this is just my hearing after all. This is puzzling to me, though.

11 (edited by KaiS 2020-03-06 08:34:09)

Re: Intermodulation distortion / Xiph.org

The distortions mentioned on that site are well within the usual audio band.
They are easily audible without bat ears, in case they occur in a system.
This is what this test is about, audible distortions caused by inaudible frequencies.


Maybe you can distinguish by yourself in which frequency range you perceive them, within or above the normal range covered by music?

Personally I've never had the sensation of intermodulation distortions created within my own ear - although I am quite often using frequency combinations that would trigger them, like 19+20kHz, a standard test I do.

Re: Intermodulation distortion / Xiph.org

I would characterize what I am hearing as very high frequencies, outside the music band, as it were.

13 (edited by KaiS 2020-03-06 12:43:29)

Re: Intermodulation distortion / Xiph.org

Very interesting!

What about single sine waves beyond 20kHz:
Do you hear them?

14 (edited by Curt962 2020-03-06 13:11:02)

Re: Intermodulation distortion / Xiph.org

ddps wrote:

I would characterize what I am hearing as very high frequencies, outside the music band, as it were.

dd...

I would characterize this as a debilitating Medical Condition.   Hyperacusis is the Term.   See a Hearing Professional at once.

Good Luck

Curt

Vintage 2018 ADI-2 DAC. "Classic Edition"
Cables:  Red Ones, and White Ones.
Speakers:  Yes (2)

15 (edited by ddps 2020-03-06 14:01:25)

Re: Intermodulation distortion / Xiph.org

Curt962 wrote:
ddps wrote:

I would characterize what I am hearing as very high frequencies, outside the music band, as it were.

dd...

I would characterize this as a debilitating Medical Condition.   Hyperacusis is the Term.   See a Hearing Professional at once.

Good Luck

Curt

:-D

Well, I've been seen by an audiologist for it, a long time ago. She was not concerned.

I'm not at home right now to verify, but I think it's worth noting that, at least in the first two tests on the Xiph.org page, there is high frequency content in the spectrum analyzer display of the ADI-2, which I believe covers only the audible range. It's just in the very high frequency end of things that there is content....I should have mentioned that earlier.

16 (edited by KaiS 2020-03-10 21:16:39)

Re: Intermodulation distortion / Xiph.org

The ADI-2's Analyser's filters do not separate very sharp.

E.g a fullscale 23kHz sinewave still shows up in the 6.3k band, 3 time below it's frequency.
In reality no signal exists in the 6.3kHz band in the original testsignals.

This is a limitation of RME's bi-quad filter design.
The advantage of the bi-quad filters, opposed to an FFT based solution: the bi-quads work in real "Realtime", so are better suited to analyse dynamic signals like music.


Therefore with this special test, activity in the Analyser does not mean there is real signal in the audio range.


The whole testsignal is outside the audio band, that's what this test is all about:
Showing how inaudibly high frequencies can cause audible distortions if a device is not prepared to handle them - or to confirm an audio chain is capable to do so.


Signals that are audible for normals like me, are created only if, e.g. a headphones amplifier has nonlinear distortions.
But these wouldn't show up in the Analyser, as this happens later in the signal chain where the Analyser doesn't "see" it.

Re: Intermodulation distortion / Xiph.org

Viele Dank Kai.  +1

Vintage 2018 ADI-2 DAC. "Classic Edition"
Cables:  Red Ones, and White Ones.
Speakers:  Yes (2)

Re: Intermodulation distortion / Xiph.org

.. is this by any chance that auto DSD-mode switching "perfect storm" from long ago that was triggered by specific test-files? I can't seem to find the thread, but i remember i ran into the issue myself, while testing out intermodulation in a dac shootout ..

Producer / Engineer / Composer @ FishCorp
RayDAT > 2 x ADI-8 QS | AES+SPDIF > ADI-2 Pro

Re: Intermodulation distortion / Xiph.org

Well, I'm doing something wrong, because I can definitely hear the noises (a whistle on one track and clicks and pops on two others) using the ADI-2 PRO fs - settings:
96 KHz
Volume -20.0 on a reference level for Main Output of +13dBu
No EQ
No loudness
Played on a Mac with QuickTime and also with Audirvana, locked to 96Khz, no upsampling. Audirvana was set to 0 dB and also -1 dB- no difference.
So, if the ADI-2 isn't at fault, where is the clipping occurring (as per MC's post) which is making this intermodulation easily audible in 3 of the 4 files, in QT and Audirvana?

Dominic

20 (edited by Basken 2020-03-06 21:28:20)

Re: Intermodulation distortion / Xiph.org

found it .. have a look:

https://www.forum.rme-audio.de/viewtopic.php?id=26651

definitely a long-shot, but might be worth your time ..

Producer / Engineer / Composer @ FishCorp
RayDAT > 2 x ADI-8 QS | AES+SPDIF > ADI-2 Pro

21 (edited by ddps 2020-03-07 02:56:09)

Re: Intermodulation distortion / Xiph.org

Basken wrote:

found it .. have a look:

https://www.forum.rme-audio.de/viewtopic.php?id=26651

definitely a long-shot, but might be worth your time ..

Interesting. I will try it out as soon as I can - this weekend at the latest! Thanks!

Update - tried it; no difference.

22

Re: Intermodulation distortion / Xiph.org

Basken: good find! Indeed I missed that having DSD Detection On will cause small clicks and cracks on the two warble files. DSD detection happens so quick and short that the Status Overview does not indicate DSD. But the clicks (any noise) are gone immediately when DSD Detection is set to off.

Regards
Matthias Carstens
RME

Re: Intermodulation distortion / Xiph.org

As I mentioned above, that setting change didn't work for me.

Here's what the files sound like as recorded by my iPhone with its mike right next to my tweeter, volume on my amp at 68/100:

https://www.dropbox.com/s/dxdqx3o9w4lel … d.m4a?dl=0

In real life, the sound is not as loud as this very near-field recording would indicate, but at least it's representative of what I am hearing.

The sound I get from the headphones connected directly to the ADI-2 is identical.

24

Re: Intermodulation distortion / Xiph.org

In your above posts you wanted to make us think you have super-bat ears, even your friend couldn't hear the 'super-high' frequencies. And now you present us something that everyone can hear, audible in the medium frequency range, typical for a not proper playback chain with these files.

If you want more help you need to list your playback chain IN ALL DETAILS.

Regards
Matthias Carstens
RME

25

Re: Intermodulation distortion / Xiph.org

I used the analog input of the ADI-2 Pro to document the real output signal at the Pro and DAC with the four xiph.org test files. There is nothing in the audible band. This changes when turning Volume up to +1 dB, then artefacts of overload and aliasing become audible and would show up in the range below 20 kHz.

http://archiv.rme-audio.de/images/xiph_song_shift.png

http://archiv.rme-audio.de/images/xiph_warbles_96khz.png

http://archiv.rme-audio.de/images/xiph_warbles_192khz.png

http://archiv.rme-audio.de/images/xiph30_33khz.png

Regards
Matthias Carstens
RME

26 (edited by DominicPerry 2020-03-08 12:54:58)

Re: Intermodulation distortion / Xiph.org

Well, no one responded to my post - I'm happy to accept that I'm doing something wrong, but I don't know what it is.
When I play back these four files at 96KHz (I know one is meant to be played at 192, but I didn't), I can hear a high pitched whistle in the 30&33 file, a tune in the Pido_O_trollbat file, and some high pitched warbling (like Long-Wave radio) in the warble files.
Set up is:
MacOS 10.15.3 on MacBook Pro 15" retina (2015).
I play the files in QuickTime, Audirvana and Presonus Studio One, all with the same results.
The ADI-2 PRO fs is connected to a USB port with no other devices on the port. It is set to 96KHz with AudioMidi setup (96,000 32bit float). Audirvana is set to 96KHz on the Bridge setting (which may have no impact) and no upsampling. StudioOne song is set to 96KHz/32bit
ADI-2 Pro fs has Main Out set to +13 ref level, and volume is set between -20 and -8, with no EQ and no loudness, no bass/treble (these are all turned OFF as well as being set to 0).
DSD detection is OFF
Pulling the volume down below zero in all three applications makes no difference - things get quieter but they are still there.
ADI-2 PRO fs is connected with balanced cables to Nord 1ET400A power amp, connected to Tannoy Eaton Legacy speakers.
What else should I try???

Dominic

27 (edited by ddps 2020-03-08 14:40:26)

Re: Intermodulation distortion / Xiph.org

@MC - I don't think I have bat ears at all (although they certainly may be batty!!!); I didn't even offer my medical defect as a possibility until later on. When i offered this as part of the story, I did so with a dose of incredulity myself. So let's please agree that we should get this out of the picture.

This is why I offered the recording. By the way, even my friend could hear that recording. However, he cannot hear that when putting his ears near the tweeters, played back at the same level that I recorded it at. It is VERY soft.

Beyond that, my playback chain is very simple:

Two different streaming chains available; both start with a Synology NAS, and neither are using methods where the files are being "pushed" upstream, but rather the files are being "pulled" over a wired network connection, directly from the filesystem.

  • The first is: Synology NAS --> gigabit ethernet --> Auralic Altair used as a streamer, with all processing capabilities disabled --> USB --> RME ADI-2 DAC (new edition). I have tested this configuration using the RME Bit Tests, and all tests pass.

  • The second is: Synology NAS --> gigabit ethernet --> Windows 10 PC with Audirvana used as a streamer, with all processing capabilities disabled --> USB --> RME ADI-2 DAC (new edition). I have tested this configuration using the RME Bit Tests, and all tests pass.

  • I am using Balanced outputs from the RME, and for "Line Out" my settings are as follows: Source: Auto; Ref Level +7 dBu; Auto Ref Level OFF; Volume 0.0dB; Lock Volume ON. I have no EQ or loudness or any other effects enabled on the ADI-2. Oh, and as I mentioned earlier, DSD detection on or off doesn't have any meaningful impact on the results for me. The playback sounds pretty much the same either way.

  • In all cases with speakers, playback is through a McIntosh MA8900, to Salk SS 9.5 speakers. Headphone playback is to Sennheiser HD280 Pro connected directly to the RME ADI-2 standard headphone jack.

I have felt since starting this test that there was significance in the fact that all Bit Tests passed (and the Bit Test files are in the same subfolder as the Xiph files, going through the same playback chain), since that should prove that there is true file transfer integrity. I had actually thought that this was the most significant thing of all; hence why I left out the other details until now. @MC, if you can shed light on how the passing of the bit tests might be irrelevant in this scenario, that might shed some light on this for me.

Please remember that using either of these configurations, I can still hear the sounds you hear above in my headphones connected to the RME ADI-2, albeit at a lesser level than my speakers, from which I recorded the example. This is likely because my headphones lack high frequency response compared to my speakers. I am also able to hear artifacts on my speakers - at a notably reduced amplitude in comparison, as previously noted - using the same chains as above, but played back on my McIntosh DA2 module instead of the RME ADI-2 DAC.

I'd like to make one more point here - all of this is very difficult to hear in the real world. Even I could barely hear the artifacts while standing to make my earlier iPhone recording. The iPhone recording only serves to amplify the effect - and even I am surprised by how well the iPhone amplifies the effect! The point of my recording was to "exaggerate for clarity" of this conversation, as it were.

It appears that @DominicPerry is experiencing the same issues.

I hope this helps clarify things.

28

Re: Intermodulation distortion / Xiph.org

Thanks for these details, much helpful and appreciated!

I have done more tests this weekend and can confirm again that the line outputs of the Pro and DAC are free of audible problems with these test files. Testing this is not that easy for you. The test files include an extremely high amount of high-frequency garbage which causes audible artefacts (down-mirroring etc) when played over various electronics. For example my monitor system in the lab completely fails. No matter what output level and what volume setting on the monitors, the warble sound is always there and easy to hear. But at the RME's outputs there is only what I have shown in the measurements above - nothing to listen to.

To verify this is easy for me, not so easy for you: one needs a sharp high-frequency filter (low pass) that only passes the audible audio. I used an ADI-2 Pro in Preamp mode (AD plus DA conversion) in-between the DAC and the monitor system. At 192 kHz sample rate the full signal gets passed through - warble garble. At 48 kHz sample rate: total silence. The AD/DA conversion operates as steep 24 kHz filter, removing all the high frequency junk. If the DAC would cause the warble to become audible that would have been unaffected. As there is no sound anymore it is clear that the issue lies in the input electronics of the monitor system.

I expect the same is happening with your McIntosh, which works much better as my monitor system (according to your description of the sound being very low volume).

The situation at the phones output depends on the impedance of the phones. It is quite easy to run into overload with these test files without noticing it. For example a 16 Ohm phones can only be driven up to about -6 dB volume in High Power mode. Everything above will make the warble become easy to hear. Everything below is very low in volume. With 300 Ohm headphones even 0 dB High Power is free of warble (free in terms of not easily audible, it might be audible for someone with very sensitive ears). But then we are talking about a non-issue, not a design fault or a problem in the DACs and Pros.

Regards
Matthias Carstens
RME

29

Re: Intermodulation distortion / Xiph.org

BTW, the fact that the 96 kHz warble file causes a click near the end of the playback, which can be solved by turning off DSD detection, is now on our list for the next firmware update. As the units do not support DSD at double speed it should not be necessary to switch DSD off.

Regards
Matthias Carstens
RME

30

Re: Intermodulation distortion / Xiph.org

DominicPerry wrote:

I can hear a high pitched whistle in the 30&33 file, a tune in the Pido_O_trollbat file, and some high pitched warbling (like Long-Wave radio) in the warble files.

That should not be audible.

DominicPerry wrote:

ADI-2 PRO fs is connected with balanced cables to Nord 1ET400A power amp, connected to Tannoy Eaton Legacy speakers. What else should I try???

Use the headphone out. It should be clean. Then the Nord is doing the 'demodulation' of high-frequency content into the audible band.

BTW, I used JRiver as player on macOS when testing.

Regards
Matthias Carstens
RME

31 (edited by ddps 2020-03-09 17:28:54)

Re: Intermodulation distortion / Xiph.org

@MC - many thanks. This is helpful and makes sense. One thing that surprises me in my admittedly limited appreciation is why the different filter selections don't seem to change this. Why wouldn't using the AK4493's "Sharp" filter block a reasonable amount of content higher than 22kHz from passing through into the analog realm at all?

And if that's not possible, is it possible to implement a (switchable) filter in the existing architecture that might do that?

Thanks again,

Drew

32

Re: Intermodulation distortion / Xiph.org

You overlooked that you play back a 96 kHz file with 48 kHz bandwidth. So the filters do NOT work at 22 kHz, but 48 kHz, and make zero difference at 22 kHz.

Regards
Matthias Carstens
RME

33 (edited by Basken 2020-03-09 20:16:06)

Re: Intermodulation distortion / Xiph.org

Hi ddps!

I think it's important to keep your "eyes on the prize" so to speak.

The intermodulation test-files are meant to "provoke" your system and coax out issues, and did just that .. however - no valid musical material will contain this kind of information, so in a real world listening-scenario this is a non-issue.

The filters in a DAC has a very specific purpose, which is to limit content above the nyquist-frequency (eg. half the given sample-rate), which can be done in various ways with various consequences. Asking for a filter to limit ultrasonic frequencies from hi-res files (eg. the very purpose of hires) - which even in an ideal reproduction environment cannot be heard (by humans) - kind of defeat the purpose of listening to hires files in the first place  .. so do think about that for a second wink ..

When examining formats such as very high samplerate PCM or DSD, you'll often see an impulse-plot with a near vertical slope - and i guess it's intuition to consider this "better" than the impulse-response of a 44,1k graph .. but - in real life, the tweeter on your speakers cannot move from 0 to 100 in "no time", so this "perfect" impulse could or most likely will result in audible distortion .. if not from the speaker, then from your amp ..

So .. yeah, don't worry bout hires - just enjoy the sound you can hear <3

Best,
Basken

PS: In a production situation (where i have my daily routine), there are somewhat objective reasons for working in a resolution above 48k, but in all honesty it's still very much of a theoretical nature, and i don't want to "pollute" this thread <3 ..

Producer / Engineer / Composer @ FishCorp
RayDAT > 2 x ADI-8 QS | AES+SPDIF > ADI-2 Pro

34

Re: Intermodulation distortion / Xiph.org

ddps wrote:

And if that's not possible, is it possible to implement a (switchable) filter in the existing architecture that might do that?

Well, I overlooked the most simple way to address this (for testing): Band 5 of the PEQ can be set as low pass, Q 0.8, Frequency 20 kHz. You'll notice that the level meter (post) now shows a reduced level. This already fixes most of the warble sound on my failing monitor system. To get full removal I need to lower the frequency to 15 kHz (this low pass is not as steep as the DAC filters).

This information might come in handy to check equipment following the ADIs, like Dominic's case of the Nord amp.

Regards
Matthias Carstens
RME

35 (edited by ddps 2020-03-09 21:12:19)

Re: Intermodulation distortion / Xiph.org

MC wrote:

You overlooked that you play back a 96 kHz file with 48 kHz bandwidth. So the filters do NOT work at 22 kHz, but 48 kHz, and make zero difference at 22 kHz.

That's a misunderstanding I had about those filters for all these years. I thought the cutoffs were sample rate independent. Thank you.

36 (edited by ddps 2020-03-09 21:12:58)

Re: Intermodulation distortion / Xiph.org

MC wrote:
ddps wrote:

And if that's not possible, is it possible to implement a (switchable) filter in the existing architecture that might do that?

Well, I overlooked the most simple way to address this (for testing): Band 5 of the PEQ can be set as low pass, Q 0.8, Frequency 20 kHz. You'll notice that the level meter (post) now shows a reduced level. This already fixes most of the warble sound on my failing monitor system. To get full removal I need to lower the frequency to 15 kHz (this low pass is not as steep as the DAC filters).

This information might come in handy to check equipment following the ADIs, like Dominic's case of the Nord amp.

That's kinda what I was wondering - thank you! This is all interesting, even if it's not 100% substantive in regard to actual music. I might try this just for my own edification. Thanks for engaging in the conversation.

Re: Intermodulation distortion / Xiph.org

I'm still doing something wrong, I don't understand what. I've tried, as suggested, using JRiver (Media Centre 26). I'm not familiar with this software but I turned all the processing off including the volume control.
Using headphone output I get the same result - I can hear the artifacts quite easily and there display shows that there is plenty of 'noise' in the audible band.
https://www.dropbox.com/s/4kj020cfkd0d6 … .HEIC?dl=0
https://www.dropbox.com/s/e86ajz2xrdpbp … .HEIC?dl=0
https://www.dropbox.com/s/dkqrrxbi55hs7 … .HEIC?dl=0
https://www.dropbox.com/s/740h3ehx1l9ov … .HEIC?dl=0

Re: Intermodulation distortion / Xiph.org

DominicPerry wrote:

Using headphone output I get the same result - I can hear the artifacts quite easily and there display shows that there is plenty of 'noise' in the audible band.

Dominic -

@MC explained this in his post:

MC wrote:

The situation at the phones output depends on the impedance of the phones. It is quite easy to run into overload with these test files without noticing it. For example a 16 Ohm phones can only be driven up to about -6 dB volume in High Power mode. Everything above will make the warble become easy to hear. Everything below is very low in volume. With 300 Ohm headphones even 0 dB High Power is free of warble (free in terms of not easily audible, it might be audible for someone with very sensitive ears). But then we are talking about a non-issue, not a design fault or a problem in the DACs and Pros.

Re: Intermodulation distortion / Xiph.org

This is nonsense. All the reasons given are changing - it was the wrong settings, then it was my amp, then it was my audio player, now it’s my headphones. Rubbish. The DAC itself shows the presence of these artifacts on its screen, well before any amp or speakers or headphone is in place.
For the record, I have four different pairs of headphones and it’s audible with all of them.

Re: Intermodulation distortion / Xiph.org

See my post #16 for the explanation why you see activity in the Analyser.

In short: the Analyser's filter bands don't seperate very high.
Even a pure, inaudible sine of 23kHz shows up down to 6.3kHz, far below it's own band.
So the Analyser cannot tell you if there is an audible signal or not with this test.

But - you hear something and I don't (on my ADI-2 Pro), there must be a reason.
The only way to find out why is to check one signal stage after the other.

Or let it be, file it under audio curiosities and forget it smile

41 (edited by DominicPerry 2020-03-10 22:24:04)

Re: Intermodulation distortion / Xiph.org

Well, I await further instructions, because now we've saying the analyser doesn't count either!
I can hear the sounds in the test files that you are not meant to be able to hear.
I've used all the players mentioned - Aurdirvana, QuickTime, StudioOne, JRiver - and I still hear it.
I've turned off the things that I've been told to turn off - DSD detection, EQ, loudness, treble/bass, and I'm still hearing it.
I've used my amp and 4 pairs of headphones - I still hear it.
I'm keeping volume below 0dB (most of the time I'm keeping it below -13, on the grounds that I'm using +13ref level, although I don't know if those are directly related).
So, I fail to understand why MC is saying that the signal isn't there at the outputs, but I can hear it using all the real-world tests so far.

Dominic

42 (edited by ddps 2020-03-11 00:08:28)

Re: Intermodulation distortion / Xiph.org

DominicPerry wrote:

Well, I await further instructions, because now we've saying the analyser doesn't count either!
I can hear the sounds in the test files that you are not meant to be able to hear.
I've used all the players mentioned - Aurdirvana, QuickTime, StudioOne, JRiver - and I still hear it.
I've turned off the things that I've been told to turn off - DSD detection, EQ, loudness, treble/bass, and I'm still hearing it.
I've used my amp and 4 pairs of headphones - I still hear it.
I'm keeping volume below 0dB (most of the time I'm keeping it below -13, on the grounds that I'm using +13ref level, although I don't know if those are directly related).
So, I fail to understand why MC is saying that the signal isn't there at the outputs, but I can hear it using all the real-world tests so far.

Dominic

Dominic,

@MC acknowledges that the signal is there, and even he can hear it on his equipment; what he has been trying to explain is that there are elements through the *analog* portion of our playback chain that can cause these very strong ultrasonic frequencies to cause interactions in a way that is audible. This is actually what part of the point of the original Xiph.org article was trying to make - every part of the playback chain is vulnerable to the side effects of these reproduced frequencies.

I understand the frustration - I have similar issues as you, but I think I understand them much better now.

43 (edited by KaiS 2020-03-11 01:26:48)

Re: Intermodulation distortion / Xiph.org

Possible reasons not yet checked:


Press button 1 until you see "State Overview"

Sync, for USB should be "conn"
SRC should be "off"
Clock Source, for USB should be "INT"
SR: should reflect the rate used, e.g. "96" or "192"


ELSE:

To check if the problem is in the digital or analog domain:
Positively use high impedance (>100 Ohms) passive (no battery, amp) headphones directly out ADI-2.
Use the EQ as High-Cut like MC proposed (Band 5, 15kHz, Q 0.8, HiCut).
If with EQ the situation gets better, ADI-2's DAC or Headphones amp is faulty.
Problem the same on both headphones outs?


ELSE, EQ doesn't change anything:
The problem is in the digital domain.

If available, instead of your Macbook Pro, try another player hardware.
Maybe a Windows PC or an iPhone with USB Adapter.
Play directly out of the browser.
Watch ADI-2's Sample Rate indicator down right in the main window's Status Bar if it reflects the playback files' rate.


Here I am at my wit's end.
Good luck!