On the Distortion of Cirrus Logic CS431xx-Based Devices: A Comparative Review

[Gitfiddle]

Your setup, voicemeter into a VM, is novel and any quality improvements need to validate before making credible claims. Simply assuming that the quality must be better is a sure way to introduce unforeseen quality degradations. The performance of Windows resampler on the other hand has been well studied and serves users even with high standards.

I never claimed anything whatsoever about windows resampler. Nothing about it's quality, no claims about voice meeter being better, and no claims about windows being bad at anything in any way; other than problems with some APO's showed in the linked thread above.

The only 2 claims I've made is after testing a second time (with higher volume and a more quiet room) I could hear the distortion in the C Major track provided in this thread on my moondrop dawn pro. Also that I didn't have any noticeable audible sampling artifacts from voice meeter. I guess that second part is what you have issue with?

My setup is not novel at all. VM was abbreviations for voice meeter, did you think I meant virtual machine? Oops, if that was the case. My setup is basic. windows 11 ->Voicemeeter -> dawn pro -> headphone amp-> headphones/IEM's.

I think this is mostly just a miscommunication.

When I get a chance, I will run a quick test of the Voicemeeter's upsampling effect as it provides a 30-day trial.

Thank you, you should be able to use voicemeeter banana without even doing the trial. The banana version is freeware and the potato is it's paid counterpart unless something changed very recently. If you want to know my specs I'm using 192k and 1024 buffer after learning about the NOS 50k oddities you found. Before I read this thread I was using the default settings.

 

[markanini]

I never claimed anything whatsoever about windows resampler. Nothing about it's quality, no claims about voice meeter being better, and no claims about windows being bad at anything in any way; other than problems with some APO's showed in the linked thread above.

The only 2 claims I've made is after testing a second time (with higher volume and a more quiet room) I could hear the distortion in the C Major track provided in this thread on my moondrop dawn pro. Also that I didn't have any noticeable audible sampling artifacts from voice meeter. I guess that second part is what you have issue with?

My setup is not novel at all. VM was abbreviations for voice meeter, did you think I meant virtual machine? Oops, if that was the case. My setup is basic. windows 11 ->Voicemeeter -> dawn pro -> headphone amp-> headphones/IEM's.

I think this is mostly just a miscommunication.


Thank you, you should be able to use voicemeeter banana without even doing the trial. The banana version is freeware and the potato is it's paid counterpart unless something changed very recently. If you want to know my specs I'm using 192k and 1024 buffer after learning about the NOS 50k oddities you found. Before I read this thread I was using the default settings.

Just passing on some possible landmines that might be helpful for you to avoid, that's all. Maybe you didn't consider that Voicemeeter can introduce unwanted/unknown alterations, or that Windows defaults are less susceptible to audio degradation. When you say your goal is to defeat windows resampler using WASAPI via Voicemeeter it's implied that you assume a positive difference. Neither is normally required or necessarily beneficial for evaluating the CMaj clip. A different issue with Voicemeeter is that it's a bit of a black box and won't tell you why if there's a difference in output, you may try other software like Foobar2000 for WASAPI exclusive capabilities because of that. The Udial clip isn't selective and also distorts(IMD) from the slightest clipping, including inter-sample peaks. Hope this helps and good luck!

 

[Gitfiddle]

Just passing on some possible landmines that might be helpful for you to avoid, that's all. Maybe you didn't consider that Voicemeeter can introduce unwanted/unknown alterations, or that Windows defaults are less susceptible to audio degradation. When you say your goal is to defeat windows resampler using WASAPI via Voicemeeter it's implied that you assume a positive difference. Neither is normally required or necessarily beneficial for evaluating the CMaj clip. A different issue with Voicemeeter is that it's a bit of a black box and won't tell you why if there's a difference in output, you may try other software like Foobar2000 for WASAPI exclusive capabilities because of that. The Udial clip isn't selective and also distorts(IMD) from the slightest clipping, including inter-sample peaks. Hope this helps and good luck!

Fair enough and I do thank you for the warning. I have managed to hear some of that audio degradation first hand when I had bad EQ settings or to low of a buffer rate years back. So your point is very valid. Good luck to you as well.

 

[boogerlad]

Is the apple usbc dac dongle affected too?

 

[MBO]

This is about my Topping D70PRO Octo.
I applied the recommendation from ASR: I set the filter to NOS, thereby I turned off DRE, but for optimal operation of the CS431xxx in NOS mode, preliminary Oversampling is necessary, which I did in Audirvana, turning on Resampling x2.
Considering that I replaced the PSU in my D70 with a more powerful and high-quality one, I got a sound quality at the output that is probably an order of magnitude higher than my previous tests with this DAC.
Thank you very much, now the sound is simply incredibly pleasant.

 

[jkim]

Is the apple usbc dac dongle affected too?

Yes

 

[jkim]

This is about my Topping D70PRO Octo.
I applied the recommendation from ASR: I set the filter to NOS, thereby I turned off DRE, but for optimal operation of the CS431xxx in NOS mode, preliminary Oversampling is necessary, which I did in Audirvana, turning on Resampling x2.
Considering that I replaced the PSU in my D70 with a more powerful and high-quality one, I got a sound quality at the output that is probably an order of magnitude higher than my previous tests with this DAC.
Thank you very much, now the sound is simply incredibly pleasant.

What OS are you on?

 

[MBO]

What OS are you on?

MacMini M1 + Asahi Linux

 

[boogerlad]

Are there any mitigations for the JCALLY JM20?

 

[jkim]

MacMini M1 + Asahi Linux

Ok. I will eventually measure the resampling quality of Mac OS and Linux. It is likely that you won't need any special software beyond what comes with the OS.

 

[jkim]

Are there any mitigations for the JCALLY JM20?

You mean you want to avoid this?

There's no way to fix this unless its firmware is updated which is unlikely. But in the case of JM20, you will rarely hear the distortion as it occurs only in a restricted situation.

 

[boogerlad]

Correct, I want to avoid that. Is there a certain volume level I can set to avoid it?

 

[jkim]

Correct, I want to avoid that. Is there a certain volume level I can set to avoid it?

Assuming you connect the JM20 directly to your headphones, how much and how often distortion occurs will depend on how loud you listen and how sensitive headphones you use. But there's nothing you can control practically. Why would you want to listen very loudly or very softly just to avoid the distortion? Why would you buy some particular headphones just to use this dongle?

You may have some control if you connect it to an external headphone amp. But using a separate headphone amp with this cheap dongle does not make sense, either.

Again, the JM20 still has its value even if it produces this distortion, which will be barely audible in most cases.

 

[Weastony]

You should hear it. The Moondrop Dawn Pro was tested by @cyril_meysson and reported here.

What’s the point of listening for distortions if it makes your ears strain (which never happens with normal music playback)? And why bother with adjusting specific signal levels or selecting certain tracks for testing? Shouldn’t we just enjoy the music instead of obsessing over negligible DAC imperfections?

P.S. I use the JM20 Max myself. It costs so little money that it's a sin to complain. It's an "endgame" DAC for most users.

 

[CedarX]

What’s the point of listening for distortions if it makes your ears strain (which never happens with normal music playback)? And why bother with adjusting specific signal levels or selecting certain tracks for testing? Shouldn’t we just enjoy the music instead of obsessing over negligible DAC imperfections?

P.S. I use the JM20 Max myself. It costs so little money that it's a sin to complain. It's an "endgame" DAC for most users.

The point is informed decision: you can decide what is important or not for you.
JCally is not exactly a mainstream brand… How did you come up with a JM20 Max? What makes you say it’s an endgame DAC for most users? Based on what criteria?

 

[jkim]

which never happens with normal music playback

Incorrect. It happens. But a more important point is not whether it happens or not. Even if it rarely happens, some individuals can perceive it and be bothered, like here.

And why bother with adjusting specific signal levels or selecting certain tracks for testing?

The test and its setup is simply to easily and clearly find out which type of distortion (full vs partial) a device produces. It does not mean the distortion occurs only at that volume setting or only with that test signal.

Shouldn’t we just enjoy the music instead of obsessing over negligible DAC imperfections?

This is a very strong argument, pointing to all the discussions happening on this Audio Science Review forum. Of course, you can enjoy the music even using a DAC/hp amp with a SINAD of 38 dB. Or people did enjoy listening to music using CD players or DACs with the 1980's technology. As you know, that is not the point here.

P.S. I use the JM20 Max myself. It costs so little money that it's a sin to complain. It's an "endgame" DAC for most users.

No doubt the JM20 Max is a great value as I reviewed although I wouldn't call it an "endgame DAC for most users."

 

[MaxwellsEq]

It's an "endgame" DAC for most users.

Welcome to ASR!

This is not an endgame DAC, for any user. It's cheap, but clearly flawed when operated using a standard setup. You should be able to feed any DAC with a steep reconstruction filter 44.1kHz sampled content and not have audible errors.

 

[Weastony]

The point is informed decision: you can decide what is important or not for you.
JCally is not exactly a mainstream brand… How did you come up with a JM20 Max? What makes you say it’s an endgame DAC for most users? Based on what criteria?

I saw it on AliExpress, then found a review here. I believe this is a DAC that will satisfy almost any user, based on reviews from Reddit and AliExpress itself. I’ve already seen many people praising the same Graveaudio DA06 (which costs $7) or jcally jm6 and not hearing a difference between it and an expensive desktop device. First and foremost, I emphasize the reasonableness of such meticulous listening for distortions and other. Music is usually enjoyed in a relaxed state

Anyway, i don't have much experience. But at the moment I do not hear any problems that could be "fixed" by buying another DAC.

 

[Gitfiddle]

I was able to run some 1k sine tests in REW using a generic microphone input and I got this comparing 48k and 192k on my moondrop dawn pro with no filter selected. I had some similar results to you regarding rise in distortion even with such poor measurement hardware (and very little idea how to use the software yet.)

 

[HissingFree]

Is it not possible to upgrade firmware to cheap CS43131 such as JM20 to a firmware from another CS43131 device with NOS filter option? or no such device exists where you can upgrade its firmware and said firmware is also compatible with JM20?

Edit: I think FiiO KA11 allows to change it's filter to non oversampled filter. KA11 should be similar to JM20 Max I think? maybe one can load 0.08 firmware to JM20 Max and enable NOS filter into JM20 Max with the FiiO control app? would be nice to test.