Luxsin X9 Smart Stereo DAC & Preamp

[Mambo95]

Hello everyone,

A few days ago, I installed the two latest available updates for my Luxsin X8 and X9, that's to say :

- v1.0.026 for X8
- v2.0.0.5 for X9

These 2 latest updates turned out to be CATASTROPHIC and DANGEROUS because I discovered, quite by chance, within just a few days that the remote controls of two other devices, completely unrelated to Luxsin, could also control X8 and X9!!! (see photos).

- ON/OFF button of Bluesound Node remote changes displays of X8/X9 (switching from one VU-meter to another).

- But most importantly, REC button of my digital terrestrial receiver increases volume of X8/X9, which is obviously VERY DANGEROUS for my expensive speakers and ears!

I don't even want to think about what would happen if I started fiddling with the buttons on all the other remotes I own...

Therefore, I reverted to the previous updates, specifically:
- v1.0.0.25 for the X8
- v2.0.0.4 for the X9

AND EVERYTHING IS BACK TO NORMAL!!!

Help, LUXSIN!

 

[Luxsin]

Hello everyone,

A few days ago, I installed the two latest available updates for my Luxsin X8 and X9, that's to say :

- v1.0.026 for X8
- v2.0.0.5 for X9

These 2 latest updates turned out to be CATASTROPHIC and DANGEROUS because I discovered, quite by chance, within just a few days that the remote controls of two other devices, completely unrelated to Luxsin, could also control X8 and X9!!! (see photos).

- ON/OFF button of Bluesound Node remote changes displays of X8/X9 (switching from one VU-meter to another).

- But most importantly, REC button of my digital terrestrial receiver increases volume of X8/X9, which is obviously VERY DANGEROUS for my expensive speakers and ears!

I don't even want to think about what would happen if I started fiddling with the buttons on all the other remotes I own...

Therefore, I reverted to the previous updates, specifically:
- v1.0.0.25 for the X8
- v2.0.0.4 for the X9

AND EVERYTHING IS BACK TO NORMAL!!!

Help, LUXSIN!

View attachment 536754 View attachment 536755

Thanks for the feedback, this issue will be fixed in next firmware upgrade.

 

[Luxsin]

Beta firmware 2.0.0.6 is now available for the Luxsin X9

1. Set the redesigned Web UI as the default web interface.
2. Optimized HTTPS service stability and performance.
3. Adjusted sample rate display formatting.
4. Headphone plug/unplug status is now synchronized with the Web UI output display.
5. Fixed an issue where remote controls from other devices could unintentionally trigger X9 operations.
6. Fixed an issue where the right-channel subwoofer delay setting would not be retained after a reboot.

 

[Honor_the_maggot]

Sorry if it's been discussed, and my understanding of ARC is almost nil; but could a firmware update enable the HDMI input to receive more than ARC? For example, my Oppo 205 has an "audio only" HDMI output that I'd like to run into that HDMI input on the Luxsin X9, even better if that HDMI could receive DSD.
(It's virtually a nonissue, since a conversion to PCM apparently means nothing in terms of sound quality. Yet: the X9 can handle DSD but apparently only through USB, the coaxial/optical will convert it to PCM. If the X9 can handle DSD, but cannot handle analog input without converting it, wouldn't it be nice to allow the DSD straight from an SACD player?)
But I don't understand the way HDMI ports are made and this might simply be a hardware issue. Maybe there is a workaround and need to just do some homework.

 

[Honor_the_maggot]

If you only select XLR out, nothing attached to RCA will receive any sound. If nothing is attached to RCA, then leave it on XLR-only. The Subwoofer output is treated separately, so setting it to XLR only will not turn off the Subwoofer output. Setting X9 to DSP Bypass will disable the Subwoofer output. Turning Subwoofer off will disable Subwoofer output. So to use a subwoofer via the Subwoofer output, you must disable Bypass and enable Subwoofer. Output XLR versus XLR+RCA has no effect on the Subwoofer output.

Sorry, this has bound to have been addressed in the thread, but why is subwoofer output disabled when DSP Bypass is enabled? Why does Bypass need exclude sub output, esp. if one preferred to send the sub signal out full-range without X9 low-pass?
I'd noticed a firmware update (1.0.0.18): "Fixed subwoofer no output issue after disabling HP-EQ".....but that refers to something other than enabling Bypass mode, right?

 

[SoySalsaPicante]

Sorry, this has bound to have been addressed in the thread, but why is subwoofer output disabled when DSP Bypass is enabled? Why does Bypass need exclude sub output, esp. if one preferred to send the sub signal out full-range without X9 low-pass?
I'd noticed a firmware update (1.0.0.18): "Fixed subwoofer no output issue after disabling HP-EQ".....but that refers to something other than enabling Bypass mode, right?

All bass management is digital domain and performed in a separate DSP module before conversion to analog. DSP Bypass completely bypasses said DSP chip. That’s why the subwoofer output uses digital attenuation (does not pass through the resistor ladder attenuator).

 

[VilmFilm]

Should I just bite the bullet and buy X9 for 2.1 stereo setup or should I way till the B9 comes out later?

 

[MandorliM]

View attachment 531888

Beta firmware 2.0.0.5 is now available for the X9

Download

USB driver

luxsinaudio.com

-Luxsin

Is Now full signal to speakers when enable subwoofer, added in this firmware?
thanks

 

[Luxsin]

Is Now full signal to speakers when enable subwoofer, added in this firmware?
thanks

Full-range output options for both sub and main speakers implemented in firmware 2.0.0.1

 

[gtg465x]

Found a bug / software design flaw with the implementation of the Loudness feature. The Luxsin X8 (and most likely X9) is directly applying the equal loudness contours from standard ISO 226-2003, but what it should be doing is applying the difference between the contours. For example, if they're calibrating to the 80 phon contour as they say, and you set the Loudness threshold on your X8 to -20 dB, then when you set the volume to - 30 dB (10 below your threshold), then they should be calculating the difference between the 70 phon and 80 phon contours and applying that correction to make 70 phon sound like 80 phon. Instead, what they're doing is simply applying the 70 phon contour when you set the volume to 10 dB below the threshold.

Here's some measurements I took of the X8 to show this. The X8 Loudness is set to -20 dB threshold with the default Bass and Treble values of 7 and 7. The red lines are the measurements of the X8 frequency response at -30 dB and -40 dB. The white lines are the eq the X8 should be applying (the difference between the contours instead of the contours themselves). I used code from https://github.com/grisys83/FIR-Filter-Maker-for-Equal-Loudness to generate the correct frequency responses (differences between contours). And if you want another source of how both the contours and the corrective eq being applied should look, you can reference https://musios.app/projects/equal-loudness/loudness-eq.html

I found evidence that Eversolo makes the same mistake:
- https://www.audiosciencereview.com/forum/index.php?threads/you-need-loudness-correction.63954/page-4

I would appreciate if Luxsin could pass this information along to the engineers to be fixed.

In the mean time, I have found that setting Loudness Bass to 5.5 and Treble to 4 better approximates the correct correction. This setting is shown in blue:

 

[Luxsin]

Found a bug / software design flaw with the implementation of the Loudness feature. The Luxsin X8 (and most likely X9) is directly applying the equal loudness contours from standard ISO 226-2003, but what it should be doing is applying the difference between the contours. For example, if they're calibrating to the 80 phon contour as they say, and you set the Loudness threshold on your X8 to -20 dB, then when you set the volume to - 30 dB (10 below your threshold), then they should be calculating the difference between the 70 phon and 80 phon contours and applying that correction to make 70 phon sound like 80 phon. Instead, what they're doing is simply applying the 70 phon contour when you set the volume to 10 dB below the threshold.

Here's some measurements I took of the X8 to show this. The X8 Loudness is set to -20 dB threshold with the default Bass and Treble values of 7 and 7. The red lines are the measurements of the X8 frequency response at -30 dB and -40 dB. The white lines are the eq the X8 should be applying (the difference between the contours instead of the contours themselves). I used code from https://github.com/grisys83/FIR-Filter-Maker-for-Equal-Loudness to generate the correct frequency responses (differences between contours). And if you want another source of how both the contours and the corrective eq being applied should look, you can reference https://musios.app/projects/equal-loudness/loudness-eq.html
View attachment 538957
I found evidence that Eversolo makes the same mistake:
- https://www.audiosciencereview.com/forum/index.php?threads/you-need-loudness-correction.63954/page-4

I would appreciate if Luxsin could pass this information along to the engineers to be fixed.

In the mean time, I have found that setting Loudness Bass to 5.5 and Treble to 4 better approximates the correct correction. This setting is shown in blue:

View attachment 538958

Thank you very much for your feedback. We’ll have our engineering team look into it.

-Luxsin

 

[Ange33600]

The new « auditory compensation » function that will be present in a future update on the X9 is THE function that changes everything, especially with the technique you choose to define the thresholds in dB upstream and by frequency range

I hope it will be possible to do it for the 2 ears together but also by ear, because hearing deficits are often different in LR

Will this "auditory compensation" be stored upstream in a new block before an EQ step or will it be considered a memorized EQ to which we can then make a copy to "add" our equalization settings?

 

[Sparky]

If the X9 had an AES/EBU output and XLR sub-out it would be an insta-buy for me. Still, very interesting product so I'll keep an eye on Luxsin.

 

[VilmFilm]

If the X9 had an AES/EBU output and XLR sub-out it would be an insta-buy for me. Still, very interesting product so I'll keep an eye on Luxsin.

Good luck with that. Luxsin seems more focused on experimental software and gimmicks. All I asked for was a form factor of x8 with single sub out but nahhhhhh lets make a DAC with a wireless knob that can detach instead of just using a remote and probably adds more to the production cost for a useless gimmick.

 

[Sparky]

Good luck with that. Luxsin seems more focused on experimental software and gimmicks. All I asked for was a form factor of x8 with single sub out but nahhhhhh lets make a DAC with a wireless knob that can detach instead of just using a remote and probably adds more to the production cost for a useless gimmick.

Haha. Yeah, I saw that! I thought it was pretty cool at first until I realised it's just another gimmick. I'll be keeping an eye on their future products but I won't hold my breath.

 

[rtraideo]

Anyone operating one of these with a Linux USB source? I'd like to know how this DAC handles these scenarios. Many modern DACs have some of these issues under Linux.

• slow to "wake up" when cold starting a track, resulting in missing audio for the starting few hundred milliseconds of the track
• spurious noises when cold starting or stopping a track
• slow to "wake up" after sample rate changes between tracks, again resulting in missing audio
• spurious noises when sample rate changes between tracks

 

[gtg465x]

Anyone operating one of these with a Linux USB source? I'd like to know how this DAC handles these scenarios. Many modern DACs have some of these issues under Linux.

• slow to "wake up" when cold starting a track, resulting in missing audio for the starting few hundred milliseconds of the track
• spurious noises when cold starting or stopping a track
• slow to "wake up" after sample rate changes between tracks, again resulting in missing audio
• spurious noises when sample rate changes between tracks

On Mac, it’s definitely slow to wake and loses the first few hundred ms of audio, but at least you can avoid the issue by setting it to never sleep. I do wish there was some way to power it on and off with the computer, though.

 

[Luxsin]

Beta firmware 2.0.0.7 is now available for the X9

1. Added Hearing Compensation feature.
2. Fixed an issue where certain EQ settings did not take effect.
3. Fixed an issue where the X9 became unresponsive after a failed OTA firmware update.

Watch the video below to learn how to complete the Hearing Compensation test on the X9.

 

[elgol]

My thoughts, a.please remove the annoying music from tutorial, b. Doesn't our different headphones affect the measurements?must calibrate the headphones first and with which curve? c. Assume we can apply this in speaker listening

 

[Luxsin]

My thoughts, a.please remove the annoying music from tutorial, b. Doesn't our different headphones affect the measurements?must calibrate the headphones first and with which curve? c. Assume we can apply this in speaker listening

Different headphones will indeed produce different hearing test results, since every headphone has its own frequency response. When running the Hearing Compensation test, please make sure all other DSP features (HP-EQ, PEQ, Crossfeed, etc.) are disabled.
And yes, the feature also works with speakers, you can perform the hearing test using speakers, and the resulting compensation curve can be applied to the RCA/XLR outputs as well.