• Welcome to ASR. 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!

Loxjie D40 Pro DAC & HP Amp Review (NTTY)

NTTY

Major Contributor
Forum Donor
Reviewer
Joined
Apr 27, 2018
Messages
1,656
Likes
10,637
Location
Switzerland
Hello Everyone,

This is a complementary review of the Loxjie D40 Pro.

Loxjie_002.jpg


Preamble

The Loxjie D40 Pro has already been reviewed here by Amir, but as opposed to the SMSL DO400, it is not the same item I'm reviewing, as I bought this one directly from the webstore (on sale at 280$).

I thought it would be a great opportunity to review a different item to check if the performances are consistent.


Loxjie D40 Pro - Presentation

This DAC is an economic version of the SMSL DO400. The functionalities are the same, to the exception of the "sound colors" option absent in the D40 Pro. The display is basic and the remote control is not specific for this DAC, making it a little cumbersome to use without looking at the user manual. Some of the options (like the selection of filters) are directly accessible from the small buttons on the front.

I got the Loxjie D40 Pro at roughly 40% less than the DO400, but do we get almost half the performances? We'll check that.

On the back, nothing is missing:

Loxjie_005.jpg


We get the same set of inputs/outputs, which is cool.

I did not take a picture of the inside of the DO400, I'll do that when I return home next week, but at least this is the Loxjie:

Loxjie_004.jpg


The main differences between the DO400 and the D40 Pro have been shared by @Roland68 on the main SMSL DO400 review post.


User Experience

I like the fact the remote control is not mandatory, and the direct access to some functions (output, filter) from the front buttons. If the interface is very basic, it is still intuitive after one read on the user manual.

As opposed to the SMSL, the Loxjie offers only a preamp mode, so the output voltage can always be adjusted. This is good news since it does not reduce the performances and allow for a better resistance to intersample overs, as I'll show in the review (same as with the SMSL).

When the display shows -0 as volume, it means 4Vrms at balanced outputs (2Vrms unbalanced) and it goes up to +2.0, for 5Vrms output (2.5Vrms unbalanced). Oddly +0.5 means indeed 0.5dB more than the -0 setting, but -1 equals -0.5dB at the output. Go figure.


Loxjie D40 Pro - Measurements

All measurements performed with an E1DA Cosmos ADCiso (grade 0) in mono mode, the Cosmos Scaler (100kohms from unbalanced input) and APU (when required) for analog outputs. I used a Windows 11 PC to connect via USB and did not have to install a driver. I also alternatively used the SMSL PO100 Pro to feed it with Coax and Toslink SPDIF for some complementary tests.

From XLR out (non inverting) I measured:
  • At -0 setting: 4.080Vrms from right channel and 4.087Vrms from Left one, i.e. 0.01dB channel imbalance.
  • At +2.0 setting: 5.134Vrms from right channel and 5.144Vrms from Left one, i.e. 0.01dB channel imbalance.
From RCA out I measured:
  • At -0 setting: 1.979Vrms from right channel and 1.982Vrms from Left one, i.e. 0.02dB channel imbalance.
  • At +2.0 setting: 2.491Vrms from right channel and 2.495Vrms from Left one, i.e. 0.02dB channel imbalance.
----

I aligned this review with the one I made of the SMSL DO400, so not all measurements are aligned with those of Amir, as I explained in that review of the SMSL.

Here we start with the standard 1kHz @0dBFS test tone, and output volume set to 4Vrms (48kHz SR, FFT 32k, 4 averages):

LoxjeD40Pro-1kHz-4Vrms-LR.jpg


The two channels perform exactly the same with a 126.6dB SINAD. When extending the span to the one of Amir (10Hz to 22.4kHz), the SINAD is 126.3dB.

If the Loxjie is nearly half the price of the SMSL DO400, the loss on SINAD perspective is a small 1dB. And note that from the Coax input, I got 0.3dB better SINAD, meaning you don't lose anything when using it:

LoxjeD40Pro-1kHz-4Vrms-Coaxin_LR.jpg


Increasing the output volume to 5Vrms got me to 127.5dB average SINAD between the two channels (20Hz to 20kHz) and again 0.3dB lower with Amir's span (10Hz - 22.4kHz).

These are crazy numbers, even if behind the SMSL DO400.

From RCA outputs, I got an average SINAD 122.5dB at 2.5Vrms and almost 2dB less at 2Vrms, which is 2dB less than the SMSL DO400.

----

From experience reviewing CD Players, I like to run the same 1khz test 6dB lower:

LoxjeD40Pro-1kHz-2Vrms-LR.jpg


The two channels are again identical and very close to what I measured from the SMSL DO400, only a very small 0.4dB weaker.

----

Stereophile was often showing the below and I like it too, so I'm used to reuse it. It is a view from 20Hz to 1kHz with a 1kHz sine at 0dBFS:

LoxjeD40Pro_PS_LR.jpg


This graph allows a better viewing at the power supply potential related noise (leakage) and here there's nothing. I had to do 0 effort to get there, meaning the Loxjie is well isolated on power supply perspective. That said, there is some correlated (pseudo-random) noise at the foot of the fundamental that we see here. It's not a concern, but more than what I measured with the SMSL DO400, meaning some cost cuts show this way, which is reasonable.

----

Since I'm here, let me add the 50Hz @0dBFS into 600ohms, for compatibility with Stereophile measurements:

LoxjeD40Pro_50Hz_600ohms_LR.jpg


Besides the higher than usual noise floor at the foot of the fundamental, this trace is very clean, with only the H3 showing itself and then no subsequent harmonics of the 50Hz.

----

There is a selection of oversampling filters, and some are really flat in audio band, but not all of them. The below is a wideband analysis of white noise @-10dBFS, volume set at 0dB, with 44.1kHz input sampling rate:

LoxjeD40Pro-Filters.jpg


They are identical to those of the SMSL DO400, and they are the ones of the ESS9039. I recommend Filter 3 which is the Linear Phase Fast Roll-off of the ESS DAC.

The only exception is Filter0, meaning "Filter OFF" as per the documentation of Loxjie. And exactly same as with the SMSL DO400, it plays 6dB higher and clips the output stage at -3.5dB relative to 4Vrms output. Look:

LoxjeD40Pro_1kHz_Filter0.jpg


With volume set at -7 (equivalent to -3.5dB), I measured an output level of 5.714Vrms. With volume set at +2 (i.e 5.5dB more), I got 6.839Vrms, so the output stage is clipped (it should be going to 10Vrms).

This filter is useless and potentially dangerous for the other elements down the line, especially your loudspeakers. Just avoid it, or at least keep the volume at -8.

----

Let's go back to good news with the Jitter Test (48kHz, 512k FFT. 32 averages):

LoxjeD40Pro_JT_LR.jpg


This is SOTA trace. The low frequency "stressor" component of the jitter test can be (very) well seen at low frequency, meaning very low noise floor. We get again that small correlated noise at the foot of the fundamental but no visible side bands while the FFT analysis goes down to -165dBr!

----

Of course, I won't forget the Multitone test (48kHz SR, 256k FFT, 16 averages):

LoxjeD40Pro-MT_LR_4Vrms.jpg


It is revealing the increased noise floor (with signal close to full scale) that I already mentioned several times, and that manifests itself up to 200Hz. This is the real impact of the cost cut of the Loxjie when compared to the SMSL DO400. That said, being at low frequencies, where our ears are much less sensitive, I don't think it is a big issue.

----

Next is the THD+N vs Frequency (192kHz, 512k FFT 4 averages, 3PPO (20Hz to 20kHz), 4Vrms output):

LoxjeD40Pro-THDNvsFreq.jpg


This test is hitting the max resolution of the ADC, but less so than the SMSL. Basically the Loxjie has up to 6dB more distortion and noise than the SMSL DO400. Below 100Hz, the limitation comes from the correlated noise and the foot of the fundamental (again).
All that said, you see from the plot at 1kHz, we get almost -113dB THD+N at 1kHz, with all noise included up to 90kHz (Filter 3). This is still a very good performance! The other channel overlays perfectly.

----

Next is the IMD SMPTE vs output level sweep, which can't be compared to the one of Amir because this measurement is based on FFT analysis of IMD distortion only (192kHz, 128k FFT 4 averages, 1dB step, scaler in auto mode):

LoxjeD40Pro-IMDSMPTE_LRt.jpg


This trace is very close to the one of the SMSL DO400. The issue I reported before does not show here due to the nature of the test (IMD distortion components are calculated around the 7kHz test tone).

----

Dynamic range is extremely good, be it CCIR-2K weighted or A-Weighted (0dBFS = 4Vrms):

LoxjeD40Pro-DR-CCIR2K_4Vrms.jpg


LoxjeD40Pro-DR-AWeight_4Vrms.jpg


Again we lose only 2dB of performance when compared with the SMSL DO400, no audible consequences, of course, these are crazy good numbers.

----

And now let's have a look at other less usual measurements. The one that I like is the resistance to intersample-overs.
This test is performed with dedicated test tones with a specific phase shift to generate digital overs. If anything in the digital path (oversampling filter / DSP / ASRC) does not have enough headroom for digital overs, it will generate clipping within the interpolator.
For that, I'm using a test tone of 11'025Hz with a phase shift of 45° that goes up to +3.01dB over, by 1dB steps. This allows me to process what's the headroom of the digital interpolator of the DAC. Here are the results (THD+N measurement up to 90kHz):

Intersample-overs tests
THD+N 20Hz - 90kHz
Peak = -0.99dBFSPeak = +0.01dBFSPeak = +1.01dBFSPeak = +2.01dBFSPeak = +3.01dBFS
Loxjie D40 Pro (Vol +2dB)-102dB-102dB-40.9dB-25.2dB-20.4dB
Loxjie D40 Pro (Vol 0dB)-102dB-102dB-102dB-102dB-40.8dB
Loxjie D40 Pro (vol -1dB)-102dB-102dB-102dB-102dB-102dB

The above means that at full output power (5Vrms from XLR), the interpolator of the Loxjie (Linear Fast filter) has a headroom of only 1dB. But because the volume control is done in digital domain, before interpolation, there is more than 3dB headroom when lowering the volume by... 3dB.

So, with volume set at -02 (=-1dB), you get the best ISO resistance. I recommend reducing the volume that way, even if you use the Loxjie as a DAC (without needing volume control), since so many master are too hot. The penalty on SINAD is negligible, especially considering how good it is at 4Vrms.
Note that the best result (-102dB) is driven by distortion, not noise, and the SMSL DO400 was reaching -115dB of THD+N. That will not change anything to your ears, since we are talking first harmonic at 22kHz.

----

Other measurements (not shown):
  • First, I now regroup all IMD tests as below, sourced from REW's results (IMD distortion and Total Distortion+Noise shown for each):
LoxjeD40Pro-IMD.jpg

  • Crosstalk: -140dB (1kHz)
  • Pitch Error (GPSDO corrected) : 19'996.95Hz (19'997Hz requested) ie -2.5ppm
The above results are all very good. The IMD tests show a little more noise, and that is consistent with the issue I reported before.

----

Last and not least, I wanted to take the opportunity to revisit the linearity test, since these latest DACs are so good. Here we go:

LoxjeD40Pro-Linearity_R.jpg


Instead of the standard view starting at -120dBFS, I decided to start it at -140dBFS (end at -80dBFS). If we define the threshold of linearity as 0.2dB deviation from the desired output voltage, then the Loxjie easily reaches the staggering result of -131dBFS (same for the two channels).
And that is with the 0dBFS defined as 4Vrms. Relative to 5Vrms, linearity is 2dB better! Let's start chasing -140dBFS :p


Loxjie D40 Pro - Measurements - Headphones

In high gain mode, I got an already very high 9Vrms with volume set at -14 (=-7dB) into 200kΩ, meaning a max 25Vrms.
In low gain mode, at -0 volume, I got a high 3.926vRMSVrms output (into 200kΩ) from the 4.4mm jack. These are the same as the SMSL D0400.

In low gain mode, 4Vrms output, the 1kHz @0dBFS is the below:

LoxjeD40Pro-1kHz-4Vrms_HeadphonesBal_LR.jpg


This is very close to the SMSL DO400 which achieved -121.5dB of THD+N on the same test.

At 50mVrms output, in low gain mode, I get a little disappointing SNR of 84.1dB:

LoxjeD40Pro-50mV_HeadphonesBal.jpg


This is only 0.5dB less than the SMSL DO400, and concurs with the findings of Amir.


Conclusion

For almost half the price of the SMSL DO400, we certainly have more than half the performances.

The issues I spotted were not revealed by the testing of Amir's device, which is probably an indication of potential variations related to some cost cuts between two devices.

Basically, if the device that Amir reviewed is the best and mine represents the worst, then you still get a super high resolved DAC in all cases.

I hope you enjoyed this review!

Flo
 
Last edited:
Hello Everyone,

This is a complementary review of the Loxjie D40 Pro.

View attachment 545031

Preamble

The Loxjie D40 Pro has already been reviewed here by Amir, but as opposed to the SMSL DO400, it is not the same item I'm reviewing, as I bought this one directly from the webstore.

I thought it would be a great opportunity to review a different item to check if the performances are consistent.


Loxjie D40 Pro - Presentation

This DAC is an economic version of the SMSL DO400. The functionalities are the same, to the exception of the "sound colors" option absent in the D40 Pro. The display is basic and the remote control is not specific for this DAC, making it a little cumbersome to use without looking at the user manual. Some of the options (like the selection of filters) are directly accessible from the small buttons on the front.

I got the Loxjie D40 Pro at roughly 40% less than the DO400, but do we get almost half the performances? We'll check that.

On the back, nothing is missing:

View attachment 545049

We get the same set of inputs/outputs, which is cool.

I did not take a picture of the inside of the DO400, I'll do that when I return home next week, but at least this is the Loxjie:

View attachment 545050

The main differences between the DO400 and the D40 Pro have been shared by @Roland68 on the main SMSL DO400 review post.


User Experience

I like the fact the remote control is not mandatory, and the direct access to some functions (output, filter) from the front buttons. If the interface is very basic, it is still intuitive after one read on the user manual.

As opposed to the SMSL, the Loxjie offers only a preamp mode, so the output voltage can always be adjusted. This is good news since it does not reduce the performances and allow for a better resistance to intersample overs, as I'll show in the review (same as with the SMSL).

When the display shows -0 as volume, it means 4Vrms at balanced outputs (2Vrms unbalanced) and it goes up to +2.0, for 5Vrms output (2.5Vrms unbalanced). Oddly +0.5 means indeed 0.5dB more than the -0 setting, but -1 equals -0.5dB at the output. Go figure.


Loxjie D40 Pro - Measurements

All measurements performed with an E1DA Cosmos ADCiso (grade 0) in mono mode, the Cosmos Scaler (100kohms from unbalanced input) and APU (when required) for analog outputs. I used a Windows 11 PC to connect via USB and did not have to install a driver. I also alternatively used the SMSL PO100 Pro to feed it with Coax and Toslink SPDIF for some complementary tests.

From XLR out (non inverting) I measured:
  • At -0 setting: 4.080Vrms from right channel and 4.087Vrms from Left one, i.e. 0.01dB channel imbalance.
  • At +2.0 setting: 5.134Vrms from right channel and 5.144Vrms from Left one, i.e. 0.01dB channel imbalance.
From RCA out I measured:
  • At -0 setting: 1.992Vrms from right channel and 1.086Vrms from Left one, i.e. 0.01dB channel imbalance.
  • At +2.0 setting: 2.495Vrms from right channel and 2.500Vrms from Left one, i.e. 0.01dB channel imbalance.
----

I aligned this review with the one I made of the SMSL DO400, so not all measurements are aligned with those of Amir, as I explained in that review of the SMSL.

Here we start with the standard 1kHz @0dBFS test tone, and output volume set to 4Vrms (48kHz SR, FFT 32k, 4 averages):

View attachment 545054

The two channels perform exactly the same with a 126.6dB SINAD. When extending the span to the one of Amir (10Hz to 22.4kHz), the SINAD is 126.3dB.

If the Loxjie is nearly half the price of the SMSL DO400, the loss on SINAD perspective is a small 1dB. And note that from the Coax input, I got 0.3dB better SINAD, meaning you don't loose anything when using it:

View attachment 545057

Increasing the output volume to 5Vrms got me to 127.5dB average SINAD between the two channels (20Hz to 20kHz) and again 0.3dB lower with Amir's span (10Hz - 22.4kHz).

These are crazy numbers, even if behind the SMSL DO400.

From RCA outputs, I got an average SINAD 122.5dB at 2.5Vrms and almost 2dB less at 2Vrms, which is 2dB less than the SMSL DO400.

----

From experience reviewing CD Players, I like to run the same 1khz test 6dB lower:

View attachment 545060

The two channels are again identical and very close to what I measured from the SMSL DO400, only a very small 0.4dB weaker.

----

Stereophile was often showing the below and I like it too, so I'm used to reuse it. It is a view from 20Hz to 1kHz with a 1kHz sine at 0dBFS:

View attachment 545089

This graph allows a better viewing at the power supply potential related noise (leakage) and here there's nothing. I had to do 0 effort to get there, meaning the Loxjie is well isolated on power supply perspective. That said, there is some correlated (pseudo-random) noise at the foot of the fundamental that we see here. It's not a concern, but more than what I measured with the SMSL DO400, meaning some cost cuts show this way, which is reasonable.

----

Since I'm here, let me add the 50Hz @0dBFS into 600ohms, for compatibility with Stereophile measurements:

View attachment 545090

Besides the higher than usual noise floor at the foot of the fundamental, this trace is very clean, with only the H3 showing itself and then no subsequent harmonics of the 50Hz.

----

There is a selection of oversampling filters, and some are really flat in audio band, but not all of them. The below is a wideband analysis of white noise @-10dBFS, volume set at 0dB, with 44.1kHz input sampling rate:

View attachment 545065

They are identical to those of the SMSL DO400, and they are the ones of the ESS9039. I recommend Filter 3 which is the Linear Phase Fast Roll-off of the ESS DAC.

The only exception is Filter0, meaning "Filter OFF" as per the documentation of Loxjie. And exactly same as with the SMSL DO400, it plays 6dB higher and clips the output stage at -3.5dB relative to 4Vrms output. Look:

View attachment 545069

With volume set at -7 (equivalent to -3.5dB), I measured an output level of 5.714Vrms. With volume set at +2 (i.e 5.5dB more), I got 6.839Vrms, so the output stage is clipped (it should be going to 10Vrms).

This filter is useless and potentially dangerous for the other elements down the line, especially your loudspeakers. Just avoid it, or at least keep the volume at -8.

----

Let's go back to good news with the Jitter Test (48kHz, 512k FFT. 32 averages):

View attachment 545083

This is SOTA trace. The low frequency "stressor" component of the jitter test can be (very) well seen at low frequency, meaning very low noise floor. We get again that small correlated noise at the foot of the fundamental but no visible side bands while the FFT analysis goes down to -165dBr!

----

Of course, I won't forget the Multitone test (48kHz SR, 256k FFT, 16 averages):

View attachment 545086

It is revealing the increased noise floor (with signal close to full scale) that I already mentioned several times, and that manifests itself up to 200Hz. This is the real impact of the cost cut of the Loxjie when compared to the SMSL DO400. That said, being at low frequencies, where our ears are much less sensitive, I don't think it is a big issue.

----

Next is the THD+N vs Frequency (192kHz, 512k FFT 4 averages, 3PPO (20Hz to 20kHz), 4Vrms output):

View attachment 545093

This test is hitting the max resolution of the ADC, but less so than the SMSL. Basically the Loxjie has up to 6dB more distortion and noise than the SMSL DO400. Below 100Hz, the limitation comes from the correlated noise and the foot of the fundamental (again).
All that said, you see from the plot at 1kHz, we get almost -113dB THD+N at 1kHz, with all noise included up to 90kHz (Filter 3). This is still a very good performance! The other channel overlays perfectly.

----

Next is the IMD SMPTE vs output level sweep, which can't be compared to the one of Amir because this measurement is based on FFT analysis of IMD distortion only (192kHz, 128k FFT 4 averages, 1dB step, scaler in auto mode):

View attachment 545098

This trace is very close to the one of the SMSL DO400. The issue I reported before does not show here due to the nature of the test (IMD distortion components are calculated around the 7kHz test tone).

----

Dynamic range is extremely good, be it CCIR-2K weighted or A-Weighted (0dBFS = 4Vrms):

View attachment 545099

View attachment 545100

Again we loose only 2dB of performance when compared with the SMSL DO400, no audible consequences, of course, these are crazy good numbers.

----

And now let's have a look at other less usual measurements. The one that I like is the resistance to intersample-overs.
This test is performed with dedicated test tones with a specific phase shift to generate digital overs. If anything in the digital path (oversampling filter / DSP / ASRC) does not have enough headroom for digital overs, it will generate clipping within the interpolator.
For that, I'm using a test tone of 11'025Hz with a phase shift of 45° that goes up to +3.01dB over, by 1dB steps. This allows me to process what's the headroom of the digital interpolator of the DAC. Here are the results (THD+N measurement up to 90kHz):

Intersample-overs tests
THD+N 20Hz - 90kHz
Peak = -0.99dBFSPeak = +0.01dBFSPeak = +1.01dBFSPeak = +2.01dBFSPeak = +3.01dBFS
Loxjie D40 Pro (Vol +2dB)-102dB-102dB-40.9dB-25.2dB-20.4dB
Loxjie D40 Pro (Vol 0dB)-102dB-102dB-102dB-102dB-40.8dB
Loxjie D40 Pro (vol -1dB)-102dB-102dB-102dB-102dB-102dB

The above means that at full output power (5Vrms from XLR), the interpolator of the Loxjie (Linear Fast filter) has a headroom of only 1dB. But because the volume control is done in digital domain, before interpolation, there is more than 3dB headroom when lowering the volume by... 3dB.

So, with volume set at -02 (=-1dB), you get the best ISO resistance. I recommend reducing the volume that way, even if you use the Loxjie as a DAC (without needing volume control), since so many master are too hot. The penalty on SINAD is negligible, especially considering how good it is at 4Vrms.
Note that the best result (-102dB) is driven by distortion, not noise, and the SMSL DO400 was reaching -115dB of THD+N. That will not change anything to your ears, since we are talking first harmonic at 22kHz.

----

Other measurements (not shown):
  • First, I now regroup all IMD tests as below, sourced from REW's results (IMD distortion and Total Distortion+Noise shown for each):
View attachment 545108
  • Crosstalk: -140dB (1kHz)
  • Pitch Error (GPSDO corrected) : 19'996.95Hz (19'997Hz requested) ie -2.5ppm
The above results are all very good. The IMD tests show a little more noise, and that is consistent with the issue I reported before.

----

Last and not least, I wanted to take the opportunity to revisit the linearity test, since these latest DACs are so good. Here we go:

View attachment 545122

Instead of the standard view starting at -120dBFS, I decided to start it at -140dBFS (end at -80dBFS). If we define the threshold of linearity as 0.2dB deviation from the desired output voltage, then the Loxjie easily reaches the staggering result of -131dBFS (same for the two channels).
And that is with the 0dBFS defined as 4Vrms. Relative to 5Vrms, linearity is 2dB better! Let's start chasing -140dBFS :p


Loxjie D40 Pro - Measurements - Headphones

In high gain mode, I got an already very high 9Vrms with volume set at -14 (=-7dB) into 200kΩ, meaning a max 25Vrms.
In low gain mode, at -0 volume, I got a high 3.926vRMSVrms output (into 200kΩ) from the 4.4mm jack. These are the same as the SMSL D0400.

In low gain mode, 4Vrms output, the 1kHz @0dBFS is the below:

View attachment 545113

This is very close to the SMSL DO400 which achieved -121.5dB of THD+N on the same test.

At 50mVrms output, in low gain mode, I get a little disappointing SNR of 84.1dB:

View attachment 545115

This is only 0.5dB less than the SMSL DO400, and concurs with the findings of Amir.


Conclusion

For almost half the price of the SMSL DO400, we certainly have more than half the performances.

The issues I spotted were not revealed by the testing of Amir's device, which is probably an indication of potential variations related to some cost cuts between two devices.

Basically, if the device that Amir reviewed is the best and mine represents the worst, then you still get a super high resolved DAC in all cases.

I hope you enjoyed this review!

Flo
I'm so impressed by these reviews, showing results we don't see elsewhere and obviously requiring deep knowledge of the subject. Just one language niggle, though: it's "lose", not "loose", which means something else. <https://www.oed.com/search/dictionary/?scope=Entries&q=lose> or <https://www.merriam-webster.com/dictionary/lose> if you want to check. If only English words were pronounced how they are spelt (or spelled)! ;-)
 
Thank you so much for the review!

Although it's quite technical for my level, there are some things I understand. It seems that with the filter off there might be a risk of clipping at high volumes. I'm going to quickly select the fast roll-off filter on my DL400! ;)
 
Great review!I'm soooooooooo happy to know that now we have two expertised reviewers frequently conducting detailed measurements of DAC. The review is of great value for exploring the limitation of the device, and for knowing the edge of current DAC technology. Thank you for sharing! Keep it up!
 
Nice!

Many thanks for the hard work, and keep them coming!
 
I'm so impressed by these reviews, showing results we don't see elsewhere and obviously requiring deep knowledge of the subject. Just one language niggle, though: it's "lose", not "loose", which means something else. <https://www.oed.com/search/dictionary/?scope=Entries&q=lose> or <https://www.merriam-webster.com/dictionary/lose> if you want to check. If only English words were pronounced how they are spelt (or spelled)! ;-)
You mean 'Bough' as in the bough in a tree or the late presenter Frank Bough, who's surname was pronounced 'Boff.'

I'd suggest English as a language isn't easy to learn properly and I'm deeply envious of those from foreign climes who learn to speak it so much better than some natives I'd suggest - and I'm using myself as a reference here as my other-language skills are next to zero, including my late mother's Welsh ;)

I'll disappear back under my stone now ;)
 
I am fan of your reviews! Complements and extends so much of the routine ASR reviews here. Is like a remixed version of an album with a different artistic take. :)
 
You mean 'Bough' as in the bough in a tree or the late presenter Frank Bough, who's surname was pronounced 'Boff.'

I'd suggest English as a language isn't easy to learn properly and I'm deeply envious of those from foreign climes who learn to speak it so much better than some natives I'd suggest - and I'm using myself as a reference here as my other-language skills are next to zero, including my late mother's Welsh ;)

I'll disappear back under my stone now ;)
:) There so many, but I was referring to how "lose" is pronounced, which I think is what led to the confusion with "loose". It's the inconsistencies more than anything and you just get to know them over time. :)
 
Hello Everyone,

This is a complementary review of the Loxjie D40 Pro.

View attachment 545031

Preamble

The Loxjie D40 Pro has already been reviewed here by Amir, but as opposed to the SMSL DO400, it is not the same item I'm reviewing, as I bought this one directly from the webstore (on sale at 280$).

I thought it would be a great opportunity to review a different item to check if the performances are consistent.


Loxjie D40 Pro - Presentation

This DAC is an economic version of the SMSL DO400. The functionalities are the same, to the exception of the "sound colors" option absent in the D40 Pro. The display is basic and the remote control is not specific for this DAC, making it a little cumbersome to use without looking at the user manual. Some of the options (like the selection of filters) are directly accessible from the small buttons on the front.

I got the Loxjie D40 Pro at roughly 40% less than the DO400, but do we get almost half the performances? We'll check that.

On the back, nothing is missing:

View attachment 545049

We get the same set of inputs/outputs, which is cool.

I did not take a picture of the inside of the DO400, I'll do that when I return home next week, but at least this is the Loxjie:

View attachment 545050

The main differences between the DO400 and the D40 Pro have been shared by @Roland68 on the main SMSL DO400 review post.


User Experience

I like the fact the remote control is not mandatory, and the direct access to some functions (output, filter) from the front buttons. If the interface is very basic, it is still intuitive after one read on the user manual.

As opposed to the SMSL, the Loxjie offers only a preamp mode, so the output voltage can always be adjusted. This is good news since it does not reduce the performances and allow for a better resistance to intersample overs, as I'll show in the review (same as with the SMSL).

When the display shows -0 as volume, it means 4Vrms at balanced outputs (2Vrms unbalanced) and it goes up to +2.0, for 5Vrms output (2.5Vrms unbalanced). Oddly +0.5 means indeed 0.5dB more than the -0 setting, but -1 equals -0.5dB at the output. Go figure.


Loxjie D40 Pro - Measurements

All measurements performed with an E1DA Cosmos ADCiso (grade 0) in mono mode, the Cosmos Scaler (100kohms from unbalanced input) and APU (when required) for analog outputs. I used a Windows 11 PC to connect via USB and did not have to install a driver. I also alternatively used the SMSL PO100 Pro to feed it with Coax and Toslink SPDIF for some complementary tests.

From XLR out (non inverting) I measured:
  • At -0 setting: 4.080Vrms from right channel and 4.087Vrms from Left one, i.e. 0.01dB channel imbalance.
  • At +2.0 setting: 5.134Vrms from right channel and 5.144Vrms from Left one, i.e. 0.01dB channel imbalance.
From RCA out I measured:
  • At -0 setting: 1.992Vrms from right channel and 1.086Vrms from Left one, i.e. 0.01dB channel imbalance.
  • At +2.0 setting: 2.495Vrms from right channel and 2.500Vrms from Left one, i.e. 0.01dB channel imbalance.
----

I aligned this review with the one I made of the SMSL DO400, so not all measurements are aligned with those of Amir, as I explained in that review of the SMSL.

Here we start with the standard 1kHz @0dBFS test tone, and output volume set to 4Vrms (48kHz SR, FFT 32k, 4 averages):

View attachment 545054

The two channels perform exactly the same with a 126.6dB SINAD. When extending the span to the one of Amir (10Hz to 22.4kHz), the SINAD is 126.3dB.

If the Loxjie is nearly half the price of the SMSL DO400, the loss on SINAD perspective is a small 1dB. And note that from the Coax input, I got 0.3dB better SINAD, meaning you don't lose anything when using it:

View attachment 545057

Increasing the output volume to 5Vrms got me to 127.5dB average SINAD between the two channels (20Hz to 20kHz) and again 0.3dB lower with Amir's span (10Hz - 22.4kHz).

These are crazy numbers, even if behind the SMSL DO400.

From RCA outputs, I got an average SINAD 122.5dB at 2.5Vrms and almost 2dB less at 2Vrms, which is 2dB less than the SMSL DO400.

----

From experience reviewing CD Players, I like to run the same 1khz test 6dB lower:

View attachment 545060

The two channels are again identical and very close to what I measured from the SMSL DO400, only a very small 0.4dB weaker.

----

Stereophile was often showing the below and I like it too, so I'm used to reuse it. It is a view from 20Hz to 1kHz with a 1kHz sine at 0dBFS:

View attachment 545089

This graph allows a better viewing at the power supply potential related noise (leakage) and here there's nothing. I had to do 0 effort to get there, meaning the Loxjie is well isolated on power supply perspective. That said, there is some correlated (pseudo-random) noise at the foot of the fundamental that we see here. It's not a concern, but more than what I measured with the SMSL DO400, meaning some cost cuts show this way, which is reasonable.

----

Since I'm here, let me add the 50Hz @0dBFS into 600ohms, for compatibility with Stereophile measurements:

View attachment 545090

Besides the higher than usual noise floor at the foot of the fundamental, this trace is very clean, with only the H3 showing itself and then no subsequent harmonics of the 50Hz.

----

There is a selection of oversampling filters, and some are really flat in audio band, but not all of them. The below is a wideband analysis of white noise @-10dBFS, volume set at 0dB, with 44.1kHz input sampling rate:

View attachment 545065

They are identical to those of the SMSL DO400, and they are the ones of the ESS9039. I recommend Filter 3 which is the Linear Phase Fast Roll-off of the ESS DAC.

The only exception is Filter0, meaning "Filter OFF" as per the documentation of Loxjie. And exactly same as with the SMSL DO400, it plays 6dB higher and clips the output stage at -3.5dB relative to 4Vrms output. Look:

View attachment 545069

With volume set at -7 (equivalent to -3.5dB), I measured an output level of 5.714Vrms. With volume set at +2 (i.e 5.5dB more), I got 6.839Vrms, so the output stage is clipped (it should be going to 10Vrms).

This filter is useless and potentially dangerous for the other elements down the line, especially your loudspeakers. Just avoid it, or at least keep the volume at -8.

----

Let's go back to good news with the Jitter Test (48kHz, 512k FFT. 32 averages):

View attachment 545083

This is SOTA trace. The low frequency "stressor" component of the jitter test can be (very) well seen at low frequency, meaning very low noise floor. We get again that small correlated noise at the foot of the fundamental but no visible side bands while the FFT analysis goes down to -165dBr!

----

Of course, I won't forget the Multitone test (48kHz SR, 256k FFT, 16 averages):

View attachment 545086

It is revealing the increased noise floor (with signal close to full scale) that I already mentioned several times, and that manifests itself up to 200Hz. This is the real impact of the cost cut of the Loxjie when compared to the SMSL DO400. That said, being at low frequencies, where our ears are much less sensitive, I don't think it is a big issue.

----

Next is the THD+N vs Frequency (192kHz, 512k FFT 4 averages, 3PPO (20Hz to 20kHz), 4Vrms output):

View attachment 545093

This test is hitting the max resolution of the ADC, but less so than the SMSL. Basically the Loxjie has up to 6dB more distortion and noise than the SMSL DO400. Below 100Hz, the limitation comes from the correlated noise and the foot of the fundamental (again).
All that said, you see from the plot at 1kHz, we get almost -113dB THD+N at 1kHz, with all noise included up to 90kHz (Filter 3). This is still a very good performance! The other channel overlays perfectly.

----

Next is the IMD SMPTE vs output level sweep, which can't be compared to the one of Amir because this measurement is based on FFT analysis of IMD distortion only (192kHz, 128k FFT 4 averages, 1dB step, scaler in auto mode):

View attachment 545098

This trace is very close to the one of the SMSL DO400. The issue I reported before does not show here due to the nature of the test (IMD distortion components are calculated around the 7kHz test tone).

----

Dynamic range is extremely good, be it CCIR-2K weighted or A-Weighted (0dBFS = 4Vrms):

View attachment 545099

View attachment 545100

Again we lose only 2dB of performance when compared with the SMSL DO400, no audible consequences, of course, these are crazy good numbers.

----

And now let's have a look at other less usual measurements. The one that I like is the resistance to intersample-overs.
This test is performed with dedicated test tones with a specific phase shift to generate digital overs. If anything in the digital path (oversampling filter / DSP / ASRC) does not have enough headroom for digital overs, it will generate clipping within the interpolator.
For that, I'm using a test tone of 11'025Hz with a phase shift of 45° that goes up to +3.01dB over, by 1dB steps. This allows me to process what's the headroom of the digital interpolator of the DAC. Here are the results (THD+N measurement up to 90kHz):

Intersample-overs tests
THD+N 20Hz - 90kHz
Peak = -0.99dBFSPeak = +0.01dBFSPeak = +1.01dBFSPeak = +2.01dBFSPeak = +3.01dBFS
Loxjie D40 Pro (Vol +2dB)-102dB-102dB-40.9dB-25.2dB-20.4dB
Loxjie D40 Pro (Vol 0dB)-102dB-102dB-102dB-102dB-40.8dB
Loxjie D40 Pro (vol -1dB)-102dB-102dB-102dB-102dB-102dB

The above means that at full output power (5Vrms from XLR), the interpolator of the Loxjie (Linear Fast filter) has a headroom of only 1dB. But because the volume control is done in digital domain, before interpolation, there is more than 3dB headroom when lowering the volume by... 3dB.

So, with volume set at -02 (=-1dB), you get the best ISO resistance. I recommend reducing the volume that way, even if you use the Loxjie as a DAC (without needing volume control), since so many master are too hot. The penalty on SINAD is negligible, especially considering how good it is at 4Vrms.
Note that the best result (-102dB) is driven by distortion, not noise, and the SMSL DO400 was reaching -115dB of THD+N. That will not change anything to your ears, since we are talking first harmonic at 22kHz.

----

Other measurements (not shown):
  • First, I now regroup all IMD tests as below, sourced from REW's results (IMD distortion and Total Distortion+Noise shown for each):
View attachment 545108
  • Crosstalk: -140dB (1kHz)
  • Pitch Error (GPSDO corrected) : 19'996.95Hz (19'997Hz requested) ie -2.5ppm
The above results are all very good. The IMD tests show a little more noise, and that is consistent with the issue I reported before.

----

Last and not least, I wanted to take the opportunity to revisit the linearity test, since these latest DACs are so good. Here we go:

View attachment 545122

Instead of the standard view starting at -120dBFS, I decided to start it at -140dBFS (end at -80dBFS). If we define the threshold of linearity as 0.2dB deviation from the desired output voltage, then the Loxjie easily reaches the staggering result of -131dBFS (same for the two channels).
And that is with the 0dBFS defined as 4Vrms. Relative to 5Vrms, linearity is 2dB better! Let's start chasing -140dBFS :p


Loxjie D40 Pro - Measurements - Headphones

In high gain mode, I got an already very high 9Vrms with volume set at -14 (=-7dB) into 200kΩ, meaning a max 25Vrms.
In low gain mode, at -0 volume, I got a high 3.926vRMSVrms output (into 200kΩ) from the 4.4mm jack. These are the same as the SMSL D0400.

In low gain mode, 4Vrms output, the 1kHz @0dBFS is the below:

View attachment 545113

This is very close to the SMSL DO400 which achieved -121.5dB of THD+N on the same test.

At 50mVrms output, in low gain mode, I get a little disappointing SNR of 84.1dB:

View attachment 545115

This is only 0.5dB less than the SMSL DO400, and concurs with the findings of Amir.


Conclusion

For almost half the price of the SMSL DO400, we certainly have more than half the performances.

The issues I spotted were not revealed by the testing of Amir's device, which is probably an indication of potential variations related to some cost cuts between two devices.

Basically, if the device that Amir reviewed is the best and mine represents the worst, then you still get a super high resolved DAC in all cases.

I hope you enjoyed this review!

Flo
Just so you know, your unit is definitely a later version featuring at least one layout change on the board. In the initial production run, the left and right RCA outputs were swapped; this was corrected on the board by manually soldering in resistors in a cross-over configuration.
In your version, that layout error has been fixed.
I don't see any changes to the component selection, though Chinese manufacturers can sometimes make poor choices when sourcing parts—such as op-amps, capacitors, or SMD components.
Naturally, this can lead to variations in measurement results.
 
NTTY, I add to the thanks here, well done. I purchased a D40 Pro for $217 during June Prime Days and it has been offered near that price on Amazon in the recent past, so it may again be in the future. Subjectively near perfect for me.

The board of the one I received looks identical to yours, so a later iteration. I understand their were minor early firmware issues. Mine showed the firmware (I think) on initial boot, but I have not been able to figure out how to see it again. The XMOS driver shows the Loxjie as version 3.23 which I believe is the latest.

My only quibble with your assessment is on the remote. It is a bit generic (the volume up/down does nothing but the levels up/down control volume) but I think more intuitive and complete than either the D40 (non-pro) or SMSLDO400 remotes. The "Mode" button directly switches outputs and I like the "All" choice they added which...well outputs HPA, SE, and Balanced simultaneously (I think).
NewRemote.jpg


No question the D40 Pro / D60 / DO400 came off the same line, but I do not think minor component differences would be due to cost cutting. That is, the initial 2023 D40 Price of $430 closely aligns with the current $500 DO400 offering and I expect substituting minor components would increase rather than decrease mfg. cost on the same line.

I expect current Loxjie (and Sabaj) discounts are the result of SMLS discontinuing those brands. thanks again for this and keep it up!
 
Insane value at these price points. A full featured 9039 Pro DAC well below the DO400 price.

And it even has all 3 headphone outputs.

The only thing that kept me back from this one is it's design. I find these looks really polarizing. Others might disagree.
 
Yeah, I don’t like remote controls, and they must stay optional and basic to me, in case I lose them. That’s why I like that there’s an interface (and or buttons) on the device itself for direct access.

I gave the Loxjie to my mum, who’s 93, and it’s the third time she calls me to complain about the volume buttons that don’t change the volume :)

- Press level buttons, mum.
- Yeah, that means nothing to me, son.
- I know :facepalm:
 
Yeah, I don’t like remote controls, and they must stay optional and basic to me, in case I lose them. That’s why I like that there’s an interface (and or buttons) on the device itself for direct access.

I gave the Loxjie to my mum, who’s 93, and it’s the third time she calls me to complain about the volume buttons that don’t change the volume :)

- Press level buttons, mum.
- Yeah, that means nothing to me, son.
- I know :facepalm:
You can find so-called "long-nib markers" on AliExpress—available in white, black, and blue, among other colors—that offer high opacity and adhere to almost any surface.
You can use them to alter the labeling on remote controls and other devices.
 
Back
Top Bottom