Gustard X18 Review (Stereo DAC)

[Nango]

Buyer beware: Update your X18 with the latest firmware. If Gustard hasn't released anything, ask them for it.

@WolfX-700 measured this DAC a few days ago. https://www.l7audiolab.com/f/gustard-dac-x18/
While most of the usual tests yielded great results, the multitone test results saw significantly higher distortions than the SINAD number suggested, and they varied quite a bit depending on the different types of filter selected, including the default filter.

After this bug was communicated, Gustard quickly sent L7audiolab a different unit with updated firmware. The buggy multitone results were visible when the DAC was fed 0dBFS signal; therefore, Gustard decided to simply lower the level of the output voltage when the buggy filters were selected to prevent the signal every reaching 0dBFS. However, the non-buggy filters still gave the full voltage.

As a result, changing the filters literally changes the voltage by around 0.2Vrms, which could be audible. I found the way of fixing this bug kinda funny. Before the fix, you might not detect the distortions rising to -80dB. After the fix, you might actually detect the lower volume when switching to a different filter.

The fix and the new firmware were developed 3 days ago. Shenzhenaudio started selling the X18 DAC 7 days ago. This means all the units in stock still have the bug, including the one that Amir received. I am not sure how Gustard will handle this update. I suppose they could release a new firmware that customers can download and install on their own, but if that was the case, why did they have to physically send L7audiolab a different unit with the new firmware? Anyway, I hope Gustard will fix this bug for every customer and not just for a technical reviewer with an APx555B analyzer.

Great insight. So All that glitters is not gold!!!

Your comments should be added to @amirm's review above.

 

[Nango]

Not again !!!!! The X16 had firmware bugs and the problem is amplified by the lack of an easy way to do a firmware update. With the X16 you needed to purchase equipment to attach to an internal circuit board after disassembling the unit.

Why don't they run the above tests themselves so they can catch these bugs before releasing?

Why should they do that if ASR does it for free?

 

[Nickelboride]

Why does the coax and toslink input of X18 have some jitter, while X16 does not?

 

[Dennis_FL]

I think they will, soon.

Beautiful rack. I didn't know they had those.

 

[StuartC]

Maybe its time for Gustard, SMSL or Topping to release something terrible, just to spice things up because right now its getting boring.

Maybe it's time to stop reviewing any of their products for a few months? We all know they'll measure excellent. The problem is they must make up 50%of DACs over the last 12 months.... At this rate the SINAD chart will be only their products before too long.

Unfortunately I'm not exactly in a position to be sending products for review due to shipping costs from the UK.

 

[Billy Budapest]

I can't believe my beloved Matrix Audio Element X is not even in the top 10 anymore....

It has essentially the same performance as the ones in front of it, though.

 

[doug2761]

You can get a better SINAD by setting volume to -2 dB on the machine's panel instead of your instrument. We integrated an analog attenuator which has better performance.

Reading this reminds me of something I read in the RME user group about their switch to ESS for the RME ADI-2. I recall RME saying that the ESS chip didn't handle overloads as gracefully as the AKM chip. To deal with that, they reserved about 2 db of headroom in the design. I prefer the solution be done that way instead of with the end user having to make adjustments. It's just one less thing for the consumer to think about.

 

[Nango]

Reading this reminds me of something I read in the RME user group about their switch to ESS for the RME ADI-2. I recall RME saying that the ESS chip didn't handle overloads as gracefully as the AKM chip. To deal with that, they reserved about 2 db of headroom in the design. I prefer the solution be done that way instead of with the end user having to make adjustments. It's just one less thing for the consumer to think about.

But if you gear ends up with only 125dB SINAD instead of 124.5dB it will most likely drop out from the Top 10 in the ASR guidance and treated like "garbage" by the audience. No matter human beings can hear only 120dB, if any.

 

[Nango]

Maybe it's time to stop reviewing any of their products for a few months? We all know they'll measure excellent. The problem is they must make up 50%of DACs over the last 12 months.... At this rate the SINAD chart will be only their products before too long.

Unfortunately I'm not exactly in a position to be sending products for review due to shipping costs from the UK.

Ohhhhh I second that!!!!

 

[GWolfman]

Worth taking a look! Thanks!

 

[solderdude]

Great insight. So All that glitters is not gold!!!

Your comments should be added to @amirm's review above.

Not needed, as Gustard already replied the issue does not exist in production models.

 

[Skinner001]

Ah, another thing that does the thing that can't be distinguished from many other things, including the previous thing in the series of things (the x16)

 

[HiFidFan]

You are right bro, there's no 'perfect software', but we are always trying to do the best in development. Big bugs were killed before batch production.
Once new bugs noticed by users, we will keep tracking and solve it as soon as possible. All the firmware updates would be released and stored on Internet. This issue of X18 is noticed before batch productions' delivery, so everyone of you can buy the fixed one.

The support will last at least one year after production stopped.

Am I the only one that feels that a comment like this gives the air of "Beta test" for the early adopters?

Excellent measuring DAC and thanks for the review @amirm.

 

[TK750]

Am I the only one that feels that a comment like this gives the air of "Beta test" for the early adopters?

Excellent measuring DAC and thanks for the review @amirm.

As is the case for every consumer product ever, especially within technology and as things have increased in complexity. Regardless of how extensive your internal and possibly even external testing is, it is always going to be orders of magnitude smaller than the amount of consumers that test it in the real world.

 

[aj625]

Great !

 

[Nango]

Not needed, as Gustard already replied the issue does not exist in production models.

As @X_=& mentioned the probability is high that stock items might be affected.

 

[Nango]

As is the case for every consumer product ever, especially within technology and as things have increased in complexity. Regardless of how extensive your internal and possibly even external testing is, it is always going to be orders of magnitude smaller than the amount of consumers that test it in the real world.

"Complexity" like purchasing the analyzer and do the measurements. Not very 'complex' at all.

 

[Carlo2AC]

You are right, though i did not understood very well.. maybe it was the language barrier?

You are able to upgrade its firmware without needing a third party tool.. or to open it up and use a programmer.. or do you have to send it back?

 

[solderdude]

As @X_=& mentioned the probability is high that stock items might be affected.

He is wrong about that... The manufacturer clearly stated that this is not the case for all sold ones. Only the prototype measured by Wolf had this and was corrected. What more can be said about this ?

 

[TK750]

"Complexity" like purchasing the analyzer and do the measurements. Not very 'complex' at all.

?? I thought it was quite clear I was referring to increasing complexity within technology and consumer products, I'm not sure you've understood what I was saying. For example an iPhone is much more complex than a 3310.