This is a review and detailed measurements of the Gustard DAC-X26 stereo audio dac. It is was kindly sent to me by Shenzhen Audio for testing. The DAC-X26 normally costs USD $1,299 but seems to be on sale at Shenzhen Audio for $1,169 as of this writing. Note that I have no relationship with either Shenzhen Audio or Gustard. And nothing was communicated other than wanting to send the unit in for testing.
The Gustard DAC-X26 is one heavy box. You can't tell it from the picture but it I promise you, it is:
I like the super clear display showing everything you need to know at a glance. There is a remote which I did not use for this testing.
The back panel has the inputs you expect including I²S input:
There are two independent linear power supplies with their own transformers and hence the two voltage switches and hefty weight of the unit.
Gustard has gone quite far in building their own digital filters using an Analog Devices Sharc DSP, discrete output low pass filter, etc. I let you read them in the Shenzhen Audio link. Truth to be told, when I hear companies doing their own custom implementations of such things, I worry that they are catering to audiophile addiction to technical terms than pure performance.
Another "worrisome" technical note is use of two ESS9038 Pro DAC chips in Gustard DAC-X26. The 9038 is the flagship DAC from ESS but testing of other DACs using it showed less, not more performance than products using lower tier DACs from their line.
Let's get into measurements and see if my worries are merited or not.
Measurements
As always, I start with our dashboard view. The output from the DAC-X26 was a bit high so I lowered it by 2 dB using the front panel control to get 4 volt in balanced and 2 volts in unbalanced. Here is what I got:
Wow. This is exceptional performance and lands the DAC-X216 the top of all DACs tested:
No other DAC regardless of price has produced better performance.
Unbalanced RCA performance was 2 dB lower as noted on the dashboard.
Also noted is the level of highest harmonic which is around -118 dB. The absolute best case dynamic range of our hearing assuming 120 dB SPL max playback level is -116 dB. By that definition, the harmonic distortion of Gustard DAC-X26 is provably inaudible!
As some of you know, the THD+N is the sum of harmonic distortion (THD) plus noise. Classical audio measurement tools couldn't separate these two but using signal processing, we can do that. So here is that break down for the DAC-X26:
In other words, performance is limited more by distortion than noise.
Speaking of noise, here is the dynamic range:
These are very low noise figures.
The incredibly low noise floor shows more jitter and spike distortions that would normally be visible:
Given all the talk in Gustard DAC-X26 I would have expected a cleaner display to please the eye. The ear though, would be super happy given the exceptionally low level of these spikes.
Multitone shows the low impact of harmonic distortion:
Linearity is nailed, absolutely nailed:
This is as perfect as it gets folks.
THD+N versus frequency shows very well behaved performance:
As an aside, I realized this test is run without the high-performance (dual ADC) feature of my analyzer. It used to not matter because the measurements were higher than what that feature is for. Not here. While I have left the above as I have always run it, selecting the high-performance analyzer lowered the distortion graphs further (not shown).
Frequency response was boring meaning as good as we want to see it:
Now the big question: do we have what I have coined the "ESS IMD hump" where distortion rises mid-level in intermodulation test? Let's see:
We do not! Some engineer at Gustard knows exactly what he is doing. There is no hint of rise in distortion at mid-levels or max for that matter. It has been a while since I have seen any DAC outperform our modestly priced Topping DX3 Pro. The DAC-X26 does that with both of its balanced and unbalanced outputs.
I will do more testing of the different filters later. For now, here are the default results starting with square wave:
And white noise response of the filter transition band:
As if fashionable these days, it is a very gentle filter, allowing fair bit of response above nyquist frequency 22.05 kHz.
Conclusions
What a pleasure it is to see a company talk about building a super high performance DAC in their marketing material and then proceed to deliver it! The Gustard DAC-X26 delivers state-of-the-art performance in every sense of the word. It is clearly designed by experienced engineers who have made sure the actual implementation matches the fancy specifications.
Needless to say, I am happy to give my strongest recommendation for Gustard DAC-X26. If you want a DAC that produces guaranteed transparency to your source content, you are there.
------------
As always, questions, comments, recommendations, etc. are welcome.
Crap! Snow came back yesterday. Need money to buy a second snow shovel. Please consider donating funds using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
The Gustard DAC-X26 is one heavy box. You can't tell it from the picture but it I promise you, it is:
I like the super clear display showing everything you need to know at a glance. There is a remote which I did not use for this testing.
The back panel has the inputs you expect including I²S input:
There are two independent linear power supplies with their own transformers and hence the two voltage switches and hefty weight of the unit.
Gustard has gone quite far in building their own digital filters using an Analog Devices Sharc DSP, discrete output low pass filter, etc. I let you read them in the Shenzhen Audio link. Truth to be told, when I hear companies doing their own custom implementations of such things, I worry that they are catering to audiophile addiction to technical terms than pure performance.
Another "worrisome" technical note is use of two ESS9038 Pro DAC chips in Gustard DAC-X26. The 9038 is the flagship DAC from ESS but testing of other DACs using it showed less, not more performance than products using lower tier DACs from their line.
Let's get into measurements and see if my worries are merited or not.
Measurements
As always, I start with our dashboard view. The output from the DAC-X26 was a bit high so I lowered it by 2 dB using the front panel control to get 4 volt in balanced and 2 volts in unbalanced. Here is what I got:
Wow. This is exceptional performance and lands the DAC-X216 the top of all DACs tested:
No other DAC regardless of price has produced better performance.
Unbalanced RCA performance was 2 dB lower as noted on the dashboard.
Also noted is the level of highest harmonic which is around -118 dB. The absolute best case dynamic range of our hearing assuming 120 dB SPL max playback level is -116 dB. By that definition, the harmonic distortion of Gustard DAC-X26 is provably inaudible!
As some of you know, the THD+N is the sum of harmonic distortion (THD) plus noise. Classical audio measurement tools couldn't separate these two but using signal processing, we can do that. So here is that break down for the DAC-X26:
In other words, performance is limited more by distortion than noise.
Speaking of noise, here is the dynamic range:
These are very low noise figures.
The incredibly low noise floor shows more jitter and spike distortions that would normally be visible:
Given all the talk in Gustard DAC-X26 I would have expected a cleaner display to please the eye. The ear though, would be super happy given the exceptionally low level of these spikes.
Multitone shows the low impact of harmonic distortion:
Linearity is nailed, absolutely nailed:
This is as perfect as it gets folks.
THD+N versus frequency shows very well behaved performance:
As an aside, I realized this test is run without the high-performance (dual ADC) feature of my analyzer. It used to not matter because the measurements were higher than what that feature is for. Not here. While I have left the above as I have always run it, selecting the high-performance analyzer lowered the distortion graphs further (not shown).
Frequency response was boring meaning as good as we want to see it:
Now the big question: do we have what I have coined the "ESS IMD hump" where distortion rises mid-level in intermodulation test? Let's see:
We do not! Some engineer at Gustard knows exactly what he is doing. There is no hint of rise in distortion at mid-levels or max for that matter. It has been a while since I have seen any DAC outperform our modestly priced Topping DX3 Pro. The DAC-X26 does that with both of its balanced and unbalanced outputs.
I will do more testing of the different filters later. For now, here are the default results starting with square wave:
And white noise response of the filter transition band:
As if fashionable these days, it is a very gentle filter, allowing fair bit of response above nyquist frequency 22.05 kHz.
Conclusions
What a pleasure it is to see a company talk about building a super high performance DAC in their marketing material and then proceed to deliver it! The Gustard DAC-X26 delivers state-of-the-art performance in every sense of the word. It is clearly designed by experienced engineers who have made sure the actual implementation matches the fancy specifications.
Needless to say, I am happy to give my strongest recommendation for Gustard DAC-X26. If you want a DAC that produces guaranteed transparency to your source content, you are there.
------------
As always, questions, comments, recommendations, etc. are welcome.
Crap! Snow came back yesterday. Need money to buy a second snow shovel. Please consider donating funds using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
