Review and Measurements of Asus STX II PCI Sound Card

[amirm]

you do have to wonder how much each individual psu is going to affect things. pc psu quality can vary dramatically

I use a Corsair power supply (model escapes me). As I noted in my review, the whole reason to spend so much for an internal DAC is to get one that is immune to changes in the computer components. If it is sensitive to such things, then might as well get a much cheaper one or use the motherboard.

 

[amirm]

@amirm I suspect that maximum output volume is not optimal for this card. Would you try to test THS/SINAD again with volume set to 1Vrms (-6dBFS).

I tested it as I usually do at lower levels and it made no improvement. You can see this by the way in the IMD versus level:

If you look to the right, you see that there is almost no rise in distortion as we approach 0 dB.

 

[Sythrix]

It would be nice to see a review of the Xonar Essence One, just to see if there are any problems with that. It's an external model also designed by Asus. Unfortunately, I wasn't able to find it for sale, as it's been discontinued.

 

[wadec22]

I now have a topping d50 on the way. may send my creative x7 to amir to see how it does. I know people here don't care for the looks but man it has a ton of functionality.

 

[Thomas savage]

Brilliantly useful review mate thanks, Iv always wondered about these. Be intresting to test a few more imo.

 

[amirm]

@amirm could you show the FFT of 1kHz sine at -90dB and -110dB? Thanks.

Here it is at -90:

 

[amirm]

And -110 dB:

It is strange that the level does not go below -95 dB or so and hence the huge linearity error. I am using their ASIO driver.

I clicked on its control panel and it says this:

So it is running in 16 bit mode and no way to change it (option above is grayed out).

 

[amirm]

Here is the output with Roon running in exclusive WASAPI mode playing a 24 bit 1 kHz tone:

So no difference. Seems even in Wasapi mode it is running in 16 bit.

I went to its control panel and it was set to 96 kHz. Changed that to 44.1 and SINAD improved a bit:

So it seems that it is doing some kind of forced resampling?

I don't see a bit depth setting.

 

[amirm]

My sense right now is that whoever wrote the driver/software pipeline may have screwed up the performance of this unit. It is not like there are that many people who test for low level fidelity so this kind of bug/restriction would not get noticed.

 

[bennetng]

My sense right now is that whoever wrote the driver/software pipeline may have screwed up the performance of this unit. It is not like there are that many people who test for low level fidelity so this kind of bug/restriction would not get noticed.

Thanks, so the Asus driver is responsible for the poor dynamic range. The dashboard and both of the low amplitude sine graphs have no spurs below 1kHz so it seems that there is no power supply related pollution.

I don't own any Asus card but someone used an Asus soundcard as the measurement device and didn't realize he was not recording in real 24-bit as well:
https://hydrogenaud.io/index.php/topic,111411.msg918540.html#msg918540

the OP (theriverlethe) re-uploaded the audio files and measurement results after fixing the bit-depth problem and got more than 10dB DR improvement when measuring an SMSL and Schiit DAC.

Actually, Peter, foobar2000's author hates drivers from Asus:
https://hydrogenaud.io/index.php/topic,92442.msg779690.html#msg779690

 

[sonci]

I clicked on its control panel and it says this:

View attachment 16675

So it is running in 16 bit mode and no way to change it (option above is grayed out).

Maybe it's grayed out, because you can't change sample rate or bit depth when using Asio driver, it doesn't mean it's resampling anything..
still, latency should be able to edit, so who knows, I only used Emu soundcards.

 

[amirm]

I decided to test S/PDIF output over WASAPI and get this:

So WASAPI *is* running at 24 bit depth. When I test analog output though, I can't get a SINAD over 100. Here it is with volume control set to 98 to get 2 volt output:

So seems like we are fighting two problems:

1. ASIO always converts to 16 bits.
2. WASAPI doesn't but analog output is still limited to 100 dB SINAD.

 

[miero]

@amirm isn't there a high frequency noise that decreases SINAD? maybe ASUS used some low-pass filter on AP input to measure their 124dBA SNR.

Also they state[1]:
Output Signal-to-Noise Ratio (A-Weighted) (Front-out) :124 dB
Output Signal-to-Noise Ratio (A-Weighted) (Headphone-out) :120 dB

Are you measuring "front-out" or "headphone-out" ?

FIY: There is a way to avoid using ASUS drivers .... Linux

[1] https://www.asus.com/Sound-Cards/Essence_STX_II/specifications/

 

[amirm]

A-weighting definitely improves things. They did not say they used one though on THD+N (SINAD):

 

[bunkbail]

There is a 3rd party driver for Asus sound cards (UNi Xonar Drivers) that claims to fix the problems which are present in the standard Asus driver. Maybe the 24-bit mode can work properly with this driver?

 

[bennetng]

Just want to say even TI's datasheet says PCM1792A's THD+N is 0.0004%.

 

[janderclander14]

There is a 3rd party driver for Asus sound cards (UNi Xonar Drivers) that claims to fix the problems which are present in the standard Asus driver. Maybe the 24-bit mode can work properly with this driver?

I was about to propose the same. Asus and standard c-media drivers are known to be buggy and implement filters that degrade sound quality.
In this respect, the author of the Uni Xonar Drivers also implemented a tweak on these drivers that disables an internal filter and resulted in subjectively better quality.

http://maxedtech.com/xonar-disable-speakercompensation-tweak-testing/

 

[garbulky]

A-weighting definitely improves things. They did not say they used one though on THD+N (SINAD):

Amir there is an equalizer on the Asus XOnar panel. I think by default it is enabled. So when you get to the equalizer and play a soiund - if it shows the bars going up and down, it is enabled even when it is flat. You probably want it disabled.

 

[sonci]

There's also a hidden jumper, if you don't remove it, the card will explode after 15min!

There was a time when everybody blamed Creative drivers, but Asus seems worse..

 

[bennetng]

@amirm Could you measure the headphone amp and impedance as well? From what I read in the specs it has up to 7Vrms headphone output. Also the ADC part... yeah it is not a studio interface but the ADC also made the card pricier than a DAC-only device.