Review and Measurements of Sound BlasterX G6

[MCH]

hello gang,
I finally have my G6 up and running.
There is though something i am struggling with. I want to use it as external card to a raspberry pi running camilla and the ***ard, no matter what configuration i set up (no direct mode, direct mode, direct spdif) outputs the sound as is without going through camilla processing... Camilladsp seems to be working ok but you can see that the output is not affected by the filters/mixers... it is just acting like a passthrough...
any experience with this or in general with the G6 in linux?
thanks!

 

[AudioManNewb]

If there any advantage or reason on PC to run the G6 through optical with USB as the power source?

 

[ZolaIII]

If there any advantage or reason on PC to run the G6 through optical with USB as the power source?

As separate power source yes you eliminate power supply EMI by using optical (you get small increase in performance anyway and potentially won't have problems if it's a noisy power supply).

 

[AudioManNewb]

As separate power source yes you eliminate power supply EMI by using optical (you get small increase in performance anyway and potentially won't have problems if it's a noisy power supply).

K have it all connected up. In the G6 SPDIF set to 87% Only question I have should volume be controlled through my onboard realtek digital output or the G6's volume?

 

[ZolaIII]

K have it all connected up. In the G6 SPDIF set to 87% Only question I have should volume be controlled through my onboard realtek digital output or the G6's volume?

Use what's more comfortable to you (both are digital volume controls).

 

[AudioManNewb]

For those interested I have the G6 for sell with all original accessories manual and original box. Selling $140 with all fee's and shipping covered. Shoot me a DM.

 

[Knoonan]

Amir, I'm working on a non audio project, was thinking of buying the G6, and found your review just today. In my project I am looking for best SNR for analog input. In your testing I was particularly interested in your test of the ADC, and of course disappointed in the result. But I'm confused how you actually tested. I haven't read all 30 pages of this thread, so please forgive me if you have already addressed this. It can be confusing with terms of DNR and SNR, DAC and ADC, but the front end analog input amplifier for the ADC will set the limit for SNR input. Not the number of bits or sample rate.
You said you used Toslink input, but Toslink is optical digital audio. So how can you use digital audio as an input to an analog to digital converter? I must be missing something. Can you describe in more detail?
For my application, a useful test would be to simply measure the noise floor with no input, just a resistor of the same value as intended system. Does anybody do that? What would it be called?
Thanks for comparison to other top audio devices. I am still searching for VERY good analog input, and the RME device looks great but $2K is pretty steep. Still looking, open for suggestions. I will check those others you listed.

 

[IBJamon]

Amir, I'm working on a non audio project, was thinking of buying the G6, and found your review just today. In my project I am looking for best SNR for analog input. In your testing I was particularly interested in your test of the ADC, and of course disappointed in the result. But I'm confused how you actually tested. I haven't read all 30 pages of this thread, so please forgive me if you have already addressed this. It can be confusing with terms of DNR and SNR, DAC and ADC, but the front end analog input amplifier for the ADC will set the limit for SNR input. Not the number of bits or sample rate.
You said you used Toslink input, but Toslink is optical digital audio. So how can you use digital audio as an input to an analog to digital converter? I must be missing something. Can you describe in more detail?
For my application, a useful test would be to simply measure the noise floor with no input, just a resistor of the same value as intended system. Does anybody do that? What would it be called?
Thanks for comparison to other top audio devices. I am still searching for VERY good analog input, and the RME device looks great but $2K is pretty steep. Still looking, open for suggestions. I will check those others you listed.

I can't speak to the measurements questions, but look at the Focusrite Scarlett series. They have excellent ADCs for a very good price, as they are basically mic/instrument interfaces first (and quite good DACs on top of that). The headphone amps do have a bit to be desired though.

 

[Knoonan]

I can't speak to the measurements questions, but look at the Focusrite Scarlett series. They have excellent ADCs for a very good price, as they are basically mic/instrument interfaces first (and quite good DACs on top of that). The headphone amps do have a bit to be desired though.

Thanks for the tip. I just found Amir's review of that. It does look good. Especially at that price.

 

[Berwhale]

@Knoonan - The E1DA Cosmos may be of interest...

E1DA Cosmos ADC

@IVX kindly sent me a sample of the new Cosmos ADC to try out. Here are my impressions of it so far. The Cosmos is housed in a compact metal case with rubber feet, USB Type C connector on one and dual XLR inputs on the other. Power is over USB only. Beside the USB connector is small LED that...

www.audiosciencereview.com

E1DA Cosmos ADC teardown

I recently purchased a Cosmos ADC from @IVX and I was pleased to see that the main envelope, from inside the package, had my name written on it in handwriting with a pencil, probably by Sunny, the nice lady that takes care of logistics for Ivan (@IVX). That put a big smile on my face immediately...

www.audiosciencereview.com

 

[Knoonan]

@Knoonan - The E1DA Cosmos may be of interest...

E1DA Cosmos ADC

@IVX kindly sent me a sample of the new Cosmos ADC to try out. Here are my impressions of it so far. The Cosmos is housed in a compact metal case with rubber feet, USB Type C connector on one and dual XLR inputs on the other. Power is over USB only. Beside the USB connector is small LED that...

www.audiosciencereview.com

E1DA Cosmos ADC teardown

I recently purchased a Cosmos ADC from @IVX and I was pleased to see that the main envelope, from inside the package, had my name written on it in handwriting with a pencil, probably by Sunny, the nice lady that takes care of logistics for Ivan (@IVX). That put a big smile on my face immediately...

www.audiosciencereview.com

Excellent! I love it, thank you.

 

[Berwhale]

Excellent! I love it, thank you.

Welcome to ASR!

 

[Paradigm5h1f7]

This is a review and detailed measurements of the Sound BlasterX G6. I purchased it last month for USD $150 from Amazon including shipping.

Typical of PC audio connectivity devices, the Sound BlasterX G6 is a combo device which includes an audio DAC, ADC, and headphone amplifier. This being targeted towards gamers, there are a lot of specific features in that regard which I did not test. My focus here is the pure audio performance.

The Sound BlasterX G6 looks better in pictures than in person due to lightweight, plastic case:

View attachment 23401​

No, there are no batteries. The unit is USB powered and I used photoshop to erase the USB cable. I hope those of you picking on my photography are happy!

The volume control is a rotary encoder so turns forever in either direction. There are no notches but overall feel/stiffness is not bad. I did however find that there was a lag between turning the dial and my Roon player registering the change.

As you see, the front panel has headphone out and microphone input. The back end has nice combo digital and analog 3.5mm jacks:

View attachment 23402​

Creative nicely includes the 3.5mm mini Toslink cable which was a good thing as I once again lost mine.

There are headphone gain button, one called SBX and the other "SCOUT" which if you hold down, causes an annoying blinking white LED to light up indicating you are bypassing all the effects.

As usual, there is a crapware called Sound Blaster Connect 2 to control the G6 from the PC. I found it attractive unlike most out there and functional in some cases such as selecting the digital audio filter type:
View attachment 23403

There is also a nice mixer to control what is happening although it was hard to figure out at times.

Anyway, let's get into the measurements and see how the G6 performs.

DAC Measurements
I initially ran into lots of difficulty measuring the G6. Even though Windows and ASIO4ALL wrapper would recognize it, the latter could not send it any signal. Nothing would happen. I downloaded a driver package from Creative website but that didn't fix anything. I don't know if it installed an ASIO driver or I had one from testing previous Creative products. Selecting that gave me audio but horrendous results. Even simple things like frequency response were all over the place.

As a workaround I played my 1 kHz tone through Roon player to G6 which worked fine so I knew the problem was the ASIO driver:
View attachment 23404

I must stay, I did not expect the results to be this good! Distortion products are kept to -120 dB so what we see in SINAD must be noise dominated.

As with the 1 kHz tone, I also played the J-test signal manually to get our jitter and noise response:

View attachment 23405

It is not picture perfect and we have the clear signature of jitter in the symmetrical 4 and 20 kHz indicating that the jitter frequency is 8 kHz (12-8 = 8; 12+8 = 20). Levels are below 120 db.

Problem with manual testing like this is that I can't run the full suite of tests. Then it hit me that there was Toslink input so I used that for balance of my testing to good results:
View attachment 23406

Toslink input is a bit cleaner than USB. What is even more interesting is that if I dialed down the level by 2 dBFS (digitally), SINAD would rocket up to 112 dB! That would drop the output voltage to below 2 volt so for the purposes of comparison, I used the full dB value in our SINAD chart:

View attachment 23407

Creative sells its audio products seeming by signal to noise ratio, usually boasting unbelievable numbers. The G6 marketing material was no exception:
View attachment 23408

We are getting almost 120 dB here which is excellent so not sure why they play games to advertise 130 dB.

Frequency response was nice and flat:
View attachment 23409

There is a lot to show in this review so I am omitting the usual filter response for now.

Linearity is nailed, absolutely nailed with flat and noise-free response:
View attachment 23410

Unfortunately problem was waiting for us around the corner in the form of intermodulation distortion:

View attachment 23411

What is going on here? The G6 was doing exceptionally well, beating my reference Topping DX3 Pro but then it shoots up like crazy. We have not seen this before in any DAC. And how is it that we are seeing clipping behavior here but not in the Dashboard which is at full amplitude just the same? The clue is in THD+N versus frequency (the dashboard is at 1 kHz):

View attachment 23413

Focus on the blue line first. Starting from right (higher frequencies) response is fine but the moment we get down to 300 Hz, distortion starts to shoot up and keeps getting worse and worse. By the time we get to 20 Hz, we are talking nearly 1% THD+N!!! A quick FFT (not shown) showed a spray of harmonic distortion at low frequencies. Dialing down the output by 2 dBFS completely fixed the issue.

The G6 is USB powered and likely doesn't have enough capacitance in its DC input to ride out the lasting peaks at low frequencies.

How did this get caught in the IMD test? It did because I use the SMPTE IMD test signal which consists of a pair of frequencies: 60 Hz and 7 kHz:

View attachment 23414

It was the 60 Hz tone that got severely distorted. In other words, the SMPTE signal not only shows intermodulation but also plain THD of low frequency tones. Those of you who advocate dual tone 20+21 kHz CCIF IMD would not have seen this benefit there.

Headphone Amplifier Measurements
Switching to headphone jack and measuring the power versus distortion using 300 ohm load we get this:

View attachment 23415

Considering that the G6 is purely USB powered, the power ratings in high gain mode are quite respectable. While I like to see more than 100 milliwatts, 85 is almost there so I expect good subjective performance in listening tests.

Low gain mode doesn't do much here seeing how it just stops early. With a digital volume control, I don't expect to have channel balance issues either (sorry, forgot to run that test).

Switching to 33 ohm load, the emphasis on current delivery causes more issues:

View attachment 23416

Power once again is respectable but now our THD+N has risen fair bit. This is all with 1 kHz tone so it is without the issues we saw in IMD test.

Output impedance is comfortably low:

View attachment 23428

Headphone Listening Tests
As usual, I started with my Sennheiser HD-650 and found the experience quite good. Mind you, there is no sensation of infinite power and you would be operating near or at max volume. But what is there is dynamic which is an issue I find with lesser powered headphone amplifiers.

Switching to Hifiman HE-400i on one loud track, I thought I was hearing static which is the signal of clipping. There was plenty of volume although the sound did get bright towards the end.

Overall, a good passing grade for most people who are not as picky as I am.

ADC Audio Measurements
Usually I am too lazy to go this far into testing a budget product but somehow I mustered extra energy energy last night and this morning to test the Analog to Digital (ADC) of the G6. Here is the dashboard with Toslink capture:

View attachment 23417

I was pleased that feeding the G6 2 volt, resulted in 0 dBFS showing no overflow. Performance though is not all that great with SINAD in the high 70s. We have lots of distortion components together with mains leakage.

Realizing we don't have good frame of reference and killing two birds with one stone with people asking for ADC measurements of Behringer UMC204HD, here is its dashboard:

View attachment 23418

I had trouble getting to 0 dBFS as the UMC204HD would severely distort there. Above is with 0.53 volts with analog gain controls set to minimum. Balanced input is used which nicely eliminates any hint of mains noise. And we are using USB interface that we like. SINAD performance though is not much better than Creative G6.

This is a good time to start our SINAD graph for ADCs:

View attachment 23419

Compared to high-end products, we are short some 40 dB! Definitely not splitting hairs here.

Hopefully over time we can fill out the table with bargain ADCs with excellent performance.

Conclusions
The Creative Labs BlasterX G6 delivers surprisingly good performance in this feature-filled package. It definitely erases my impression of them of shipping audio junk these days. This is all the more impressive given the USB powered nature of the device.

I am unhappy about lack of good ASIO support. This is listed on the website as supported but I could not find any ASIO drivers to download and what I have is badly, badly broken.

The ADC input is passable but don't look to it to provide transparency.

Overall for someone looking for a gaming sound card with good audio performance, they have it in Sound Blaster G6.

------------
As always, questions, comments, recommendations, etc. are welcome.

I will make this simple: you can never have enough money. So please consider donating 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).

Awesome review! Thank you.

Also, have you considered that the clipping might be from your USB power source? Many USB adaptors, and most PC USB ports are 500mA. It may be that the clipping would be removed with a higher current rated USB port, which on my ASUS motherboard requires drivers to implement, and not all ports support higher current ratings.

 

[usersky]

Awesome review! Thank you.

Also, have you considered that the clipping might be from your USB power source? Many USB adaptors, and most PC USB ports are 500mA. It may be that the clipping would be removed with a higher current rated USB port, which on my ASUS motherboard requires drivers to implement, and not all ports support higher current ratings.

It was addressed in the review if I remember well as a possible lack of reserved (capacitors). That may (or may not) be improved by shorter, thicker cable, high current port, increased voltage like some MB allow specifically for DACs (Gigabyte offers 5.1 and 5.2 voltage on USB port designed for DAC usage). All this are just hypotheses, not confirmed by measurements though. But G6 is really good and convenient for the price and features already the way it was reviewed.

 

[Paradigm5h1f7]

It was addressed in the review if I remember well as a possible lack of reserved (capacitors). That may (or may not) be improved by shorter, thicker cable, high current port, increased voltage like some MB allow specifically for DACs (Gigabyte offers 5.1 and 5.2 voltage on USB port designed for DAC usage). All this are just hypotheses, not confirmed by measurements though. But G6 is really good and convenient for the price and features already the way it was reviewed.

I think you are onto something. I thought the same thing, and went back today to re read the review and noticed your note on needing more capacitance on the DC input, especially since it is a low frequency issue. Would be an interesting test to try a different charger, like a phone charger, which will have a larger DC capacitance, unlike the USB ports on a PC which share a small cap between all devices.

 

[Paradigm5h1f7]

It was addressed in the review if I remember well as a possible lack of reserved (capacitors). That may (or may not) be improved by shorter, thicker cable, high current port, increased voltage like some MB allow specifically for DACs (Gigabyte offers 5.1 and 5.2 voltage on USB port designed for DAC usage). All this are just hypotheses, not confirmed by measurements though. But G6 is really good and convenient for the price and features already the way it was reviewed.

Hey, I think I found the issue. I bought myself a G6 to try it out, and during my research (I am an EE) I pulled up the datasheet for the Cirrus Logic DAC. It has a pin for "charge pump" to utilize a Class H designed amplifier. I am curious to see if pins 25 and 26 on the DAC (for charge pump) are connected. If so, when the volume is turned up past 90% the charge pump enables for higher voltage in the amplification stage, and this might (likely will) further increase the PSU requirements. I am thinking the PSU / Class H design of the Cirrus Logic chip, is indeed the issue you noticed. The charge pump circuit should output up to 3V, rather then the 2V measured.

 

[Paradigm5h1f7]

It was addressed in the review if I remember well as a possible lack of reserved (capacitors). That may (or may not) be improved by shorter, thicker cable, high current port, increased voltage like some MB allow specifically for DACs (Gigabyte offers 5.1 and 5.2 voltage on USB port designed for DAC usage). All this are just hypotheses, not confirmed by measurements though. But G6 is really good and convenient for the price and features already the way it was reviewed.

You might find this interesting from the datasheet. Mentions if the ADPT PWR pins are set to 011, strict Class AB. Class H set to 111, and will result in clipping until Class H takes over.

Ironically, it says for about 5.5 seconds, it waits at clipping to ensure the IC won't continually toggle between Class AB and Class H making sure the added voltage is needed. If you are using an oscope which is capturing the first rising edge signal, it will indeed result in clipping if Class H is enabled. This would also mean they might be getting that 130db SNR?

 

[usersky]

I just don't want to chase that perfection. G6 really is stellar considering performance/price/features. Moreover, if you need to listen to absolute volume over 90, you probably need a more powerfull hp amp. Even so, distorsions at low frequecies need to be really high, and i mean like10% to be noticeable. So just enjoy it, it's practically perfect Your observation is indeed VERY interesting, this may be a transitory aspect of it, just while deciding to trigger to another mode/class. Really wow if just in theory!

 

[Paradigm5h1f7]

I just don't want to chase that perfection. G6 really is stellar considering performance/price/features. Moreover, if you need to listen to absolute volume over 90, you probably need a more powerfull hp amp. Even so, distorsions at low frequecies need to be really high, and i mean like10% to be noticeable. So just enjoy it, it's practically perfect Your observation is indeed VERY interesting, this may be a transitory aspect of it, just while deciding to trigger to another mode/class. Really wow if just in theory!

Agree whole heartedly. I have an old school Yamaha AVR with a 24bit DAC that is so so. The purchase of the G6 is simply experimental to see if there is an audible difference when using it to decode instead. I also purchased a newer Yamaha WXA-50 I will be experimenting with this on. If it doesn't work out, I will be using it on my gaming computer, so no loss. I just like to experiment, especially with the hype of DACs these days.

Insofar as much as class H, I was surprised to see it here, and am a huge fan of the technology. I've been into audio since the 90's and BASH was my favorite. It never took off because of the license required but I have 3 auto amplifiers (Blade SE's and Grenade) which used a class H topology. They used a PWM power stage (essentially class D), which mirrored the class AB music stage giving AB sound, with class D efficiency. The amplifiers I have are about 1"x8"x6" in dimensions and never get hot to the touch even with small heatsinks even though they are rated for 250w x 2 at 0.05% distortion. The patent expired last year, so I am hoping for a revival, but who knows? BASH technology was used in some home audio Klipsh, Technics, SVS, and Infinity gear, but never really hit mainstream. My computer speakers are the first model Klipsh ProMedia 4.1 which were BASH, but have since been replaced with lower quality gear.

If you are interested, here is a video on the Blade SE amplifiers from the 90's. You will notice that the second test the amplifier reproduced the sound with a 72% efficiency.

 

[usersky]

BASH was Technics's, wasn't it? Those were the days of our lives...