I have not noticed any pops, cracks, echo distortion (distortion that follows right after original sound) or other general distortion. There was one issue with digital noise that increases in decibel with mouse movement. This is a very old issue with digital audio, especially for the desktop PC users. Luckily the community at Head-Fi.org found a fix, rather quickly in fact! New user “Dżonel” first suggested this simple mod and a handful of other users confirmed the fix works, including myself (you can find more on that here).
What you want to do is remove the USB’s ground connection all together. The DAC still has a ground regardless, it’s either from the 12V Molex or the normal DC power input, whichever you are using gives the +5V its return path to ground. We know there must be one by KVL, Kirchhoff’s Voltage Law, which says all voltage drops & voltage rises in a loop, must sum to zero volt. This mod simply negates a 2nd (and thus optional) ground of the system. You always want to do this, especially for audio circuits that can amplify hum from a ground loop. If you cut open your USB cable and clip the black ground wire, this noise will cease, or at the very least, drop in decibel by 75%. Some XMOS modules require the GND from the USB cable, usually those without a grounded chassis, so pick the cable you cut up wisely! You can also kill the USB ground at the DAC, but this is a more permanent option.
I’m going to say that the high current drawing parts, such as the CPU and GPU, highlight the dual path to ground problem whenever they are under a heavy load. There is likely increased noise appearing on the ground, during this scenario.
The issue of mouse induced noise/artifacts is an issue that I have brought up to companies like Creative before the X-Fi series ever launched (over a decade). I am of the opinion that it has to do with high amperage draw GPUs and CPUs. There is the possibility that a noisy AC line/outlet could contribute as well. You can minimize this chance by minimizing the number of active components in your PC (and in your audio signal chain, start with a small, but high quality system). With regard to minimizing the number of components, I am speaking about the non – system critical hardware (as in don’t remove your CPU/RAM/Mobo/PSU). When we are dealing with high res audio, this noise usually isn’t an issue, but when dealing with a lower quality (like streaming) it sometimes is. I have also picked up on similar distortion when converting multi-channel audio to stereo with the Play. When noise on the digital line is an issue, the ES9038P is almost always the winner. According to “The Art of Digital Audio 3rd Rev.”, the ability of a DAC to reject noise/jitter/artifacts/aliasing/etc.. is the “greatest measurement of DAC performance”. But as we have already mentioned, it was simply a ground issue and was easily fixed, so not everything is so cut and dry.
I’ll do more tests using compressed audio tracks vs. high res audio to see if I can duplicate it. I found out that the ES9038P is able to do gain compensation dynamically and as a result, lower the THD (Total Harmonic Distortion). I am building a 2nd PC for Hallman Labs, nearly identical to the main PC except this one has an i5 3570K de-lidded (heat spreader removed and higher quality paste added) w/ Cooler Master V8 GTS + ASRock Z77 Extreme6 + G.Skill DDR3 2133MHz 16GB (4x4GB) + Corsair HX750 (750W/62A) + Corsair 650D rig, running the Linux tailored to Hi-Fi, Audiophile Linux (“AP-Linux” link here). This will allow us to see if this is an OS related issue, or truly just a hardware grounding problem (well, in theory it should allow us).
From the ESS datasheet “(The ES9038P has) Programmable THD compensation to minimize the THD caused by external components” and “Full-Scale manual/auto-gain calibration reduces device-to-device gain error.” This is exactly what Burson Audio said is causing the noise over a digital signal, gain offsets between devices. Alex said the next revision of the Play will be able to adjust this manually. Now we know after troubleshooting, the cause of our digital noise has to do with the ground in the USB cable.
I also had to use a standalone power supply from a computer with a 24 pin lab PSU module with an on/off switch to give the 12 V Molex power. I used a Corsair HX750 (high end PC PSU) for this to ensure the cleanest possible power (nothing else is drawing power, even the fan only runs when necessary).
To make a long story short, I had a bad power brick and I ended up preferring how the PC power supply sounded over the replacement I got. I estimate having over 400 hours of use on the new revision of the Play. Even when putting it up against stout competition, the Play’s V6-OPA “Classic” op-amps really shine, especially as the amplifier. In fact, doing a direct comparison with just the Play and its internal amplifier vs. the entire other setup, the Play isn’t that far behind (using its own amp vs. the ES9038P on RCA out to sAp-10)!
Table of Contents:
- Review Main Page
- Competition Hardware Descriptions:
- Review Hardware Info v2.0 vs. v1.6 Burson Audio Play:
- Digital Noise/Mouse Distortion + Solution: (USB Ground Headaches)
- Opening Thoughts on Voicing of Burson “Play” vs. ES9038Pro:
- Listening Tests Streaming: ES9038 on Coax + RCA out
- A Closer Look at Oversampling & Noise Filtering: (It’s Not Magic!)
- Listening Test Music (Pt 1): (Adele with ES9038P on I2S)
- Listening Test Music (Pt 2): (Tedeschi Trucks Band with ES9038P on Coax)
- A note on Converting PCM to DSD with Foobar2000: (DSD64 Clicking)
- Balanced vs. Unbalanced: (Understanding the ES9038’s Performance)
- Adding I2S Input To ES9038 DAC: (Power Tools and Soldering Irons)
- Measurements/Technical Page
- Closing Remarks: (The Verdict)
Copyright © 2018 Hallman Labs™, All rights reserved. No material written or photographed on this site may be replicated/copied/quoted/etc. without written permission from the site author/owner of Hallman Labs.