Detailed THD(+N) vs. Output Level Measurements of ESS ES9039Q2M. Balanced output quality from Hidizs AP80 PRO-X DAP. And Tekton's self-inflicted PR disaster.

Hey everyone, I wanted to post a follow-up of sorts to a couple of previous articles. First, the article "Hi-Res THD(+N) vs. Output Level Measurements (ESS "HyperStream" vs. AKM vs. TI/Burr-Brown). And a bonus R-2R!" was published back in the summer of 2022 showing an interesting sinusoidal pattern in the harmonic distortions with ESS DACs.

Since then, with the release of their newest ES9039 DACs which employ their latest "HyperStream IV" modulator, I thought it'd be cool to have another look...

Another follow-up is a look at the 2.5mm balanced output from the Hidizs AP80 PRO-X Music Player which was measured last year. At that time I didn't do the balanced output measurements so let's capture a few metrics and compare this with the single-ended 3.5mm output.

We can then end off with some audiophile social commentary from this week's drama.

Hi-Res THD(+N) vs. Output Level from ES9039Q2M vs. ES9038Q2M DAC

As you may recall from the previous article, I found something interesting with the ESS DACs which is the presence of sinusoidal variations in the harmonics across output levels when we examine the stepped-sine test closely. For example, here's the output from the SMSL DO100 which features dual ES9038Q2M chips (I see there's an updated SMSL DO100 PRO with dual ES9039Q2M and HDMI now, nice):

We saw this kind of pattern among the HyperStream II architecture chips like the ES9038PRO as well as older HyperStream (I) like the ES9018Q2C. In contrast with the ESS chips, the AKM DACs and TI/Burr-Brown products did not show this pattern.

So, how did the newest ES9039Q2M perform? (Used the E1DA #9039S pre-production DAC as recently measured.)

Notice I put the cursor at -6dBFS and -120dB to show the relative resolution of the DACs; clearly the ES9039Q2M achieving better performance.

Nice to see that the new HyperStream IV DAC does not have that sinusoidal harmonic distortion pattern!

Pretty cool, eh?

Of course, no need to freak out or get too excited... I don't really believe that this sinusoidal pattern was audible with the older generation ES90XX chips given how low distortion is already. More than likely outside of the ability for humans to hear. Nice to see objective improvement in unexpected patterns like this nevertheless across the generations of products!

---------------------

Hidizs AP80 PRO-X Digital Player - Balanced Output quality

Another quick follow-up is for the Hidizs AP80 PRO-X digital audio player reviewed back in June 2023. It's a nice little portable player which measures quite well (more than good enough for on-the-go listening!).

However I did not measure the balanced 2.5mm headphone output back then so let's have a peek at how it does with a few measurements and graphs.

I. Peak output level & 1kHz 0dBFS THD+N, unloaded

Peak output - Lo-Gain = 2.1Vrms, Hi-Gain = 4.16Vrms

Nice suppression of even harmonics.

With single-ended output measured last year, Lo-Gain = 1Vrms, and Hi-Gain = 2Vrms peak. The balanced output is therefore able to double the output level (+6dB); expected. As discussed previously, balanced headphone output is really about increasing power through "differential drive" (bridging amplifiers); depending on the amplifier quality, there's not necessarily an improvement in distortion or noise level.

 

We can see that pushing the player to "Hi-Gain" drops our THD+N slightly in this "no load" (to be exact, 36kΩ using the RME ADI-2 Pro FS) condition even though the output level increased by 6dB. 

II. Output impedance

Balanced output impedance is generally about double of single-ended assuming about the same circuit design. As expected then, I'm seeing an increase, measured as just shy of 0.3Ω which is higher than the 0.1Ω with the single-ended 3.5mm phono jack. Good that the impedance is flat across the audible spectrum.

This audio player will have no problem with any normal headphone impedance.

III. 1kHz THD(+N) vs. Level into representative headphone loads

This is meant to be a portable headphone audio player so let's have a look at power under representative headphone-like loads. I used the E1DA Load Board set to balanced mode at 20.5 / 63 / 300Ω for simplicity:

I typically estimate power levels with "hi-fi" distortion threshold at 0.1% THD hence at 20Ω we get 126mW, at 63Ω 153mW, and into 300Ω 58mW. It's a slight improvement over single-ended where into 20Ω I got 98mW @ 0.1% THD. There was a greater gain in the mid-impedance for headphones say between 60-80Ω where at 63Ω we can achieve up to 3Vrms of non-distorted output.

At 1% THD which is how most manufacturers rate their devices, we see 160mW into 20Ω. The official Hidizs specs say 190mW into 32Ω, presumably at 1% THD which is probably consistent with my findings using the lower impedance.

Summary:

Nothing surprising with the Hidizs AP80 PRO-X (about US$160 these days) balanced output measurements. If you need the extra power, by all means, use the 2.5mm balanced jack with "Hi-Gain". Don't expect any significant lowering of distortion however - 1kHz THD+N at best around -96dB like with single-ended which is excellent for a mobile device.

(For portable balanced headphone output with high-fidelity resolution into low-impedance loads, the E1DA #9039S is the one to beat with better-than -120dB 1kHz THD+N into 16Ω, excellent multitone performance.)

--------------------

Eric Alexander's Self-inflicted PR disaster...

To end, it was quite the drama this past week with the company Tekton Design and the owner Eric Alexander's hotly litigious threats against those who publish reviews of his products; especially independent objective measurements like on Audio Science Review (the M-Lore speakers) and Erin's Audio Corner. Notice that Tekton doesn't seem to mind less technical reviews like this that challenge underlying core selling points about the purported benefits of the "patented tweeter array" midrange.

Here's Erin's final take on the Tekton Troubadour speakers:

Obviously, it's not the video review itself that was particularly unique (Erin actually enjoyed the speakers, and I suspect it would not have been especially attention-grabbing), but Mr. Alexander's unreasonable approach with his impulsive threats of litigation. He then puts up a YouTube video a few days later saying that he wasn't going to sue anyone with his idiosyncratic view of what he meant by "litigation" (some kind of friendly inter-lawyer chat?), namedropping of Wilson Audio, among other oddities, video since withdrawn. Obviously he must have felt the pressure from the audio community to have backpedaled like this after news of his behavior went public. We might remember 5 years ago there was some brouhaha with a New Record Day's review (long since withdrawn, a summary here of the madness) of one of their speaker models as well. 

It's really good to see that despite all the craziness, frauds, and snake oil in this hobby, there is appreciation from the public that objectivity expressed in good faith is valued. Obviously, actors such as Eric Alexander responding emotionally with bullying tactics rather than engaging at the level of rational objective analysis showing his own measurements or other data is rightly seen as unreasonable, and his actions regressive among those interested in the pursuit of high fidelity. 

The irony to all this is that I don't think any technically experienced audiophile or DIY speaker hobbyist would imagine speakers like the Troubadour would measure without blemish. We might have even been a little surprised that Erin's original subjective assessment was reasonably positive. It's fascinating though that in Alexander's withdrawn video, he focused so much on the impedance graph and the small bump at around 250Hz (box resonance) as if this would have been a major issue impeding prospective consumers' desire for the speaker. Interesting because we're then left wondering whether it's out of personal insecurity that this seemed to be such a psychological injury or his own lack of appreciation for what is or is not meaningful when interpreting measurements!

Furthermore, now we all know that he drills holes through the speaker cabinet and depends on screw-in feet to plug them. 🤔 (Not surprising that plugging these small screw holes only resulted in minor measurement changes for the ported design.)

The only question now is how Tekton is going to move ahead. Since this whole chapter was escalated to public awareness due to Mr. Alexander's self-inflicted words and actions, what's the guy going to do now? He has clearly overvalued the usefulness of the few positive soft YouTube reviews and interviews out there and grossly underestimated the potency of his prideful (something about 18 awards), litigious character with repercussions to his company and those he employs. As the saying goes from Wayne Dyer: "Your reputation is in the hands of others. That’s what the reputation is. You can’t control that. The only thing you can control is your character."

[While I try to keep politics out of these pages it's hard not to make comparisons. I believe this is no different in a much smaller scale of course from a certain prideful ex-president who brings misery to himself and chaos to those around him, and will fail (again) at the elections later this year. Reasonable, conservative-leaning individuals also have limits as to what is unethical and unacceptable in the character of leaders possessing low-quality intellect.]

So, will we see the emergence of a more contrite spirit to rectify Mr. Alexander's image (nobody's beyond redemption, right?)? Will we see measurements for each new speaker coming out from this company as he promised in his retracted video? Supposedly the company has made 250 models in 19 years, a huge number considering a small company - how much R&D engineering time and money went into each one? I guess we'll get a sense based on the quality of the measurements to come, assuming they do come...

--------------------

Alright audiophile friends, that's all for now. Remember to have a listen to the "High-End" DAC Survey from last week and get me your impressions! Thanks to those of you who have submitted results already.

Enjoy the music...