Hidizs S9 PRO (short) Measurements
Hi folks,
Here is a
quickreview, measurements-based, of the Hidizs S9 PRO portable USB DAC/Dongle. It has been kindly sent to me by a French fellow. The S9 PRO has been introduced only few months after the S9 (1st gen), that Amir reviewed a year ago. As opposite to the previous model, which used AKM AK4493, the "PRO" now comes with ES9038Q2M from ESS. No doubt, the change has been closely related to the AKM shortage that started late 2020. It costs 109€, tax incl as of this writing.
The Hidizs S9 Pro is a very light (11g) and tiny unit (18 x 59 x 8mm) made out of aluminum and glass. It has no physical volume control, the latter being done in software domain. The logo is backlighted and indicates sampling rates using different colors. You see two outputs: one 3.5mm TRS (SE) and one 2.5mm TRRS (BAL). Due to lacking of both correct cables and personal time, I only tested the balanced output for today's review.
Only input you will find is (OTG) USB-C. Note that, being a dongle, the S9 Pro is not self-powered but feed by USB. Besides, be aware that it can run quite hot and may drain your phone battery fairly fast when used on the go...
Measurements
Disclaimer: Measurements you are about to see are not intended to be as precise or extensive than what you get from a 30k€ AP. There is obviously both hardware and software limitations here, so not quite apples to apples comparison with Amir's testing. Still, this data is enough to have a pretty good idea if the gear is bad or not, stellar, broken, or sub-par...
- ADC : E1DA Cosmos (Grade B), set to 4.7Vrms input. Minimum phase filter.
- Software : RMAA 6.4.5 PRO, Multitone Loopback Analyzer 1.0.15. After countless requests, I now measure unweighted THD+N with REW V5.20.7.
- Method : 8 runs for each test, then I choose the closest to the average. All regular tests are running 24bits / 44.1Khz except for Jitter and Multitone ones.
- Hidizs S9 Pro: BAL TRRS output at max volume, measured at 4.254Vrms @1Khz for both channels with a DMM.
- Main limitation : lack of load board to measure output at different impedances.
- Teasing for reviews to come : @IVX sent me a preproduction Cosmos APU, that is intended to improve both THD+N and AES DR accuracy in REW.
Summary
| Test | Results | Rating |
| Frequency response (from 40 Hz to 15 kHz), dB | +0.03, -0.23 | Very good |
| Noise level, dB(A) | -122.7 | Stellar |
| Dynamic range, dB(A) | 122.4 | Stellar |
| THD, % | 0.00041 | Excellent |
| THD + Noise, dB(A) (RMAA) | -103.1 | Excellent |
| THD + Noise, dB (REW) | -104.6 | Excellent |
| IMD + Noise, % | 0.00060 | Excellent |
| Stereo crosstalk, dB | -123.4 | Stellar |
| IMD at 10 kHz, % | 0.00050 | Excellent |
We are clearly landing in the desktop-class territory. It is in fact even better than many huge DACs sold by classic manufacturers. Main bottleneck here appears to be absolute THD, which is not the best I've seen and explains SINAD of about 105dB, despite superb Noise figures. It indeed performs several dB worse than the previous model measured by Amir. One hypothesis could be its output impedance being too low to be properly measured by the Cosmos... These are still excellent results, even more when talking about a dongle. Note that THD+N measured at Max volume matches specs quite accurately (0.0006%).
REW FFT
Frequency response
Left | Right | |
| From 20 Hz to 20 kHz, dB | -0.48, +0.03 | -0.48, +0.04 |
| From 40 Hz to 15 kHz, dB | -0.23, +0.03 | -0.23, +0.03 |
Noise Level
Left | Right | |
| RMS power, dB | -120.9 | -121.6 |
| RMS power (A-weighted), dB | -122.3 | -123.1 |
| Peak level, dB FS | -87.1 | -86.9 |
| DC offset, % | -0.0 | +0.0 |
Dynamic range
Left | Right | |
| Dynamic range, dB | +121.1 | +121.8 |
| Dynamic range (A-weighted), dB | +122.1 | +122.7 |
| DC offset, % | -0.00 | +0.00 |
THD + Noise (at -3 dBFS)
Left | Right | |
| THD, % | 0.00046 | 0.00036 |
| THD + Noise, % | 0.00065 | 0.00056 |
| THD + Noise (A-weighted), % | 0.00074 | 0.00065 |
Intermodulation distortion (swept tones)
Left | Right | |
| IMD + Noise at 5000 Hz, | 0.00056 | 0.00066 |
| IMD + Noise at 10000 Hz, | 0.00041 | 0.00050 |
| IMD + Noise at 15000 Hz, | 0.00037 | 0.00051 |
Intermodulation distortion (SMPTE)
Left | Right | |
| IMD + Noise, % | 0.00061 | 0.00058 |
| IMD + Noise (A-weighted), % | 0.00042 | 0.00039 |
Stereo crosstalk
Left | Right | |
| Crosstalk at 100 Hz, dB | -123 | -123 |
| Crosstalk at 1000 Hz, dB | -123 | -122 |
| Crosstalk at 10000 Hz, dB | -121 | -116 |
In a nutshell : It nailed just about every test. I just noticed Left channel being constantly worse when looking closer at harmonic distortion and SMPTE IMD. Yet, variations are too subtle to be any kind of audible concerns.
That was analysis for fixed output at max volume. Let see how it behaves at lower levels:
Output levels
Test | 0.9Vrms | 1.7Vrms | 2.7Vrms | 3.5Vrms | 4.25Vrms (Max) |
| Frequency response (from 40 Hz to 15 kHz), dB | +0.03, -0.23 | +0.03, -0.23 | +0.03, -0.23 | +0.03, -0.23 | +0.03, -0.23 |
| Noise level, dB (A) | -112.4 | -117.2 | -120.2 | -121.2 | -122.7 |
| Dynamic range, dB (A) | 112.4 | 117.1 | 120.0 | 121.0 | 122.4 |
| THD, % | 0.00029 | 0.00021 | 0.00031 | 0.00035 | 0.00041 |
| THD + Noise, dB (REW) | -105.2 | -106.6 | -106.2 | -105.0 | -104.6 |
| IMD + Noise, % | 0.00103 | 0.00071 | 0.00058 | 0.00061 | 0.00060 |
| Stereo crosstalk, dB | -112.4 | -117.4 | -120.7 | -121.7 | -123.4 |
That is a bit unexpected. Usually, THD+N (in fact SNR/DR) goes down in parallel with digital attenuation. Here, we got THD going lower and lower, which still results in better THD+N despite the lower DR. Note that we do get 105.2dB SINAD even below 1 Volt ! As it is, I see no issue using the Hidizs S9 PRO as a competent preamp. Needless to mention, it should also be an acceptable headphone amp as well. Specs by the way are 200mW/32Ω (or 100mW/32Ω with SE output), that unfortunately, I cannot verify.
THD + Noise (at -3 dBFS)
REW FFT 
Great figure, even at 0.9Vrms. Highest harmonic peak is now below -120dB, when related to 0dBFS level.
Jitter
(48Khz, 1-24Khz bandwidth, 16 averages, 262K FFT)
No comment. It absolutely nailed the J-Test.
Multitone 64
(192Khz, 20Hz-22Khz bandwidth, 10 averages, 262K FFT)
As it does for multitone. Excellent job !
Conclusions
My main regret here is to publishing some kind of "half review". Yes, I would have liked to test unbalanced output and more important, to find how the S9 PRO behaves relative to typical HF loads. Yet, just looking at primary results I've got out of this tiny unit, I cannot imagine it doing a bad job as a little headamp. Now, when seeing it like a regular DAC (USB in, BAL out, done!), it already performs way above its class (for both size and price) and got my easy recommendation, even for this purpose only !
Flanker rating: Competent
