Laboratory test report
The Korg DS-DAC-100m provided exemplary technical measurements. The noise level from the line outputs when running with 24-bit audio was –107dBA, while the THD+N figure was measured 0.0013%, and IMD+N at 0.0018%. Insignificant in both cases. There was tiny bit of measurable leakage of 50Hz mains (peaking just above –120dB and the third and fifth harmonics at even lower levels) when measured initially with the Mac. When the PC was running on battery even this was gone.
With DSD the performance was identical on the noise and distortion front to 24-bit PCM, except for rising noise levels in the ultrasonic, beginning around 21.5kHz in DSD64, peaking just above 50kHz. The DSD128 noise started appearing above 40kHz.
PCM at 192kHz had the most extended high-frequency response, with a 0.1dB boost at 20kHz, then down by 0.5dB at 42kHz and 3dB at 55kHz. DSD128 was down by 3dB at 52kHz and SACD standard DSD64 was at –3dB by 44kHz.
With CD-standard 44.1kHz signals there was none of this gentle roll-off stuff. There was a slight rise in the output by around 0.15dB at 20kHz, with a brick wall at 20.8kHz. With 128× multibit oversampling, you can be pretty much as aggressive as you like in the filtering without adverse affects, and that’s what Korg has done.
The frequency response at the headphone output was identical to that at the line output, except for a slight reduction of 0.25dB at 20Hz. Despite being USB powered, the unit managed a decent 6 milliwatts into a high impedance (552Ω) load, suggesting about 8dB above the sensitivity rating for very high-impedance headphones. Into the low impedance (32.6Ω) load the output had to be reduced by a couple of notches to avoid clipping. Even so it was slightly above spec at two of the test frequencies: 84mW at 100Hz and 1kHz, and only slightly below (77mW) at 10kHz. The lower frequencies will deliver 19dB relative to 1mW. That ought to be plenty loud enough for most.
My measurements suggested an internal impedance at the headphone output of around 0.5–2Ω at 1kHz. The frequency response of your headphones therefore should not be badly influenced by the impedance profile of this unit. Stephen Dawson Readers should note that the results mentioned in the report, tabulated in performance charts and/or displayed using graphs and/or