Should I use optical or coax into my Qutest?

Short answer

For the Qutest, Watts's stated preference is optical. In July 2018 he told Qutest owners to "use the optical as your benchmark" and said coax "will always be the brighter sounding input" Head-Fi, Jul 2018. He did not say coax should never be used, and none of his posts found in this archive puts the Qutest's coax sound ahead of optical Head-Fi, Jul 2018Head-Fi, Jul 2018.

What he said about the Qutest (July 2018)

• Coax is brighter. He said coax "has no galvanic isolation, and direct coupling to the grounds," and that it sounds brighter because RF noise interfering with the analog stages creates more noise floor modulation Head-Fi, Jul 2018.
• Optical is warmer and smoother. "Optical is warmer and smoother, due to the complete galvanic isolation" Head-Fi, Jul 2018.
• Why optical is different. He said optical sends data through light only, with no electrical connection, so the coupling capacitance is almost zero Head-Fi, Jul 2018. He also said the only way sound quality can be impaired is RF noise connected via the ground, and optical does not suffer from this Head-Fi, Jul 2018.
• Why he did not filter coax. He said he could not put RF filtering on coax "without degrading lock abilities" Head-Fi, Jul 2018.
• Brighter can be mistaken for better. Replying about another listener's report of a "fuller" sound and "better soundstage" (the archived post does not show the question), he said that if fuller means brighter, extra noise floor modulation can be mistaken for transparency. He added that coax can have worse depth: flat and wide, against a narrower but much deeper soundstage Head-Fi, Jul 2018.
• Grounding matters for the Qutest. He said the difference appears once the DAC is grounded, "which normally happens with Qutest, as you will connect an amp on the outputs," and that listening tests need all other sources disconnected Head-Fi, Jul 2018.

Jitter is not the reason to choose

In 2014, on the original Hugo, he noted that optical is generally disliked for having more jitter than coax, and said his DACs remove incoming source jitter Head-Fi, Apr 2014. In the Qutest thread in July 2018 he said source jitter is not an issue with his DACs Head-Fi, Jul 2018. In May 2019, in the same thread, he said you can put 2 microseconds of 1 kHz jitter on his DACs via S/PDIF and measure nothing, down to an FFT noise floor of -180 dB Head-Fi, May 2019.

Practical points he gave on the Qutest

• Optical-to-coax adapters (July 2019): if it is battery powered, he said, "I wouldn't worry." A mains-powered one sets up ground loops via its power supply and could make the sound brighter than it should be Head-Fi, Jul 2019.
• Optical switchers (July 2019): he said he has used them, they work well, and he could not see them degrading the sound, though you can run an AB test to check Head-Fi, Jul 2019.
• Sample rate (September 2019): the Qutest can run at 192 kHz over optical, but only if the source and fiber are up to it. He said current optical parts are not as good as the older Toshiba ones Head-Fi, Sep 2019.
• DSD (July 2019): the Qutest decodes DoP over S/PDIF, so DSD64 works over optical if your source sends it, and DSD64 is the optical maximum Head-Fi, Jul 2019.
• Noisy sources (April 2021): to a Qutest owner using a CD player, he said not to worry about the player's DAC over optical, since "Optical is immune to RF," but warned the player can inject RF noise into the mains that the DAC or amps then pick up Head-Fi, Apr 2021.

How USB compares

On the Qutest he said USB "comes close, sometimes identical to optical" Head-Fi, Jul 2018. With a low-power, simple source, particularly a battery-powered one, he said USB sounds the same as optical Head-Fi, Jul 2018. In February 2019, in the same thread, he said "Technically USB is better from the clock POV" but that in practice optical had sounded identical to USB Head-Fi, Feb 2019.

Over time: in 2016, for DAVE, he favored USB Head-Fi, Jul 2016; in January 2018 he wrote he had been wrong about that Head-Fi, Jan 2018; in October 2022 he wrote "there is no technical benefit to using USB over optical when using my DACs" Head-Fi, Oct 2022, which differs from his February 2019 wording.

Later statements on other products

These were made about other products or in general, not specifically the Qutest. In May 2022, in the Hugo 2 thread, he wrote: "If you have a choice, always go for optical over coax" Head-Fi, May 2022. In December 2022 he called optical "the best digital connection" Head-Fi, Dec 2022. For the Mojo 2, in February 2024 he said Toslink optical should be considered the reference source Head-Fi, Feb 2024, and in August 2024 that coax is the most sensitive input to common mode RF noise currents because ferrites on coax stop S/PDIF decoding reliably Head-Fi, Aug 2024. Earlier, for the original Hugo in 2014, he preferred optical as smoother and more natural Head-Fi, Apr 2014Head-Fi, Jun 2014.

What he has not said

He gave no Qutest measurements comparing the coax and optical inputs, and did not say whether any coax cable removes the brightness. His Qutest statements on this question are listening impressions and design reasoning Head-Fi, Jul 2018Head-Fi, Jul 2018Head-Fi, Jul 2018.

Quotes

“Coax has no galvanic isolation, and direct coupling to the grounds, so will always be the brighter sounding input.”2018-07-29 Head-Fi, Jul 2018

“Optical is warmer and smoother, due to the complete galvanic isolation.”2018-07-29 Head-Fi, Jul 2018

“so use the optical as your benchmark.”2018-07-29 Head-Fi, Jul 2018

“The only way SQ can be impaired is via RF noise connected via the ground, and optical does not suffer from this problem.”2018-07-29 Head-Fi, Jul 2018

“If you have a choice, always go for optical over coax.”2022-05-07 Head-Fi, May 2022

Dated positions

Caveats and open points

• All Qutest-specific statements come from Head-Fi posts, mainly July 2018 to 2019, plus one April 2021 reply; no transcript or article addresses the question.
• He gave listening impressions and design reasoning, not Qutest measurements comparing the inputs.
• Statements about Hugo, Hugo 2 and Mojo 2, and his general 2022 posts, are reported for context and are not Qutest-specific.
• Several archived posts lack the question they answer, so the context of replies such as Head-Fi, Jul 2018 and Head-Fi, Jul 2018 is inferred from his wording only.

Topics: Choosing and comparing inputs, RF noise: mechanisms and audibility