Meet my desktop. When Singxer's I²S pins matched Wandla's default, HDMI became my input of choice. Audirvana's r8brain upsampler remained set to my usual 176.4/192kHz, digital files were either local or cloud-bound via Qobuz Sublime, both routing through Audirvana Studio to circumvent Window's own sound engine.
Ferrum's default filter choice is HQ Apodizing so I left that for my first date. Ditto for the standard 24V Hypsos setting which at 100 volume showed ~12W draw at 0.5A.
Below shows the remote wand close up and how the display temporarily does large numbers when we change volume. With a touch display, this master rotary needs no push option like Hypsos's which turns to highlight an item then pushes to select. Wandla options are I²S input DSD pin and format, up to ±12dB analog gain trim, up to -12dB digital input trim for 9.5/4.75Vrms to 2.4/1.2Vrms XLR/RCA out¹, HT bypass, TV control/trigger and various adjustments to display/logo brightness and behavior. 'No volume' lives under its own 'bypass' tab while 'volume control' itself shows digital/analog. When the outputs are fixed, the display shows 'bypass' for a visual reminder that we now best have volume control elsewhere. The digital filters open in the 'upsampling' tab to offer no insight into how much upsampling each applies. We can appy our own in upstream player software. Wandla's display confirms incoming sample rate and bit depth. I confirmed that Wandla handles 48kHz source material upsampled to 768kHz without any issue.
¹ Setting digital input trim to -8dB creates standard 2V/4Vrms RCA/XLR outputs. By default, Wandla is set for the pro standard of 4.75/9.5Vrms "for better S/NR". At 9.5V, most home systems expecting 2V will need very stiff attenuation to not get too loud far too soon. If we believe that the best attenuator is no attenuator, Wandla practices different religion.
At my SPL, Suvsara + AMP-23 at max gain wanted ~80 on Wanda's display in both digital/analog volume mode. That prompted questions. An opamp buffer follows Wandla's Muses controller. Do bypass and digital attenuation hit the same buffer? Does Wandla's 0-100 volume display reflect actual dB of attenuation or another scale? While I had Ferrum's ear, I also asked what upsampling multipliers Wandla's current five digital filters apply. From Pawel Gorgón we learn that "the upsampling multipliers are always integer. All ESS filters upsample x 8 by default. Our HQ filters upsample to 352.8/384kHz."
"Volume scale is in decibels so 0 = minus 100dB, 80 = minus 20dB. Bypass and digital volume completely circumvent the Muses chip and its buffers. Analog signal from the anti-aliasing filter at the DAC output past the I/V converter proceeds straight to the output line driver." Power users in fixed gain spot a potential tuning feature between analog volume at 100 vs bypass? For power, the stock supply is a Globtek switcher of 24V/2.5A. It's as long as the remote, as tall as that's wide. Its width is 1½ times that of the remote's own. Now we're done with hard facts. First takes with Ferrum's default filter, volume bypass, default 4.6V RCA out and Hypsos at 24V were decidedly un-Sabre. That's when we invoke that early if diehard ESS perception of a detail-über-alles, slightly pixilated, wiry or hyper-enunciated sound. Wandla presented warm, soft, dense, darkish and ultra friendly. Reducing supply voltage to the minimum 22V got a bit leaner or perkier, moving to 30V prioritized the weightier darker aesthetic. When Audirvana Studio upsampled to 705.6 or 768kHz, it caused some frost with unpleasant edge limning or pixilated transients. I quickly undid that. Punching my way through five filter options proved that Hypsos sweet-spot tuning is a far bigger decider. The filters are the final fine-tuning.
I ended up with -10dB digital input trim, Hypsos at 22V, HQ Apodizing and I²S signal pre-set to 176.4/192kHz. That gave me the spatially deepest most 3D-sculpted imaging with the best walkabout visibility. The sound was still proper dense but not as soft, thick and subliminally fuzzy as it had been with far higher power voltage. I'd locked in 'my' sound. Specifics are immaterial. Relevant is that you'll be able to lock in 'your' sound just as easily. It's why the above settings render virtually see-through. They must remain transparent to your own needs and preferences.