plumpudding2
500+ Head-Fier
Most people here will also be subscribed to the Roon thread (https://community.roonlabs.com/t/which-hqp-filter-are-you-using-2023/244185/584) but i'm crossposting here anyway in the hopes it will help this thread on Head-Fi gain traction
:
I’m always sceptical of my own listening impressions as the placebo effect is strong and I’m not good at articulating differences between sound sources.
Yesterday my girlfriend came over and nowadays I’ve got my listening station set up so that two people can listen at once, and I decided to see if she could perceive modulator/filter differences. It was an A/B test where I announced every change (so there is expectation bias). However I didn’t tell her what I was changing or how it “should” sound. No bias to what she should perceive.
Set up: Holo May dac, Benchmark HPA4 into Focal Utopia, HPA4 also as a preamp into Ferrum Oor + Hypsos in poweramp mode + Hifiman Susvara. She preferred the Susvara so I got the Utopia.
Used Roon’s volume leveling function to listen to exactly 80dB k-weighted average volume.
Songs were an album and artist she knows very well, Pasaka by Jessica Shy (Open Qobuz), not particularly audiophile but also not a bad recording by any stretch.
Differences were very apparent to her untrained ear, and she was actually able to articulate them better than me. Filter used was sinc-MGa.
PCM at 1.536 MHz “sharpest”, most attack, least space/2D sounding
DSD1024 AMSDM7 512+fs, also quite sharp, much more space
DSD256 ASDM7ECv3: softer, but smaller space similar to PCM
DSD512 ASDM7EC-light 512+fs: softest modulator, space similar to 1024 configuration, her clear favourite.
Once I established 512 with EC-light as her favourite I put on the Gauss-long filter to see if she preferred that, she said the balance of vocals to instruments changed and that the instruments were more forward compared to MGa, and that she preferred MGa.
Finally 512 EC-light wasn’t her favourite on every track, if the track was mixed with forward vocals the instruments moved too far back and she then preferred ECv3 at 256 or even PCM.
It has become a long post but I was just excited to share the results of this little experiment with the community

I’m always sceptical of my own listening impressions as the placebo effect is strong and I’m not good at articulating differences between sound sources.
Yesterday my girlfriend came over and nowadays I’ve got my listening station set up so that two people can listen at once, and I decided to see if she could perceive modulator/filter differences. It was an A/B test where I announced every change (so there is expectation bias). However I didn’t tell her what I was changing or how it “should” sound. No bias to what she should perceive.
Set up: Holo May dac, Benchmark HPA4 into Focal Utopia, HPA4 also as a preamp into Ferrum Oor + Hypsos in poweramp mode + Hifiman Susvara. She preferred the Susvara so I got the Utopia.
Used Roon’s volume leveling function to listen to exactly 80dB k-weighted average volume.
Songs were an album and artist she knows very well, Pasaka by Jessica Shy (Open Qobuz), not particularly audiophile but also not a bad recording by any stretch.
Differences were very apparent to her untrained ear, and she was actually able to articulate them better than me. Filter used was sinc-MGa.
PCM at 1.536 MHz “sharpest”, most attack, least space/2D sounding
DSD1024 AMSDM7 512+fs, also quite sharp, much more space
DSD256 ASDM7ECv3: softer, but smaller space similar to PCM
DSD512 ASDM7EC-light 512+fs: softest modulator, space similar to 1024 configuration, her clear favourite.
Once I established 512 with EC-light as her favourite I put on the Gauss-long filter to see if she preferred that, she said the balance of vocals to instruments changed and that the instruments were more forward compared to MGa, and that she preferred MGa.
Finally 512 EC-light wasn’t her favourite on every track, if the track was mixed with forward vocals the instruments moved too far back and she then preferred ECv3 at 256 or even PCM.
It has become a long post but I was just excited to share the results of this little experiment with the community

Last edited: