I have a technical question regarding DSD to PCM conversion in HQPlayer.
I converted a DSD track with clipping up to +1.5 dB to FLAC using HQPlayer. After conversion, the peak levels dropped to -6.7 dB (left) and -6.5 dB (right).
My understanding is that this does not mean the clipping has been fixed — the clipped peaks have simply been attenuated to a lower level. However, I wanted to confirm: does HQPlayer perform any kind of peak reconstruction or signal reshaping during conversion that could restore the original waveform, or is it merely applying gain reduction and/or a soft limiter?
I suspect the latter, but I would appreciate your clarification.
DSD doesn’t clip yet at +1.5 dB. It can go momentarily up to +3.15 dB without clipping. Although many DACs will clip the output when presented such data, unless they have the necessary headroom built-in.
Also HQPlayer doesn’t internally clip with any level, but the output has a soft-knee limiter to ensure output from HQPlayer doesn’t exceed 0 dB and that the limit curve is smooth.
You can also engage the 6 dB conversion gain if you prefer, and lower the HQPlayer’s main volume by suitable amount if you’d like some other output level. Just as long as you keep enough headroom no to trigger the limiter (Limited counter would increment).
For short duration clips, usually most of the lost peak can be recovered if the output has enough headroom.
P.S. That MusicScope (from now defunct company) doesn’t really provide accurate information for DSD sources. So I would recommend to check with HQPlayer instead.
Thank you very much for the detailed explanation. Everything is much clearer now.
The more I learn about HQPlayer, the more I like it.
Your comment about MusicScope not being accurate for DSD sources explains the discrepancies I was seeing between it and other analysis tools. I initially liked MusicScope, but over time I started to notice various inconsistencies.
