Comparison of SBC (Bluetooth Audio) and PASC/MP1 with fascinating conclusions

First of all, excuse the clickbaity title, I failed to come up with a better title. Feel free to suggest one.

SBC is the default Bluetooth audio codec, still commonly used where aptX or other proprietary more modern codecs are not available.

It uses 4 or 8 subbands which is far less than the 32 subbands of PASC and 345 kBit/s compared to 384 kBit/s of PASC on DCC.

The Wikipedia article lists

Frans de Bont has based the SBC audio codec on his earlier work,[4] and – in parts – on the MPEG-1 Audio Layer II standard. In addition, the SBC is based on the algorithms described in the EP-0400755B1.

He was a Philips employee, if I interpreted everything correctly none of the Layer II improvements with the added scale-factor are not used and Philips was also involved with the patent dating back to the development timeframe of DCC.

This leads me to believe that an even more simplified version of PASC/MP1 is still in widespread use today and even that does not sound too bad. The subtle audible differences of SBC allowed me to experience what subband coding actually sounds like which is almost impossible with the much finer PASC. With my previous knowledge of the concept of subband-coding, I was able to tell that segments with few different frequencies/subbands sound the same but segments with many different frequencies/subbands sound different, often dull. This suggests a lack of certain higher subbands/frequencies.

Could the perceived warmth of the sound of DCC related to that? What do you think @Jorn? I might build an experimental PASC/MP1 decoder with subband visualization and statistics to investigate that among other details.

2 Likes

@Max

This is an interesting topic,
My first thoughts is less sub-bands saves data, so not directly in favor to the quality,

Objective quality definitely not, but maybe subjective like Vinyl which is objectively worse subjectively better, but in a much more subtle way.