(In reply to Takashi Iwai from comment #21) > If the decompression speed (which is the most important factor for switching > to zstd) is comparable with lower compression levels, then it can be a good > win. That was a consistent finding in my testing. It's also documented in the technical papers about zstd, it's a property of the algorithm by design. I haven't digged deep enough into the math to explain why.