"These findings have been confirmed by ComputerBase and UL Benchmarks, and the latter is working on an update for its benchmark utilities that will account for VBS and help users compare scores fairly."
I don't understand... if VBS is slowing it down by 10-20%, it's fair to give a 10-20% lower score. Nobody's talked about (in the past) artificially raising Wine-based scores to match their Windows counterparts, or more recently (now that the Wine D3D9/10/11/12 support at times outruns native Windows) artificially lowering the scores to match.
Some benchmarks have had slowdowns (both Windows and Linux) from the Spectre, Meltdown, etc. mitigations -- early on people were seeings 20%+ slowdowns for some workloads on some CPUs (I think both Windows and Linux have refined their workarounds so it's less now but still non-zero). Benchmarks did not talk about trying to adjust for this difference.
Oh well!
