I don't think people understand what is being said about accuracy here. Accuracy is the degree of correctness that a device reports it's position in space. In that regard, the Moves are VERY accurate, when they are not occluded they report their position in space with very close accuracy. The classic test of accuracy has been to use real world anchors -- place a controller in one spot at start up, move it away, then move it back to the origin spot, and see how much the virtual position diverges from the start position.
Things like the Windows VR headsets are not accurate, but they get away with it because it's not obvious without outside calibration that you're actually off a few millimeters from your origin position. Since the tracking is relative to the headset, the origin moves as well, hiding the inaccuracy.
RoadToVR understands all this. I have a ton of problems with the publication, especially how they hired Heany as an editor, but they DO understand this type of discrepancy.
Now, the playstation moves are accurate, but they are not reliable. They are very subject to occlusion, meaning you can move between line of sight. In this case, they cannot track, and the software goes wonky. This is not a problem of accuracy, this is a problem of reliability, and no duh these headsets are more reliable. What is being reported is NOT the reliability. This is a comment on the accuracy, and it absolutely is news if a markerless inside-out system is able to best the PSVR's outside-in tracking in accuracy.
