Power doesn't just depend on avg force and avg pedal speed during a pedal stroke: it also depends on how these vary relative to each other. Quarq makes a certain assumption about how they vary. That assumption needs to be different for round rings versus eccentric rings. But the assumptions won't apply equally well to all pedaling styles. If you're slogging through the mud or up a steep hill or into a block headwind, the variations will be different then if you're hammering a descent. In other words, if you produce exactly 250 watts, it may report more in some cases, less in others.
Powertap doesn't have this problem: it doesn't care about the details of your pedal stroke. Powertap has other issues, but ring shape shouldn't affect its accuracy at all. Neither would it affect the Garmin Vector.
Well their "assumptions" are very accurate. Friend of mine has a powertap, he has used both round rings, and biopace. We have done the same race/group training rides. Then I send my Garmin file to him.
He takes them, just like I do in WKO training peaks software, overlays the files to compare. They are basically identical the entire time, the only difference is he weighs 20-25lbs heavier than me, and his power output will show about 10% higher on average due to his fat butt being hauled around. Otherwise, the curves, peaks, lows etc...all are identical. It is easy to correlate and see the route and how we are riding within 50ft of each other the entire time.
The proof is in the pudding as they say. So Quarq is doing the calculations very similar to a powertap no matter what rings on the crank once calibrated properly.