I have been trying to get a better sense of the potential power advantage from a modestly blended plunge cut. Needless to say, not everyone who has already tested it is in a hurry to speak up but I have gotten some information and attempted to corroborate it through an independent source.
One source which has tested the modifications to a late 1.8 head on a flow bench found a surprisingly significant improvement from taking off a small amount of metal to just break the edge of the plunge cut. Sharp edges can be very disruptive to flow and this is at the edge of a critical area, so in a simple flow test the results were promising. However, with the intake manifold bolted to the head the benefits dropped dramatically, and with the restrictor plate added the advantage was largely lost at all cam lift points. More aggressive blending did not significantly improve flow of the assembled unit. So, according to this source, what looked like a possible gain of up to a couple HP turned out to be likely well under 1. Granted even a flow bench can't account for a lot of real-world variables, but it seems unlikely that something would restore that lost advantage.
But that was just one source which may or may not feel compelled to be entirely accurate. Fortunately someone not directly involved with this mess offered to reproduce the same tests on his flow bench. In this case it was a 1.6 head but the configuration is the same and the results would be enlightening if not 100% comparable.
In this independent test, a "slight blend" yielded an average flow increase of 1.8%, interestingly most beneficial at higher lift (above .250"). I confess, that surprised me a little as I expected closer to 1%, but flow dynamics are far too complex to predict all that closely, at least for me. With the intake bolted on the flow decreased, dropping the blend benefits to 0.7%. And although the 1.6 doesn't run a restrictor one was added to test the claim that it further decreased the benefits of the blend. The impact of that was less than claimed for the 1.8 but did reduce the average gain to 0.6%. However, even with intake and plate in place there were gains as high as 1.5% above .250" of lift. (Miata cams max lift is quite low, ~.310 for the 1.6 if I recall correctly).
So the separate test of the 1.6 does at least generally support the results given for the 99+ cars. If we use the more optimistic numbers from the 1.6 then the real world benefits might be in the neighborhood of 0.6% more flow potential on average. Certainly not insignificant if you are the guy without it, but likely less than 1 HP and far from an observable advantage.
Actual testers are of course welcome to elaborate or correct anything I have said here, directly or through me. If anyone else has done similar testing or is willing to do so, please chime in.
Meanwhile, I am having trouble finding a shop which has done a bunch of plunge cuts who will let me film the tools and process as they do a head for me. Not sure why? But if you know someone, I am willing to travel to their shop.