Even in water saturated with calcium carbonate, there is a constant exchange of solid material with ions so the surface is constantly reforming. This is similar to the effect one finds with ice in a freezer -- eventually the ice surface goes from smooth to jagged even though the total amount of ice may remain the same. However, with well-made plaster this is a very slow effect which is why such plaster lasts for decades. pH is probably a more important factor after saturation of calcium carbonate; temperature is also an important factor. Higher pH and lower temperature both slow down the rate of this surface exchange.
If the saturation index were consistently high, there would be a slow buildup of scale while if it were consistently low the plaster would slowly dissolve, but for well-made plaster these effects are slow. Ideally, one wants the saturation index at 0, but onBalance has shown that for well-made plaster no degradation is seen until the saturation index gets into the -0.7 to -1.0 range so having it be in the -0.3 to +0.3 range should have the plaster last for quite some time. However, for poorly-made plaster, the degradation rate is faster if the saturation index isn't at 0 so if you are unsure of the quality of the plaster then it would be best to err on the side of saturating or slightly over-saturating with calcium carbonate.