First of all I agree that ammonia is usually a bad thing in any tank, especially in the water column, but in the substrate is it ALL bad??? After all I have heard people really raving about the initial boost of Amazonia???
The metabolic pathways for Nitrogen assimilation (from what I have read, it's not an area I have experimented in) prefer ammoniacal Nitrogen as it is easier for the plant to uptake it. Indeed I'm sure I have read somewhere (can't remember where, but probably APC) that plants have to reduce Nitrate to allow it to be assimilated.
Ammonia and even urea seem to be the favoured form of nitrogen added to terrestrial fertilisers, partly because of the production costs, but also because of the results of their tests.
I have a feeling that fluctuating ammonia may be the real cause, because, if you think about it, ammonia in a tank will usually come in bursts, e.g after a feed, when a tank's disrupted etc. It's not something you'd want to maintain at a relatively stable level!!!
Totally agree with the comments on fiddling more when the tank is new, but surely this could equally apply to a tank with a soil substrate? Why should a soil substrate have to be free of ammonia when an ADA tank doesn't???
A lot of points on reducing algae and boosting plant growth, in my humble opinion, seem to settle on stability as the key. Stability favours higher, specialised organisms over more basic ones, and it always seems to me that an older, more settled tank (well maintained) works better, rather than there being any particular 'magic' value of nutrients,CO2, light, etc. I have had big problems with the EI method and algae, which I feel is completely down to me simply not being organised/consistent enough to maintain stable levels and daily dosing, rather than any flaw in the method. For me low nutrient tanks work best, as it suits me and my fish.