Have you ever looked at what happens to air temp as boost increases on a data log? On non-intercooled motors, I've seen the air temp increase 150 degrees in the duration of a single dyno pull. Compressing the air really heats it up and it makes a very big difference. I suppose you could "make up" for it in the VE table, but you really should have the ATS in the compressed air stream. The relationship between air temp and air density is known, and the speed/density calculation that happens inside the ECU figures that in.