Wow... it's like a reunion here. Ara, John, Coleman, Jenn and George
The whole Matrix experiment was a fun one. We all knew we were in for a challenge but there was no cost to us individually so we gave it a shot. With several hundred thousand readers watching the progress of the car and being subject to fun lovingly abusive journalists, I was quite conservative with the tuning, staying pretty far away from MBT and running a fuel curve that would never be mistaken for a leanburn EPA calibration
Whether or not we know actually what happened, I think the point Dave makes is that there isn't much of a margin of safety with boosting a high CR motor. At least not enough for most users in most applications. Bad gas, sustained torque loads, sustained charge temps, etc,. will all take its toll. If not now, then definately later. And both Dave and I are considered optimists
When NA engines are designed to rev high and make a high specific output (unaided by turbos/scs), they are designed with very high static compression and very lightweight componentry. Both of which make them very grumpy when trying to support substantially higher-than-stock power loads. Even without the presence of detonation, ultra-high peak cylinder pressures (high CR x Boost = massively high dynamic compression loads) will overwork the piston ring seal (possibly causing excessive blow by), bend lightweight rods and do other nasty stuff that we all assume will never happen to us. Everything will look and feel good until you start noticing a loss of power, a nasty clanking noise and/or a lot of oil in your intercooler.
Is trying to make more power (through boost) a hopeless proposition in the Elise? Certainly not. But it's going to require a bit more thought and assumed risk than most shops and customers are willing to assume.
Just my 2c,
shiv