What was the point of doing this prior to selling a subscription to unlock it?
There are/were legitimate considerations for this too--I've had a few GTIs that ended up 'tuned.'
Typical failures on tuned cars, past a certain power, level were clutches (both in MT and DSG models) and probably other ancillary components.
I'm sure there's also lifespan calculations for components at the stock power levels too. Probably a shorter lifespan/the projected failure rates only account for the stock output etc etc.
1) Build engine with $market_leading_horsepower.
2) Throttle engine so that base models of a car have $market_leading_horsepower - $throttle_amount.
3) Price discrimination for enthusiasts removes the throttling back to the original $market_leading_horsepower while also saving the effort of having to design a faster engine.
Is there no extra cost associated with machining the faster engine? That is, would manufacturing a slower engine for the base model not be any cheaper than just building the fastest engine they can into all models?
There probably is but it's much less than the cost of having different engines for different performance specs of the same model of car (which is very common).
In reality cramming a little more air through the thing is no big deal because it'll be within the fudge factor of all the other parts.
Nobody is engineering a clutch to within an inch of it's life on a 200hp shitbox, they'll use the same clutch as the 300hp version of the same engine. Yeah the trans shifting tune will be different but that doesn't necessarily mean that the components in it are different.