Well the ephem keypair is only required for traditional stealth payments because that's how both parties can calculate the shared secret. That's not necessary in my first scenario because only the public address of the payee is required to encrypt the randomly generated shared secret. It's probably easier to think about my second scenario, the one I gave in the edit. In that scenario the ephem keypair is still required to calculate the first shared secret like normal, but we use it as the key for a symmetric encryption algorithm so that we can encrypt the randomly generated shared secret. The second random shared secret is what we actually use to calculate Q'.
Ok, I see it now. I think that (using Q X e as an encryption key) would work in principle,
but you'd have to test it in practice. Probably there's some hitch or two, otherwise
this would have superseded Diffie-Hellman already in computing stealth keys. Maybe
it comes down to the fact that this requires an additional layer of encryption, whereas the
original method works just as well and very elegantly too.
.