A few things to remember here:

A "Big Rip" is full blown curvature singularity scenario -- and a singularity that occurs "everywhere" at finite proper comoving time. The universe just "blows up". You can see this as coming from a type of "dark energy" that behaves as an increasing Lambda. Lambda, and the scale factor, just blows up, becomes infinite after some finite time.

In that Rindler/deSitter expanding hyperballoon picture, the proper accleration of the hyperradius would go to infinity. A comover sitting on it and using deSitter static coordinates would see the Cosmological Horizon shrink down on him to a point. The "rip tide" would go to infinity.

So I would say that is indeed an end of the universe, not just as we know it.

And second, energy conservation does not have to hold globally in GR. It will always hold *locally*, but not globally. That's happenning right now in the cosmological picture.

And third, I'm just not exactly kosher with considering "dark energy" as a "regular energy density". You can move Lambda over to the right in the EFE and say, hey, that looks like some sort of source term. Such a thing there is certainly not conservered in any sort of sense as you "make more of it" as space expands.

You can leave it on the left and consider Lambda to be "just geometry" and nothing to do with a source term, which is the regular "stuff" that exists in space-time. So keep in mind the distinction between the energy of "stuff in space-time", matter and fields, vs the "energy of space-time itself". Dark energy is something in the latter category, whatever it may be.

And finally about global energy -- this is one of the most vexing things in GR that the "high priests" have puzzled and worked for a long time, with different approaches and even different pedagogies.

If you space-time is asymptotically flat, an observer there can conserve energy globally, and one can invoke that and say it *is conserved*. And I think the restriction is actually just asymptotically static/stationary and not flat, but I'm not sure.

Our universe meets none of those conditions. Now, you can (try) to define a "gravitational field energy". The Landau Lifshi-tz psuedo (hyphe to avoid the profanity filter kicking in) tensor is a popular way to do it. But it is not invariant, and no such invariant thing can be defined. Note the LL energy has nothing whatsoever to do with gravitational potential energy, or field energy in any EM-like manner. There is no "energy" associated with a static space-time at all. It's only when it's dynamic and seems to be "carrying energy in or away" does the LL thing kick in.

So, by using the LL construction, one can conserver energy in a particular coordinate system by saying the difference of energy is going in or out of the gravitational field. And that is just not invariant. Another observer would say that flow is different.

-Richard