kodama said:
i understand the issue you raise about pressure.
I don't think you do, since you leave it out of the rest of your post, even though it's critical for understanding why the three cases you give do not all behave the same. See below.
kodama said:
based on the equivalence principle
The EP is irrelevant here, because the three configurations you mention all have different pressures, and pressure also gravitates. So the cases are not equivalent in any relevant sense.
If we idealize all three cases as homogeneous perfect fluids, then how they "gravitate" can be expressed as follows:
$$
\frac{\ddot{V}}{V} = - \left( \rho + 3 p \right)
$$
where ##V## is the volume of a small ball of test particles centered on some point in the fluid, ##\rho## is the energy density of the fluid, and ##p## is the pressure of the fluid. For all of the cases you give, the fluid has an equation of state of the form ##p = w \rho##, where ##w## is a constant. So we can rewrite the above as:
$$
\frac{\ddot{V}}{V} = - \rho \left( 1 + 3 w \right)
$$
This tells us the percentage "acceleration" of ##V## as a function of ##\rho## and ##w##. Now let's look at your three cases:
(1) The fluid is made of ordinary matter. For ordinary matter, ##w = 0##, so we have ##\ddot{V} / V = - \rho##. The small ball of test particles accelerates inwards--its volume decreases. This is what we mean when we say that gravity is attractive for ordinary matter.
(2) The fluid is made of photons. For photons, ##w = 1/3##, so we have ##\ddot{V} / V = - 2 \rho##. So gravity is also attractive for photons, and a given energy density of photons gravitates twice as strongly as the same energy density of ordinary matter. (In practice, the photons we observe have energy densities many orders of magnitude smaller than the energy densities of the ordinary matter we observe, which is why we don't notice this effect.)
(3) The fluid is made of vacuum energy. For vacuum energy, ##w = -1##. This means that we have ##\ddot{V} / V = 2 \rho##. Notice that the RHS is now positive, not negative; the small ball of test particles accelerates outward, not inward. (In the case of the universe as a whole, this shows up as accelerated expansion instead of decelerated expansion.) So vacuum energy gravitates the opposite way from ordinary matter or photons.