I found one online calculator (though I'm not sure how they are implementing the RD or how accurate it is) but assuming the RDs were essentially the same, the lower rated player would lose around 30 points (38.8 on that calculator but the USCF one, which uses a formula from the same creator put it closer to 30).
edit: it appears the calculator I was using might not be the best or I need to figure out how they do the RD values
http://www.bjcox.com/?page_id=2
That's the calculator I was looking at. It doesn't allow a numerical RD input but uses !! for really high confidence and ?? very low and I left it in the middle for the calculation. Using both a !! and having the 1300 lose to a 1375 gave a 10.6 point loss to the 1300.
Divergent RD values will impact the values too, though I don't know by how much. Since the numbers can be different they could move the numbers up or down some (probably not much if the RDs are withing a close-ish range), even if both have a bunch of recent games.