In, IIRC, 2006, Charles Petty had a series of articles on testing components, including primers, in a .223 Remington Cooper Phoenix. He was using a 55-grain A-max and 24 grains of Reloader 10X in all the primer tests, IIRC. I forget whose brass it was. He got velocity ranging from 3150 fps with his mildest primer (Federal 205, I think), and 3000 fps from the warmest magnum primer (I've forgotten which one it was).
If finally got to read Pearce's article on .308 loads. In the larger capacity cases, the velocity difference is typically smaller. If I simulate Pearce's 3000 fps pressure increase in QuickLOAD, using the start pressure over a load 4895 with a 168-grain match bullet, I get only about 20 fps increase. It's not a lot, but it is measurable. However, what is more interesting is Pearce's description of increasing peak pressure with a Federal 215, but getting a muzzle velocity drop. What explains this is the higher peak pressure accelerating the bullet early enough in travel (the peak occurs within an inch or two of the throat) that the expansion rate became too great for the powder choice to make gas fast enough to keep up, so late-barrel pressure and post-peak acceleration dropped. It's an example of how this stuff can fool and, under some circumstances, be counterintuitive. The main point I would make is to watch out for velocity changes (in either direction) as an indicator ignition and pressure have changed and then, of course, keep watching for pressure signs.