The vise has problems. So did the table. Chips being sandwiched between the vise base and table put numerous craters in both and pulled them convex around the central pivot stud. I got the table reasonably flat again by scraping and stoning, then took apart the vise to measure and study. The vise had a twist to the rails (warpage, I think, not abuse) and a pretty good saddle from wear.
I scraped the bottom of the vise frame flat as a reference point, then scraped the top flat and parallel to it.
My granite plate is AA grade which means flat to something like 80 millionths of an inch, so the blue print on the vise is near that now. I had to correct it from a maximum of .0025" dip in the middle and .0016" sideways tilt. The tilt accounted for the .004" in 4" taper I was getting in my parts, nott to mention all the tolerance stacking from the base to table and base to vise frame. The goal here is to make the blued surface above flat and parallel to the ram ways under load. I also scraped the bottom of the vise base flat to match the table, so now I need to put it all together and tram it to the ram and saddle to check my work. Before I do that, though, I have to set up the vise frame upside-down on some 1-2-3 blocks on my mill table, clamp it carefully so as not to induce warp, and cut the waves and low spots out of the articulating jaw's guide ways on the underside to make them even and parallel to the top of the frame, then mill the jaw itself to restore a very close fit so it doesn't lift when tightened against a workpiece.