The voltages on pins 15-18 of IC1 are a good 10% higher and the current of 27mA is 20% lower than the sim predicts, but I don't think that's the root of the problem.
The sim shows that a discontinuity in the +12V supply line, somewhere between R15 and R21, would result in a step function like the one you're seeing. So worth getting the magnifier out and checking for a hairline crack in the pcb trace in that area, also metering the whole length of the +12V line.