This will be a bit of a long answer.
Honestly, despite what some folks think, the MSFS team does actually take accuracy and realism quite seriously. One of the complaints levied at the sim was that the sim was doing altitude partially incorrectly, conditions depending, and they were correct. Only that the pitot static simulation dates back to probably FS9 or maybe even earlier, so it has actually been incorrect for some time. However, the team (Seb leading this effort) endeavored to correct this. The timeline is roughly:
- Old FS9 or FSX code was made (probably close to 2 or more decades ago now), which uses only sea level pressure vs baro setting to apply an offset to the geometric altitude using a 1000ft per 1inMg rule of thumb. This is roughly OK-ish when the atmosphere is always following ISA pressure lapse rate, but horribly wrong otherwise. This may have been mostly suitable back in the day when there was no full atmospheric model.
- A change was made for SU5 to simulate altimeter temperature / pressure error vs ISA. However, it was discovered late in the beta by our awesome beta flight that this was not acting properly due to unexpected input data problems that did not surface in testing, past the time where it was feasible to make another build.
- The previous issue was resolved prior to release and scheduled for the week 1 patch.
- However, after week 1 patch release, a number of users noted a difference vs the simulator pressure altitude simvar when the altimeter was set to 29.92, when there should not be. Additionally, some users also reported angry ATC asking for altitude corrections. In a sense, we can actually thank the continued crazy temps/pressures at altitude issue for surfacing this, because the error is proportional to the deviance from ISA.
- We’ve now discovered that deep in the bowels of legacy code, the pressure altitude simvar uses code copied and pasted from the pitot static sim, but does not actually call the pitot static sim, which is an entirely unexpected result, and the copied and pasted code is the old rule-of-thumb-sea-level-pressure method. Doubly unfortunately, the in-sim ATC also calls this simvar to validate your altitude.
- The next fix, therefore, is to unify the pressure altitude simvar and pitot static sim, which will both close the loop on this decades old altimeter issue, making it compatible with non-ISA atmospheric conditions, properly simulating altimeter error due to non-ISA pressures, as well as rectify the issue with in-sim ATC.
Hopefully that helps everyone reading along. 
-Matt | Working Title