Pressure altimeters have 3 basic classes of error (forget that PITCHBLOT nonsense - it doesn't explain things correctly!)
1. They're not perfect machines. But capsule hysteris, internal temperature effect and lag can be reduced by clever design, e.g. in servo-altimeters.
2. The connection between the altimeter and the outside atmosphere is not perfect. For example, aircraft static pressure errors and mach no. affect the actual static pressure. These can be corrected by the Air Data Computer and/or Pressure Error Correction Unit - a major need for RVSM flying.
3. The atmosphere in which the aircraft is flying is non-ISA. Or the local static pressure may have been affected by hill/mountain effects. These errors are non-aircraft specific and will affect every aircraft in the same vicinity equally. They cannot be universally corrected - because an
identical atmospheric pressure compensation would be needed for every aircraft. Which is impossible to achieve.
GPS altitude may be useful for gross error checking, but can never be relied upon as a primary altitude reference.
Suppose that should read QNH? Mixing up QNH with QFE, perhaps?
No! The RVSM datum is 1013.25 mb (or hectopascals if you must) / 29.92 inches of mercury. This is often called 'Standard Pressure Setting' (SPS) - dear old Q-code junkies refer to it as QNE. Although I understand that the old morse query 'QNE'
actually meant "What is your indicated touchdown elevation with 1013.25 as the pressure datum". The reply "QNE (n feet)" meant to the receiving aircraft that if it landed using 1013.25 it would see the (n feet) indicated aerodrome elevation at touchdown.