This paper presents the results of an experimental campaign designed to compare and understand the performance of passive protection under exposure to standard furnace tests and natural fires. As part of this campaign, five natural fire experiments were performed with partially protected cross-laminated timber (CLT) compartments under a range of ventilation conditions. In all the tests, only one side wall was left completely unprotected, and all other timber surfaces were protected with either two layers of 18 mm standard gypsum boards (GB) or two layers of 25 mm standard GBs. The structural CLT ceilings were subjected to a superimposed dead load of 1.35 kN/m² during the natural fire tests, and the fire load was (on average) 950 MJ/m²; chosen to represent the Eurocode 1991-1-2 characteristic value for dwellings. The performance of the passive protection was mainly evaluated with regards to the time to reach a protected timber surface temperature of 250°C. The testing confirms that the resulting fire protection performance of a given gypsum board layout depends on the ventilation conditions of the fire compartment, with more severe (and closest to ISO testing) outcomes when testing under ventilation-controlled scenarios. This paper provides data that sheds light on the co-dependency of the passive protection design and compartment fire dynamics and underlines the importance of considering the safety objectives of a building when defining the performance criteria of its structural elements.