Two three-story full-scale sub-standard reinforced concrete buildings were tested under self-weight and reversed cyclic lateral displacements to examine their behavior during earthquakes. While one of these buildings was a part of an actual existing building (TB1) built in the beginning of 1990's in Istanbul, the other was constructed by the authors as a representative building (TB2) that reflects the most common structural deficiencies of existing building stock in Turkey. Both buildings were constructed with plain bars, low strength concrete and inadequate lateral reinforcement. The differences between these buildings were axial load levels of first story columns, expected location of major structural damages (weak beam-strong column for TB1 and strong beam-weak column for TB2) and connection details of longitudinal bars in columns at the foundation-column interface (continuous for TB1 and lap-spliced with 180A degrees hooks for TB2). Both buildings were pushed and pulled at increasing displacement amplitudes up to near collapse well beyond the life safety performance level. While TB1 was damaged significantly at 1.5 % inter-story drift ratio, TB2 reached near collapse damage at 4.0 % inter-story drift ratio. In this paper, details and test results of these two sub-standard buildings are presented. In addition, the available nonlinear modeling techniques and performance predictions of the Turkish Seismic Design Code (Specification for the buildings to be constructed in disaster areas. Ministry of Public Works and Settlement, Ankara, 2007), ASCE 41-13 (Seismic rehabilitation of existing buildings, ASCE/SEI 41-13. ASCE, Reston, 2014) and Eurocode 8-3 (Eurocode 8: Design of structures for earthquake resistance. Part 3: Assessment and retrofitting of buildings. Comit, Europ,en de Normalisation, Bruxelles, 2005) are compared with the experimental results.