Also consider the theoretical film formation models e.g. by J. Krug (e.g. PRL 70 (1993) 3271 http://dx.doi.org/10.1103/PhysRevLett.70.3271), F. Family (e.g. PRL 61(1988)428 http://dx.doi.org/10.1103/PhysRevLett.61.428, or Physica A: Statistical Mechanics and its Applications 168 (1990)561), M. Kardar (Physica B 221 (1996) 60), G. Palasantzas (Int. J. Mod. Phys. B 9 (1995)599, DOI: 10.1142/S0217979295000239) and others!
In general, it also strongly depends on all the details of the deposition process, the nature of the deposit and the substrate, the temperature of the substrate, evaporation rate and vacuum conditions, ... so it is hard to make clear predictions, and there are many practical studies that prove this!
As mentioned above, the structure zone model (which relies on melting temperatures of your thin film material, i.e. on thermodynamical properties only) may give a first hint. Low temperatures may result in loosely bound, eventually even porous films, and the substrate may not be coated homogeneously. Higher temperatures may results in compact films, because of the onset of diffusion, leading to the annelaing of incorporated defects such as voids (on a macroscopic scale) as well as on a microscopic scale (i.e. interstitials or vacancies or even grain boundaries). But, and now all the details come into the game, all these processes depend on the speed of the deposition, the vacuum quality (i.e. uptake of residual gas - even noble gases may be incorporated at low substrate temperatures) etc. So it is hard to make a straightforward prediction for all kinds of thin film/substrate combinations at all temperatures, unless there are some more or less general trends. Check the available literature for your particular system of interest!