I am modeling a building in Open Studio, and was thinking about shortening the timestep per hour in my simulation. The epw weather file consists of hourly weather data, the default timestep in OS is 6 per hour (10 min), and it seemed logical for me to reduce my time step to 1 per hour. The difference in the results was notable (0.35 % reduction for net site energy, 1.2 % reduction of heating energy and .0.9% increase in cooling energy), but not extreme. My surface heat balance algorithm was Conduction transer function and air heat balance algorithm was Analytical solution. My model runs on ideal air loads.

Can anybody explain to me why the number of timesteps matters, if the outside conditions stay the same for one hour?

Which heat balance algorithms are favourable for the surface and the air heat balance ? From what i read in the ergineering reference document, the analytical solution seems the most stable solution for the air heat balance, but for the surface heat balance algorithm, the options stay wider open.