Currently, two-speed time engines in "RTS" games work like this. There is a turn-based time segment (where civil, economic and strategic decisions are made and then there is a real time segment where battles take place. The Total War series is an important example. Of course, turn based play is NOT real time. Even RTS time is not real time it is "scaled time". Nobody, to my knowledge, has attempted a two-speed time engine where both times are real time. The way to do this would be as follows: Tactical real time would be 1:1. This means if a battle engagement would truly take one hour then it would also take an hour of the player's real time sitting at the game. Strategic time would 60:1. This means that in one minute of the players time, strategic actions that would need one hour in-game can occur. Imagine you task engineers to build a section of fence. If this fence would 50 workers 1 hour to build in real life, so to speak, it would take 1 minute in strategic time. The best way to imagine tactical time and strategic time and the necessary time switch is as follows. Think of the standard Cossacks game screen. It is always in tactical time. The mini-map shows the same time progress but it is just scaled down to dots for formations. If strategic time was added, what would happen is this. The tactical screen would expand to full screen. The dots for formations would become standard formation symbols used for military maps (a rectangle with an X in it is a cavalry formation and so forth). The strategic display encompasses the whole map and the whole screen (with fog of war still in place as appropriate). Things are sped up sixty times now in strategic time. Maps would be large (at least 4x) and the formations would move as per the 1 second = 1 minute ratio. Positioning large formations would occur at this speed (or if that is too fast for the design it could be at some lesser ratio like 1 second = 1/4 minute. Only when significant forces engage (move into potential engagement range meaning the effective range of cannon) would the engine switch from strategic time to tactical time. Significant forces would probably be about 5% of total forces or if key units (cannon, generals are in imminent danger). Engagements smaller than that would be resolved by the engine in strategic time. To back this design, the building of towns on map would no longer occur. The player would receive units from off-map by triggers. In this respect, the game is a bit like a campaign. Building on map would consist of peasants and engineers working to create forts and outposts. To start the game peasants and engineers of a set number are on map (as formations essentially on the strategic map). They must first set up the home base (it's not a town.) The building of required structures triggers the arrival of cavalry (for scouting and captures further into the map) and also of some garrison infantry and supplies. The capture of neutral outposts (towns, stores, arsenals etc.) on a designed and symmetrically fair map triggers the arrival of more troops. These troops are deployed in strategic time. Units will have considerable sight into the fog of war. Thus, a player will see enemy formations and be aware that significant formations or even vanguards coming into action range will trigger the game to switch to tactical (1:1) time. When triggered, the strategic map zooms down to tactical mini-map size and become the tactical mini-map. The full screen becomes the tactical map we are all familiar with. Engagements occur in 1:1 real time until such time as game victory or until disengagement moves engaging numbers back to less than 5% of remaining forces. There might also be a slight time delay on this automatic game engine switch back to strategic time.