Telcontar's Warcraft II domain logo

Quo vadis, real time strategy?

The beginning

Dune from Westwood had introduced a new genre when it came out: the now very popular genre of real time strategy (RTS) games. So far, any strategy games that involved warfare had been turn based; Blue Byte's Battle Isle series and the preceding History Line were the top games of that genre. Unlike Dune, these games already involved multiplayer capabilities. However, because the different turns were not equally time-consuming for all players, there was always a player who had to wait until the other one had finished his turn. This is a major drawback in the concept of turn based games (especially if one player consults the help function while playing) and make real time games particularly interesting.


This is the key of the real time genre and had been introduced by Sid Meier's Rail Road Tycoon. Even though this game was for one player, it was now possible to respond to the opponent's actions immediately - and if you missed the opportunity, he would continue with his action until you stopped him...

Dune 2 was the first successful real time strategy game as we know it today. Featuring some improvements from his predecessor, an enthralling plot and a well made tutorial, it had set the standards for all future games of that genre. Apart from a few unit commands, it offered everything we know today: a fog of war, the fight for resources, different units with their own strengths and weaknesses and upgrades.

It was a single player game, and all opponents had been controlled by a computer; therefore, a multi player version of this game was destined to be a success.

Multiplayer games

Blizzard's Warcraft and Westwood's Command and Conquer were the first RTS games that adapted multiplayer capabilities successfully. This started an enourmous boom of the genre, and it was only logical that a sequel would eventually be released. These surpassed any expectations and are still the top games of the genre, despite some other titles that have been released so far.

With the increasing popularity of the internet, multiplayer games were no longer bound to null-modem or local area network connections. People wanted to play their games on line, and the TCP/IP protocol and Kali made it possible, along with some minor programs. The community of players increased vastly, and strategy attained a level that would never have been possible before.

New features?

However, the features of the games remained relatively unchanged. A truly remarkable addition are the spells from Warcraft I and II that elevated Warcraft's strategies to a matter beyond mere firepower. Apart from that, new ideas for the games were relatively sparse; some features had already been invented before but could not be used due to lacking CPU power. A truly three dimensional terrain and a line of sight instead of a fixed sight radius are the most obvious examples.

The unit AI has also been improved since the early days. In Warcraft II, a unit has a fairly limited life of its own and reacts mainly to your commands and not much to the situation as such. In Bullfrog's Dungeon Keeper, though, the units have almost a life of their own, specified by a wide range of parameters. They also find their way through a twisted maze, something that was not at all the case in Warcraft 2.

In the new games, there will also be some new multiplayer features added like sharing resources, units and so on. Other things have been perfected; for instance, player can now also decide to stay neutral instead of joining a particular side.

Revolutionary ideas have not been seen since Dune, though. Usually the improvements applied to the graphics engine and the interface. The gameplay remained widely unchanged, and only few new ideas have been added. For instance, Microsoft's Age of Empires combines a research tree like in Civilization with real time combat like in Warcraft. Due to stunningly beautiful graphics and a solid, fast game engine, this title is the most promising one this year (together with Starcraft from Blizzard).

Computer opponents - still the weak spot

Something that no programmer has achieved yet is creating a computer opponent whose strategy can beat a good human player. So far, computer opponents were only a challenge because they possessed far more resources to start with and sometimes even used dirty tricks. On an equal footing, computers still are not even remotely a challenge for a human player. This is a consequence of the high complexity and the huge variety of feasable actions that a RTS game involves; classical approaches to compute the optimal moves such as checking all possibilities, as done in chess games, are virtually impossible.

New approaches to these problems have to be taken. Usually, the strategy of a computer opponent is a variation of "gather a certain number of units and attack". The producers of the real time strategy games that are going to be released this year (more than twenty titles) claim that this is different in their games; however, I don't believe that until I have actually seen it. Since the major part of people play in single player mode, this will be a crucial feature for the success of a game.

The internet and continously increasing hardware resources

The hardware requirements and the latency time and bandwith of the networks are the variables that determine the development of multi player games. The hardware requirements must be lower than in single player games to increase the potential number of players. Up to now, any game has depended open a minimal hardware and set the limits accordingly. For instance, Warcraft has a limit of 600 units and buildings and a 128 x 128 map size, so it can run with 8 MB RAM. This is similar for other games. I have never seen a game yet that adjusts these values to match the hardware of the worst computer. This is something that should be improved: why do people with 32 or 64 MB RAM have to play like people with 16 MB? The programmers should let them exploit the potential of their machines.

The major limitation factor for games over the internet is the lag. The latency time depends upon countless factors: e. g. your modem, your PC, your ISP and the people connected to it, the connection that the routers choose for you. This is why you sometimes have games with no lag while at other times it is unplayable. A solution to improve the performance are better compression algorithms for the transmitted packets and a network model optimized for the internet. For instance, Warcraft II waits half a second until an action is carried out, giving the computer some time to transmit the signal and get an answer. Kali compresses the packets to decrease the amount of data that has to be transferred; however, if the game itself packed the data, better compressions would be possible since the programmers exactly know the internal format of the data.

With multiplayer projects that involve more and more players - Quake World allows 32 players to play, Ultimate Online will even have hundreds - the lag will show his immense effects more and more drastically. Usually the lagging player slows a game down for all players if the lag is very high; the most effective way to reduce the lag is to reduce the data that has to be transmitted. I hope that there will be some improvements in that matter, as well as better connections between Europe and America.

My wishes for the future

I hope that the high number of releases this year will in the long term reduce the quantity of games that are made and improve their quality. Right now, the major difference between the games are the graphics and the quality of the user interface. The brutal competition on the market forces the software producers to release some titles prematurely, without being fully tested.

While good graphics are a must to sell the game well, I am sure that people do not want to see mere clones of old games. New ideas and impulses are wanted.

Perhaps the next generation of games (that is developed after 1997) will concentrate again more on the gameplay and less on the fašade. An indication of this trend is Origin's Wing Commander 5, where the developers went away from the "interactive movie" like in WC3 and 4 back to the original game (part 1 and 2). Obviously, people have become satured with multimedia and want more than that now.


Main - Forum - Links - News - Pictures - PUDs - Strategies - Tools/Tilesets - KPUDs