Oddly, despite the widespread idea-nicking, the genre has remained relatively conservative in the face of radically new ideas.
1998's Battlezone for instance was still primarily an RTS, but one viewed from inside a powerful tank rather than an overhead perspective. You could drive around and take part in the battle, as well as order your units to assist, and it was fantastic. Yet aside from its sequel, nobody seemed to care.
Homeworld (1999) offered a similarly impressive new take on things, with its persistent carrier, and use of proper 3D space for the battles, which again went almost unused outside its own sequel and expansion pack.
Only Sins of a Solar Empire (2008) really stands out as an attempt to mine the same territory, although there have been a couple of forgotten others.
One of the main reasons for this is that RTS games are such hard work to play, and simplicity tends to work best.
There have been games that try to do clever things like have an overworld and an underworld where battles can take place simultaneously (notably 2005's Dragonshard), it tends to be too much to handle.
Much as FPS games now expect us to be able to mouselook without a second thought, so are RTS games built on the assumption that we know at least roughly what we're doing.
Anything that deviates from the standard pattern is a gamble at best, and an uphill struggle at worst. Which isn't to say that it can't ever work.
Early 3D games were generally shunned due to just being confusing – camera angles, far too many controls and so on – to the point that while like every other genre, RTS games are almost all now in 3D, the 3D has little to do with the action except making it look prettier.
We still rarely see elements like a height advantage having an impact on the battle, or aerial combat being anything other than another flat layer overlaid on the ground action. To all intents and purposes, most of these games still follow 2D rules.
One of the few real splits in recent years has been the removal of key elements to put more focus on tactical play. This can be switching resources for simple time, as with the infamous Z (the more factories you own, the faster you produce stuff), or the introduction of resources, as in World in Conflict (2007).
The ultimate version of this is games where you're not allowed to build at all, and are entirely reliant on either your starting units, or scripted reinforcements.
These tend to be less than popular for a number of reasons. While theoretically making more sense than being able to magic up a hundred grenadiers for the cost of some gold, they suffer from the fact that early mistakes can completely screw you for the rest of the map, and it's often impossible to know what's coming up.
The result is that these missions often feel more like puzzles, even when done well. Base building at least gives you a chance to hold out or turn the tide in your favour.
Ruling the world
One series of games that definitely doesn't worry about complexity is Total War, from the original Rome through to the recent Empire.
These games are so good, they were even turned into a TV show – Time Commanders – where contestants refought great historical battles in the engine.
They're easily the most impressive RTS games around, covering both general empire building from a turn-based world map, and up-close fighting in gloriously brutal RTS sequences.
This split has been done before, although rarely as impressively. Rise of Nations for instance used a Risk style world map as opposed to Total War's diplomacy and empire building, with magic cards offering boosts in each new area.
Syndicate, while more of an action game than RTS, offered something similar, with research and purchasing new equipment essential if your cyborg agents were ever going to keep up with the enemies' tech level.
Total War's complexity is largely why we haven't seen any real competitors yet – both in terms of building the game in the first place, and actually playing it. A game like C&C may not have the depth, but it has the explosions, the fast pace, and the fact that even with over a decade of complexity and bolted on gameplay elements, it's still relatively easy to understand.
At the very least, it offers the advantage that you can't screw yourself over, and the missions you're playing can, at least in theory, be defeated with skill – even if you really have to understand how you're expected to play.
As for the future, the obvious gap in the market is for the MMORTS – massive battles persisting even when off-line. It's a challenge, especially given the fast pace of the average conflict, and we've never seen a particularly impressive example.
Command and Conquer: Firestorm (the Tiberian Sun expansion pack) offered World Domination mode, with the two in-game factions taking territories based on the number of victories each side's players had had in specific areas.
Mankind (2008) would text you if your empire came under attack, but it took a long time before bases were actually able to defend themselves without their human overlord online. Much of the playerbase later emigrated to Eve Online, which while nominally an MMO, in practice shifted almost entirely to huge player-run corporations locked in RTS battles for cash and territory. So complicated is Eve Online's game world, its developer, CCP has even hired an economist dedicated to the in-game universe.
As for us, our lingering hope is that all these victories will one day pay-off as we take our rightful place in charge of real armies, fighting for the future of the world. Less dangerous than the frontline, and you don't have to be as physically (or mentally) fit.
Until then, we can but dream. And hope against hope that whoever the government picks a fight with, they don't hire South Korean generals…
First published in PC Format Issue 230
Liked this? Then check out Why are movies based on games so terrible?
Sign up for TechRadar's free Weird Week in Tech newsletter
Get the oddest tech stories of the week, plus the most popular news and reviews delivered straight to your inbox. Sign up at http://www.techradar.com/register