Why I Hate Most Real-time "Strategy" Games

Hello again, dear readers, it's Guardian here again, writing from an "undisclosed TDY location". I've been on a hiatus for a while, so I thought I'd give you a quick update before getting to the meat of this post. It was fun to write the "Guns 'n' Gear" column, but I'm going to suspend it because I feel like hands-on T&E is a lot more valuable than my armchair speculations about toys I read about on the Internet. I just don't have the time or money to do it justice except for an occasional post on noteworthy items. Instead, I'm going to just do my best to contribute some semi-regular posts on the full range of topics here on GrogNews, from funny stuff like my "Fast Roping 101" link a few weeks back to thoughts on games, game design, and so forth. With that, on to this post's main topic.

According to this article on Ubergizmo, competitive players in real-time strategy games like StarCraft take 300 actions per minute. That is, you need to make and execute 300 decisions (moving a unit, attacking, exploiting a resource, etc.) every minute. That is, in other words, 5 decisions every second.

I've been in Tactical Operations Centers (TOCs) during real-world troops-in-contact (TIC) and other intense situations. Nobody in the TOC is making and executing 5 decisions every second. I doubt that even a world-class shooter doing free-flow dynamic CQB or a fighter pilot in a dogfight is consistently making and executing 5 decisions every second.

Sure, modern warfare, especially kinetic engagements, are fast-paced and intense, but not this fast-paced and intense. This kind of hyperactive pace detracts from depth of gameplay, richness of strategy, realism (for reality-based games), and other important design values in a game.

This is especially problematic when you consider that the best RTS AIs still leave a lot to be desired. In a real war, a leader can establish rules of engagement and expect that his units will at least defend themselves in accordance with the ROE. If he's fortunate, they will take initiative to achieve their specified objectives and satisfy the commander's intent. On the other hand, I've seen many RTS AIs where friendly forces will let themselves get flattened by loitering strike fighters and bombers unless I click the little button that says "Yes, Avenger battery, you are actually supposed to at least make an attempt to shoot down the Hinds that are slaughtering the infantry company just in front of you."

So, what do I propose instead of these hyperactive click-fests?

What I'd like to see in real-time strategy games is time compression, like in the Harpoon series of computer wargames, where the game can run anywhere from some speed slower than real time (for example, 5 seconds of real time = 1 second of game time) to a speed much faster than real time (IIRC, Harpoon went up to a 1:32 time compression factor). I think 1:1 should be the lower limit, to maintain some sense of time pressure. If there's a lull in the action, the player can crank up the time compression. If things are getting intense, he can turn it down to 1:1 so he actually execute a reasonable OODA loop. As long as the player remains in ultimate control, the game itself can (optionally) adjust the time compression automatically.

So, how do you handle multi-player? I have a simple answer: multi-player games would run at the slowest time compression rate selected by one of the players. If A wants to run at 1:1 and B wants to run at 1:4, the game runs at 1:1 unless A also selects 1:4 (or faster) time compression.

I've mostly been playing console games lately. Do any modern PC real-time wargames or RTS's get this right?

By: Guardian