What priority should multiplayer have?

Last week I wrote about whether strategy games need multiplayer.  There are definitely strong feelings on the matter.

In the article, I referred to blogs written by top reviewers Bruce Geryk and Troy Goodfellow.  I want people to understand something specifically -- I referred to them because they are top reviewers in their field. They did not take any points off of GalCiv II because it lacked multiplayer. They were putting forth the reason why many people are passionate about multiplayer in strategy games.

For the people who have a group of friends that they play multiplayer with, then a multiplayer feature in GalCiv II would be ideal.  And for those people, having multiplayer allows the game to have a much longer lifespan on their hard drives.  My favorite PC game of all time is Total Annihilation. But if it didn't have multiplayer, it wouldn't have survived very long on my hard drive. It was its multiplayer that made it so popular.

In many respects, what I'm writing about and what multiplayer advocates are writing about go right past each other.  My argument is that multiplayer advocates have plenty of options to choose from.  Not only have most strategy games of the past 5 years put a lot of energy into multiplayer (Age of Empire 3, Civilization 4, Rise of Nations, etc.) but even this year there will be turn based games with significant multiplayer components such as Space Empires V, Sword of the Stars, HOMM 5.  The developers of Sword of the Stars argue that putting in multiplayer has no negative impact on the single player experience.  I'll have to disagree.  In having either designed, developed or been heavily involved in Entrepreneur, Trials of Battle, Stellar Frontier (a multiplayer space game with the Drengin, Arceans and Terran Alliance btw), The Corporate Machine, and The Political Machine, (GalCiv being the only game I've done that isn't multiplayer btw), I think I have some experience in being able to say that yes, having multiplayer changes the way the game is designed.  When I designed The Political Machine, I imagined how it would be played multiplayer first and then wrote the AI as a simulated on-line opponent.  I think The Political Machine turned out pretty well as a single player game still, but it's a very different game than it would have been had it only been a single player game. I would have made it so that top players would have a lot more data and information to put together far more sophisticated strategies.

Making GalCiv II have multiplayer as a checkbox is pretty easy. You could just send the saved game back and forth between players. But would people be satisfied with that?  Some would.  But even doing that adds time to the development schedule. It still adds budget, and you still have a bunch of other things that could make the experience non-ideal.  I'd rather a game decide what it wants to be. If you're going to have multiplayer, do it right.  And in my experience, doing it right means having multiplayer be part of the design.  In GalCiv II, we put in the piping so that we could do multiplayer later on.  But in terms of time and energy, we wanted to focus everything on making the single player experience satisfying.

The point on multiplayer isn't whether a strategy game should have it or not as much as what priority it should have in the development of a game.  Heck, if I had more development resources, I'd take the GalCiv II engine and make an RTS version of it. I'd love to see my designed ships up against someone else's in battle.