Obscured View

A few chosen words on the world of video games

A U@W design post-mortem (my point of view)

But first–

A DISCLAIMER: Since I’m no longer an employee of Petroglyph, know that this is simply my own opinion and not that of my previous employer or anyone else for that matter. I’m not speaking for them, just for myself where I see errors in my own decisions.

Also, there were, of course, others involved with some of these decisions. Games, unless they’re small indy games, are a result of team efforts rather than just one person. However, for the sake of not pointing any fingers anywhere but at moi, it’s just me, me, me, taking blame (or kudos) in this writeup for how things turned out. That’s not 100% reality, but I’ll take my share of both good and bad.

Well, I must say that U@W was one of the most interesting projects I’ve worked on. It was incredible how much went right, and how much did not on the title. The point of this post is to concentrate on where I think the design went right and where it went wrong, rather than on any other factors that appeared in the final game.

What went right for design:

  • Pre-production was long enough to figure a lot of elements out. You don’t get that luxury usually, as pre-production is the first thing to be cut down on when publishers want a game done by X date. You typically only get to fool around with a few things before you have to start creating the real game. I actually had a good chunk of time to get things sorted. A lot of the core design (like the walkers) we had enough time to figure out how they’d work, which made solving the other problems with the factions that much easier. Solving the walker issue also helped to further differentiate the other factions, since having massive, building crushing war machines available in the first 2 minutes of game play isn’t the norm in RTS, and forced the other factions designs to evolve to counter that threat. We tackled some big issues first, and that paid back in spades.
  • The factions. I am really happy with the way they turned out. They’re very distinct from each other, and each has a little to a lot that’s different from factions in other RTS games that have come before. They play very much like they were intended from the original paper concepts, each catering to different types of players and strategies. I was very worried about creating factions that were too copycat to anything from C&C, ‘Craft, Dawn, or any of the other current sci-fi RTS franchises. I killed a lot of ideas that were too similar, and yet I think I found some very fun stuff for them in the long run that isn’t mirroring other games. The key here is to always go back to your core tenets for a faction and make sure that any system or design you create for them reinforces one of those tenets. If it doesn’t, drop it.
  • The research system. This worked out rather well in providing players opportunities to get fun upgrades, but make hard choices at the same time. The original system was crazy, but where it ended up I’m pretty happy with. It was relatively easy to understand once you got into it. Although some people hated hearing in multiplayer that an enemy project was completed, it was there to light a fire under your ass to get into research and use it, as it made a huge difference.

What went wrong in design:

  • Replaced a cut faction with another rather than dropping to two. About a year into our development, I was asked to switch out one of our planned factions, the humans. I said OK and used the faction that was planned for the expansion (the Masari) rather than drop the game to two playable factions. In hindsight, this was a terrible idea on my part, but I believed that a 2-sided competitive RTS wouldn’t sell in this day and age. What I should have done is sidelined the humans and had Novus and the Hierarchy fighting around and over them. That was PLENTY for a game, with enough narrative to set up a lot of great missions. It would have also allowed more time for the two remaining factions to be fleshed out, rather than cramming a third in there was well and trying to explain their background in a limited number of missions that divided the map design pie up even further. A lesson that more is not always better, unless you mean “more refined” instead of “more quantity”.
  • Linear campaign vs. open-ended. I went with the feedback we got from players in the previous “at war” titles on this one, and it turned out to be the wrong idea. All the feedback that I was seeing was pointing to people not liking the open ended structure for a campaign, but rather preferring that the story play itself out in sequence so that you didn’t get “filler” missions and could just follow along to the end. This turned out to be a major complaint of many people because they saw the globe and wanted to play in that sandbox as well. The lesson learned here is that if you have a major component like a global mode, make sure it’s incorporated into every mode of the game, and don’t sideline it at any point.
  • No research in campaign. I believe this is now rectified in a patch (and on by default in the 360 version), but many people kept telling me that research was just too much for players – and map designers – to handle in the campaign, and I was worried about it as well. I thought OK, you’ve got it in conquest and multi-player, so perhaps we can sideline it in the scripted campaign. Bad idea. Just a bad idea. One of the key components of U@W’s factions (on-the fly customization) wasn’t seen by many players… and many reviewers, since many people only play the campaign. The lesson here is that minimizing the use of key designs that help distinguish your product in any game mode is not a good idea for any game mode.

So that’s it. Short and sweet. A few lessons learned.

1 comment

1 Comment so far

  1. UAW Design Post-mortem at words April 28th, 2008 5:58 pm

    […] Isgreen, Petroglyph Creative Director during the development of Universe at War, shares his thoughts on what worked, and what didn’t. Filed under MLP, video games, work […]

Leave a reply

You must be logged in to post a comment.