Review

Homeworld is a no-holds barred, fully 3D RTS. Can it be done? Yes, Homeworld is proof of that. Unfortunately, what Relic accomplished with Homeworld was creating a fully 3D RTS that was just as frustrating to play as your average RTS. So yeah, call it a success if you want. What Relic didn't do, unfortunately, was make an RTS that's fun to play. That's just my opinion, obviously. There seem to be no shortage of gamers who'll belly up in droves to play every frustrating RTS that hits the shelves. And if they're not complaining, then the developers will keep turning the crank. As for me, I played Homeworld for a week or so and walked away feeling virtually identical to how I'd felt after playing AOE2. Namely, I felt that the unit micro management completely sucked the fun out of the combat. Success in Homeworld comes just as it does in virtually every Warcraft clone I've played: churn out a bunch of units, group them together, and then point them at individual enemy units, one after the other. There is no concept of 'engagement'. Your entire fleet is generally best employed by you as a single uber-unit.

Homeworld has two fatal flaws, one in the UI and one in the single player. First, the UI: you can't give orders to your ships while the game is paused. The ability to do this would probably have mitigated the ubiquitous sense of frustration just enough to keep me playing. Truth be told, I actually consider the fundamental UI flaw to be the fact that the only means of managing the fleet is through the 3D view. But even so, if I could've paused the game, and utilized the fabulous camera UI to command my ships rather than just admire them, I'd give the game a thumbs up, despite the aforementioned micro managed combat. Pausable micro management is tolerable. Real-time micro management is a headache. The only exceptions to that are vehicular games, and avatar-based action games. In those games, you're micro managing the control of a vehicle, or micro managing an in-game avatar. Either way, the real-time micro management is intuitive, and you're generally not constantly flitting around the keyboard for hot keys.

The single player flaw stems from something that I think is actually a very good concept, which is that your fleet composition carries over from mission to mission. In other words, you don't start any given mission with a predetermined number/type of ships. You start any given mission with whatever you had when you finished the previous mission. I think in general this is a good idea. The problem manifests when you combine this with the fact that Homeworld's single player missions are highly scripted and very specific in nature. Basically, you need to know what's going to happen in the next mission in order to properly manage your resources at the end of the current mission.

Here's the best way to progress through Homeworld's single player campaign: at the end of every mission, save the game. Then go to the next mission. Plow into the new mission as fast as you can, in order to trigger all of the scripted events, attacks, etc.. This will let you determine how to best prepare for the mission. Now go load the saved game from the previous mission, and completely tap the entire map of all available resources. Spend them wisely based on what you witnessed in the next mission. Now go attempt to play through the next mission again, 'for real'. Does that sound like fun? Not to me. Homeworld is amazingly cinematic, and has a pretty good story, but it's all wasted since the scenarios are so brutally scripted that the margin for error/discovery is so tiny, and therefore the only reliable way to progress is to bulldoze your way into every mission, and then go back and play it again. This completely murdered the immersion for me, and left me totally uninterested in the story, and totally apathetic towards the mission strategies.

And lest you think I'm exaggerating, let me quote the Prima Strategy Guide (which comes with the Homeworld GOTY edition): "When you arrive, Fleet Intelligence suggests you position a Resource Controller near the asteroid vein to expedite your collection activities. Don't do this."(pg 88) And they're not kidding, if you do follow the suggestions of your own fleet intelligence, which gives the impression of providing you with tutorial-like suggestions, you make the mission ridiculously difficult for yourself. In other words, you can't even trust the game. Here's another example: "First, ignore Fleet Intelligence's suggestion to send a Probe to investigate the Taiidan position beyond the dust clouds."(pg 92) Why shouldn't you send the Probe? Because doing so triggers a scripted attack by a huge enemy fleet, and if you don't bulk up before triggering the attack, you're basically hosed. And here's a good example of needing a priori knowledge of a subsequent mission: "If you didn't build up a contingent of at least 15 Attack Bombers and 15 Defenders at the end of the last mission, it may be too late to do so now." (pg 96) Well guess what happens if you, in fact, didn't do what they described? You get to go back and replay the end of the previous mission. And if you didn't manually save the game near the end of the previous mission, with enough surplus resources to build the appropriate ships? You get to replay the entire previous mission! Wheee, are we having fun yet?

But enough of that. Suffice to say that Homeworld is busted in a traditional RTS fashion (micro managed combat) and in its own unique way (bad single player design).

Actually I've got one other nit to pick. Technology research in Homeworld doesn't require resources to complete, and I'm fine with that. However, since it doesn't require resources, you never have to "buy" tech from your own facilities, and instead you really want to be researching something ALL THE TIME. Whenever a technology is completed, your research ships go idle, and each second that they're idle is wasted research time. But Homeworld doesn't provide any way of creating a research queue. Why? I have no idea whatsoever. It's a terrible oversight, in my opinion. There is no infusion of strategy by forcing the player to manually pick a tech to research every time. All it does is punish the player who's busy micro managing the combat when a tech happens to get completed. You'll get a cheerful message from the Fleet Command stating that technology so-and-so has been completed. And from that point until you manually go pick the next tech, your research ships are just sitting there. Doing nothing. What Homeworld should've done was provide a research queue. My guess is that most players develop a fairly standard research order anyway, making the manual selection process EVEN MORE irksome. With a queue, I can plan a research path, and only adjust it if need be, based on the progression of the game. That's called "strategy".

For the record, Homeworld is gorgeous, gorgeous, gorgeous. I loved the graphics. It feels like space. The textures could use some higher resolutions, but the ships are cool and crisp. The weapon and explosion effects are great. What I really wanted to do with Homeworld, once I'd gotten fed up with the RTS combat, was to be able to jump into one of the fighters, or even a capital ship, and pilot the thing around in the middle of a big battle. I guess I should just go play one of the space combat sims ( Freespace I/II, IWAR I/II ).

Apparently, the stand-alone follow-up game by Cavedog, Homeworld: Cataclysm, addresses many player complaints regarding the UI and the single-player, but I doubt they redesigned the fundamental mechanics of the combat, and that's what really pushed me away from Homeworld. In short, I was seduced by the gorgeous graphics, only to find myself lying in bed with a run of the mill RTS, and I don't want to immediately repeat that mistake. So I think I'll steer clear of Cataclysm.

Oh yeah - the soundtrack is outstanding. Buy the GOTY edition of Homeworld if you see it. The extra soundtrack CD justifies the purchase all by itself.