Scope Matters

Not flustered, and if we weren't weeks away from an alpha, I might even enjoy the mental exercise, but.. just not sure how it's relevant at the moment. I have one focus at the moment, and it's initials are Planetary Annihilation.

By the way, if I sound defensive, that's not my intent. I originally posted this here because a lot of the conversations here originally inspired the blog post. I'm happy to let you all test me on my own failings when it comes to objectively analyzing feature ideas. But for now, I was hoping it might trigger some more introspective discussions of some of the oft repeated but rarely fleshed out ideas that come up here.

And the main reason for that is.. it actually helps us down the road as we start to look at other features that are most important to the community. We can come up with plenty of ideas on our own, but being sensitive and really being able to give serious consideration to some of the most popular community ideas is very important, and well thought out rationales about how players really see them impacting the game is important, especially when we don't have the mental bandwidth to analyze a million ideas ourselves.

For me personally, I also find it fun to read a well thought out idea, even if I know we won't be able to do much about if for a while. It's the random "Oh, please add X!" posts that come through that do, admittedly, get repetitive since there's rarely thought put into those requests. Or at least none put into the post, which is effectively the same thing, since my ESP is still quite limited.

I too enjoy reading those posts/discussions as they develop. Generally the "pls add xyz" threads spawn some valid discussion but it requires parsing through some pages of discussion.

Admittedly I am guilty of sometimes stating my position and only later defending it, e.g.
viewtopic.php?f=61&t=46037&start=29

I think I'd like to add one more level to the whole scope thing, in terms of getting your idea from 'OMG SO COOL' to 'Hey Uber is going to try this out'. Namely, prototyping.

If you're really exciting about an idea, and you think it'll work out, the best way to get it considered & communicated would be to prototype it. Make more than just words on a forum post, and work out the UI, gameplay effects, and possible implementations - even better, make a mod to showcase the idea. It's the best way to start to answer the final question (how much work is the idea).

I haven't yet heard an answer on this for sure but I am going to assume that part of the point of the Uber Labs will be for experiments, which if they work, could get slotted into the main game.

I completely agree with the spirit of the blog post linked in the OP, that the vast majority of suggestions read along the lines of "wouldn't it be cool to add space bears!" With a strong focus on simple, deep gameplay, suggestions like this will clutter the game without adding very much.

However, I did notice an abrupt shift in the argument used against things like weather and space combat, and the argument against localizing the economy. The basic argument against localizing the economy is that it would be burdensome to the player, whereas on the other counts the claim is that implementing them is burdensome to the developer, or that it detracts from game functionality.

Especially this section:

This actually misses the single biggest advantage which is minimizing the strength of consolidating build power. Construction in serial is more efficient than construction in parallel, and there's no cost to serial construction with your entire economy if the economy is completely universal. Ten units built one after the other, instead of ten units being built simultaneously in ten different places. But construction in parallel is more strategically interesting, as there will be more activity on the map, more locations of interest, more units in different places, more need for movement and greater decentralization of both economy and military across more space.

Localizing the economy arguably might increase the burden on the player. However another one of PA's stated goals is to have powerful command and control tools available to the player. Features like automating transportation and production which are already going to be present, but which are made less significant by a universal economy.

The "cool features" of a local economy have very desirable and significant gameplay effects. However the additional burden on the player would be relatively easy to mitigate or even eliminate with fairly simple automation. But the player would remain unable to completely serialize production on one thing at a time. Players will divide their economy and production across geographic space, and then transport units and materiel where they are needed. Both these tasks can be assisted with simple automation techniques, while their strategic impact remains.

This depends on the implementation of the feature. This is very possible, but not necessarily true.

I imagine similar to air vehicles in that they move in their own "space" layer (along with orbital structures) and have similar utility with the exception of being able to independently move between planetary bodies.

Again, I imagine similar to air vehicles. For example lets say there are 3 types. Fighter, bomber and transport. for fighter, not at all. They only interact in their own layer fighting other fighters, bombers, transports and orbital structures. Bombers Can target the ground (and maybe orbital structures), very similar to air bombers but with extra mobility (between planets) yet quite vulnerable. Transports are just like air transports, but can take units between planets for use when a unit cannon just wont due. I imagine the space transports would be the only unit able to transition between air and space (similar to how supcom's UEF Experimental Aircraft Carrier moved from sub to surface).

That sounds hard, so don't do that. You can add space combat that is satisfying and strategically rewarding without having to do that.

Like I said before, don't.

This game isn't homeworld, nor sins. It is it's own game and it's worth looking into a new way to do space combat to address new challenges brought by a genre reinventing game like PA. You already have sub naval, naval, land, air and orbital confirmed for the game. An orbital structure that produces orbital units does not sound so drastic to me.

So, to go through your list:

Space Combat!

That might take some time

Orbital "space" units. They impact by allowing easier interplanetary fighting and adding greatly adding to the feel of intergalatic forces at war.

Yes. It makes the orbital layer more exciting and important and it gives players new ways to launch preliminary strikes on incoming or foreign planetary bodies.

Of course it adds some complexity, but with the orbital layer already confirmed in your stretch goals I wouldn't say it does by much.

Without a doubt, yes. You've already developed air units which can be used for core functionality. You already met your orbital stretch goal. You already have methods for units to travel between planetary bodies. Essentially everything you need is already there. If you stop thinking about other games and focus on your own you might find you already have the tools you need.

Article was a great read. It is very helpful to see an organized test of ideas. In the article you allude to some of the core design principles you guys started with. Would you be willing to enumerate them for us. So we can be on the same page when contemplating ideas, and evaluating them using the criterion from the article.

While the temptation is there to think of homeworld as a model for space units I don't think it is the way I would go.
The orbital shell is just another 2 dimensional layer to the game and I think units that exist in the shell might simply be able to transit from one planet or moon to another and once they have arrived behave similar to air units in the control mechanics. I don't think combat in all of the in-between spaces would be relevant. these units would simply be the units you use to rapidly reinforce another base that may not have time to build the reinforcements that they need.
This would essentially make it possible to have an interplanetary force that can be built in one location and then be transitted to another when needed.
The controls would essentially be: select the group of units, set them to transit mode, select the destination world and off they go to provide immediate support to a base that may be getting overrun or is in danger of being overrun very soon.

So first off I don't make decisions directly based on core design principles, but I do let them be a guide. There are *always* exceptions.

There isn't a huge list of core principles written down anywhere. It's more about how I think about solving problems and what I thought was important about how these games work. I'm also trying to channel the spirit of TA a bit although the game has started to get it's own "soul" so to speak.

- prefer to show things in game no the map instead of in UI elements, let the game world speak for itself
- physics matter but there are always limits to this
- scalability, moddability (is that a word?) and extensibility

There are a bunch of minor points that come out of all of this.

Of course the biggest input to the game is how much time/money we have. Our by far biggest design principal is that we have to actually be able to implement the feature. As Marc has stated we have to make constant tradeoffs. Things like weather, quite frankly, isn't high enough on the list to even merit putting time into thinking about. There are literally dozens of other features backlogged that would be more important. Keep in mind we have a very tight budget / small team. So bang for buck is hugely important. Things that are simply aesthetic don't really make the cut at the moment unless it's a core feature that we are going to build a lot of future stuff on.

Generally speaking after making games for as long as we have (20 years this year in my case) is that you realize polish is often more important than number of features. I am definitely not a "kitchen sink" kind of designer who will simply put ideas into a game without solid reasons for it's existence. Unless it's a pet theory I have or something...

Like adding cats?

Nice read , I agree with the space combat and weather examples (though I am still hoping for orbital space battles which should be relatively easy to mod in anyways, right?)

about the local economy example... that is more of a design choice issue, than a scope issue right? Allowing for both would add to the scope, but mainly because of the balancing required I think. Implementing the feature on its own would be that hard I guess?

Don't get me wrong, follow the teams' vision in this regard . So choose either one the team think is best (which seems to be global economy)

However I believe in the poll a large minority (about one third) voted for local economies, me being one of them. I know people do not really know what they want, so I will not request this feature on the base of some speculations and assumptions.

I do however hope this would be one of the hooks that could be moddable. This way the team is not obliged to balance it, as it is not part of the main game, taking away the largest time cost for this feature (I would guess). I think this would be an opportunity to cater to people who like base building (like you said in your blog) as well as keeping a strong vision for the main game.