Geoff the Medio wrote:
We probably need to make a distinction between "what you can do" and "how you can do it". There can be choice within a branch / category of game content about what way something is accomplished, while that branch is still collectively about doing that something. If instead, multiple strategy / category branches have duplicate ways of doing the same thing, it would seem to mean a lot of duplication and loss of distinction between the available choices. Hence it's beneficial to have a player's strategy consist of picking several branches from the available options, with some variation within the branch still available.
Speaking for a moment in terms of "paths" rather than "categories", I think that regardless of which path a player takes through the tech tree, he will still want to be able to do some basic things: He will want to be able to defend his planets, he will want to be able to build ships, he will want to be able to improve production of various resources. Different paths through the tech tree will let him do some of these things better than others, but will be balanced overall and include a bit of everything. If this is the case, a path could comprise a choice of two or three categories out of several categories. Some of these categories could be about how you can do things, and others could be about what you can do. For example, there could be a category that focuses on buildings and one that focuses on ship parts. These would have high interdependency with other areas of the tech tree which focus more on resource production or space tactics or diplomacy, but nonetheless, if a player didn't go into the production category very much, there would still be techs in the ship parts or buildings category to unlock something to increase production, just not to a level comparable with a player who had also focused on the production tree (should such a seemingly limited category exist). Perhaps going through the buildings category would give some very powerful defensive buildings when combined with the space tactics category, and the ship parts category would give powerful offensive weaponry... but there would still be some defensive buildings and offensive ship parts in the space tactics tree regardless.
Regardless, if techs are grouped together in a categories, what would those categories be called, if not something about what the techs let you do?
They could be named after how they let you do something, or they could be named more abstractly, like MoO2's Constuction, Physics, Biology, Power, etc.
Reduces diversity in terms of what the cateogy lets you do? (Yes, that's the point) Or reduces diversity in by not having multiple ways to do the same thing in the category? (I don't think so)
If there are multiple ways to do the same thing, they should be in different categories, or at least different "paths".
There are indeed other ways of doing things which should also be introduced, but this fact by no means indicates that direct, game-enforced mutual exclusivity should not be included. I believe that this kind of immediate here-and-now game-enforced mutual exclusivity can add a great deal of enjoyment to the game, the absence of which cannot quite be made up for with other less direct implementations of mutual exclusivity.
Could you explain or repeat why you believe this?
It can be a difficult, heart-wrenching decision to (not) research a specific tech if you know that you'll never be able to have access to a tech again. Similarly, it adds a new feeling to diplomacy, when you're trying to get a tech that you need, but can't possibly get for yourself anymore. (Since mutually exclusive techs are under the same theory/theories, this applies even when trading with a player who is following a very similar tech path).
What is particularly intrusive about them, and how are you adding them? Likely any cross-category tech dependencies should be necessary for core theories or the majority of important or useful applications, but you seem to object to any such linking...
If core theories are dependent on theories of other categories, it reduces the ability to specialize on a single category. It should, IMO, be possible to get to a high-level theory in one category without having to research any other categories. On the other hand, it might be interesting and less intrusive if this cross category dependency applied only to applications, a mutually exclusive set of which might all be unlocked by the same two theories of different categories, for example.
MoO1&2 had 6 tech pillars or columns. There were no branches on that tree. The fact that there were randomly appearing gaps in the 6 tech categories is clever, and it made what would have otherwise been a monotonous progression of techs interesting. These gaps never interrupted your forward progress.
That's a pretty accurate description of MoO1's tree, but MoO2's tree had no random gaps. Instead, each level of a category had a certain number of techs from which you could choose, and once you chose to research one of them, you could never go back and research the others. This is the mutual exclusivity to which I'm referring. And IMO, it is better than the MoO1 approach of random gaps.
Civ (and FO) have actual trees, where each tech may require one or more prerequisites, and lead to multiple later techs. The interest is in deciding how to navigate the intertwining branches of the tree. Especially in Civ IV where you have more freedom to navigate the tree in different ways, you have interesting strategic decisions to make, though of a different kind than in MoO.
Random gaps are interesting in a MoO non-branching system, but they don't lend themselves to an actual tree. Randomly making an important branching tech inaccessible could easily cripple the unfortunate empire. There may be ways of adding exclusivity in an actual branching tree, but it won't work and feel just like MoO without getting rid of the entire concept of branches. Otherwise we may very well end up with the worst of both worlds instead of the best.
My hope is to have a system of mutual exclusivity, rather than random gaps, which as you point out, can end up crippling a player. How can you create a long term strategy if you don't even know what techs will be available? My intention is not to create a tree that works and feels exactly like the MoO2 tree, but to create one that utilizes the best of the available examples. People usually associate a system they like with its best aspects most readily. By combining aspects of MoO2 with aspects of Civ/HoI, it is possible to create a tree that fans of MoO2 will find feels like the MoO2 tree with some interesting elements of branching and interconnectivity between categories, and fans of Civ/HoI will recognize as feeling like a Civ/HoI tree with the interesting element of mutual exclusivity added in.
Personally i think our research system pushes toward over-complexity already. Adding on more "features" from other games-- no mater how good they are in their original context-- will tend to hurt FO research overall.
I feel that our current research system is sufficiently simple as to be able to profit from the inclusion of another feature.