More-Flexible Tech Prerequisites

[labgnome]

So I want over my old files, and it wasn't really as legible as I'd like, or as easy to split into section to show easily, so here is something quick and general I made from scratch. Keep in mind this is a very general idea, that's meant as a concept for a guideline or skeleton. This would be applicable to the current construction, production and knowledge branches, and maybe with some re-working stealth or defense.

General Tech Tree Branch

Tech1.png (39.2 KiB) Viewed 4653 times

The main point here is that there is a "core" or "trunk" of pure theory techs leading to transcendence in each major branch, with various application technologies branching off. We can of-course set-up whatever inter-dependencies or cross-dependencies we like. General bonus techs generally shouldn't lead to each other, and probably shouldn't be related to other applications, unless there's a good argument for a logical association or dependency.

[The Silent One]

In short, I like this concept. Each "tech level (theory)" unlocks new options (buildings, applications); the connection with other categories by "applications" is non-confusing/straightforward. Improvements don't require the player to advance in tech level. Nice.

[labgnome]

Also "theory" is a tag that techs can have and sorting legibility of the tech tree display has been an issue. Though it's well outside of my capability, if we implement this: we could possibly have the algorithm take the "theory trunk" into account when sorting the tech tree connections.

[The Silent One]

Here's a quick sketch how the growth category might look using the concept above.

- theories take long to research and are expensive, increasingly so with each "level"
- applications must be researched individually after research on the theory has been completed, but are cheaper and quicker to research
- refinements take more time and cost more with each iteration (genome bank could grant a population growth bonus increasing with each refinement?)
- interconnections between categories are only used for applications (and not presented by connection arrows)

The player can unlock transorganic sentience either following the genetics or the planetary ecology pathway. The first takes longer, but grants more applications, the latter path is shorter, but grants less applications.
There would be more theories, maybe a 4th level ("ultratrophology" or similar).

[Geoff the Medio]

The last few posts sound very similar to the Theory - Application - Refinement structure of the tech tree in the v0.3-era FreeOrion design.

Main problem is that you end up with a bunch of do-nothing theory techs that players complain about quite a bit, and which aren't really that interesting or fun to research.

[labgnome]

Here's a quick sketch how the growth category might look using the concept above.

- theories take long to research and are expensive, increasingly so with each "level"
- applications must be researched individually after research on the theory has been completed, but are cheaper and quicker to research
- refinements take more time and cost more with each iteration (genome bank could grant a population growth bonus increasing with each refinement?)
- interconnections between categories are only used for applications (and not presented by connection arrows)

The player can unlock transorganic sentience either following the genetics or the planetary ecology pathway. The first takes longer, but grants more applications, the latter path is shorter, but grants less applications.
There would be more theories, maybe a 4th level ("ultratrophology" or similar).

Not exactly

You are right in all of your points, and I'd agree with them, so it looks like we are at least onto something that we can get a number of people behind (hopefully). However I think we're not quite on the same page on a few things:

• There will probably be just one line of theories that will be just as "long" as what we currently have (although I am open so side-branches for specializations).
• No either/or prerequisites. So either both would the prerequisite; or one would be a prerequisite, and the other reduce the cost.
• Some "refinements" (I prefer "improvement") would also be dependent on the next theory tech, depending on how "big" of an improvement they are.
• Aside: though not shown, because where they would connect would vary, the different ship-lines would also be connected to a "theory trunk". So organic hulls should be in there somewhere.

The last few posts sound very similar to the Theory - Application - Refinement structure of the tech tree in the v0.3-era FreeOrion design.

Main problem is that you end up with a bunch of do-nothing theory techs that players complain about quite a bit, and which aren't really that interesting or fun to research.

I think what I am actually proposing is fairly straight-forward and actually can help us solve a number of our tech tree issues.

1. Legibility: having a theory trunk, and clearly grouping techs into the appropriate category should make sorting easier.
2. Balance: currently the way bonus techs work, makes just balancing them troublesome: they would no longer be requirements for anything else.
3. More Balance: you can beeline for transcendence, but you have to devote time and resources away from that goal for better population, research, industry weapons or ships. The same would go for influence once that's a thing.
4. Editing and Modding: changing the tech tree becomes easier, as "where" things are supposed to fit should be fairly clear, and fewer issues are created if we need to move an existing application tech.

I never played that version of FO, so *I can't speak to how research worked, so maybe this is actually a step back, but also we have moved along quite a lot since then, so we might also be at a point where that kind of strategy (or something similar) can actually work. We have a lot of things that "do stuff", but balancing them tech wise is a headache because of the interdependence. If you have to move one tech, then you have to account for all the other thech's that are dependent on it.

*I am not going to read through all 8 pages of that thread to try to figure it out either.

[The Silent One]

The last few posts sound very similar to the Theory - Application - Refinement structure of the tech tree in the v0.3-era FreeOrion design. Main problem is that you end up with a bunch of do-nothing theory techs that players complain about quite a bit, and which aren't really that interesting or fun to research.

I think the logic behind the "TAR" structure really is sound, it's the relationship between the number of theories and aplications that must be right. Granted it's not fun if you research stuff that doesn't do anything all the time. However, and please take a good look at my sketch above, if balanced right - a small number of theories, a large number of applications - the theories become thresholds that you need to overcome before you get research the juicy stuff. If properly used, it can create lots of interesting strategic options that the current tech tree does not offer. Like in the sketch above, the player can choose to research only the genetic or the planetary eco pathway to rush to transorganic sentience, or research both pathways if has enough RP or isn't in a hurry to get to the late-game tech.

Also, as visible in the sketch, the tech tree becomes much more organised. The theories help to structure the tech tree into tech levels, as labgnome suggested.

There will probably be just one line of theories that will be just as "long" as what we currently have (although I am open so side-branches for specializations).

Well, correct , I don't agree with this. Only one core theory branch may work for some categories like intelligence; but for other categories, it will just take away strategic options, making it somewhat boring. On what I agree is, we should have a clear, not too complex structure of a limited amount of theories per category that only unlock applications (buildings, ship parts and hulls, boosts e. g. for population), but don't do anything else otherwise.

No either/or prerequisites. So either both would the prerequisite; or one would be a prerequisite, and the other reduce the cost.

Why not? Again, see how it creates interesting options above.

Some "refinements" (I prefer "improvement") would also be dependent on the next theory tech, depending on how "big" of an improvement they are.

Sure, that's fine by me. On your other points (legibility, balance, interdependence etc.) I think I completely agree.

[labgnome]

No either/or prerequisites. So either both would the prerequisite; or one would be a prerequisite, and the other reduce the cost.

Why not? Again, see how it creates interesting options above.

It can be interesting, yes; but it's not KISS. I believe that the "cost reduction" rout offers a similar effect without any of the potential headache or confusion. Also, should balance issues emerge (and they will continue to), you don't wind up with de-facto "primary" and "alternate" routs, because one has all the cool toys.

FYI: given what MatGB & I seem to have in mind for growth, I've been intentionally leaving that specific branch out of this discussion, as it will probably be the most different from all of them. It would need to use a highly modified version of this system, and cost relationships will be much more important, as cost increase will need to be an important factor. I even have half a mind to say that terraforming and gaia transformation might be more at home in the construction branch just to make everything simpler. But I'd also put the starlane bore (and nexus), artificial balck hole and (probably) *scrying sphere there too (maybe intelligence? Whatever branch influence gets?).

*No I do not think it should continue to share a tech with the starlane bore, that make no sense to me from either gameplay or theoretical physics perspective.

[The Silent One]

It can be interesting, yes; but it's not KISS.

I don't agree. Especially if we switch to a system as Geoff initially suggested:

I'd like to consider a large but simple change to the tech tree: the requirement to be able to research a tech is to have any of its prerequisites researched, rather than all prerequisites as is the case now.

The tech tree stays KISS, but offers branches/options. The bifurcation is only one possible scenario. Interesting might also to have "dead end branches" that don't take you further down the "level road", but have some interesting applications to offer if you choose to pursue them. Just a single theory/level branch is dull, in my humble opinion.

I even have half a mind to say that terraforming and gaia transformation might be more at home in the construction branch just to make everything simpler. But I'd also put the starlane bore (and nexus), artificial balck hole and (probably) *scrying sphere there too (maybe intelligence? Whatever branch influence gets?).

I'm not opposed to move some techs or applications to different categories where it makes sense.

[labgnome]

The tech tree stays KISS, but offers branches/options. The bifurcation is only one possible scenario. Interesting might also to have "dead end branches" that don't take you further down the "level road", but have some interesting applications to offer if you choose to pursue them. Just a single theory/level branch is dull, in my humble opinion.

Firstly: you do know we can do that now: and technically the ships, weapons, defense and intelligence tech lines could be "dead end" branches in that they don't help get you to transcendence. Honestly I think this is where we're just having a miscommunication. What I am suggesting is that applications would be dead-ends/side-branches off of the "main road" of tech progression.

Secondly: While it's not terribly complicated, the general consensus form others seems to be that they'd rather keep AND-gates if these are introduced to the tech tree, so unless Geoff is going to do all the legwork himself, it's probably going to not be quite that simple in actual implementation. Also: tech queue sorting (as the tree is already barely legible) and AI (which is the current big obstacle to any major tech modifications) become an even bigger complication than they already are. I also don't think that it would be KISS from a standpoint of future maintenance or modification of the tech tree; which is where I want to introduce more KISS. Right now the tech tree is such a mess because we really only can put band-aids on the problems. I think what I am proposing could allow for a whole-sale transplant.

I would be amenable to or-gates in application-to-application, application-to-improvement, or improvement-to-improvement "either/or" dependence, however I'd like the "theory trunks" to stay fairly solid. The intent is that they would be an immutable framework to prune from and graft to as we need. Hopefully even allowing for variable tech trees. Basically everyone always gets the theory trunks, and gets the same theory trunks. However the branches off of those trunks can be as different as we like.

Specifically on the OR-gate question in relation to my proposal, as this is something I have started to actually put work into making this a thing on the front end: the or-gate is not something I am willing to invest my time and/or energy in relation to making it work with what I have in mind. If someone wants to introduce OR-gate functionality after I get this off the ground, they are free, I'm just not interested in making it work.

I'm not opposed to move some techs or applications to different categories where it makes sense.

I mean geoff and I might be on different pages on what changes we'd like to see in the tech tree, but I think we're both on the same page that it's abysmal. Half the dependencies don't make much logical sense, and the other half don't make much gameplay sense, really except for ships and defense (who get all the love and attention). The rest of it looks more like some "tech hairball" rather than a "tech tree". Right now we have no way to effectively prune or graft as we need to and that's really where I think the structure I'm offering gives the most benefit.

[The Silent One]

As I see it, the problem is that theories and applications aren't differentiated, and applications are not grouped with their theories.
A little variation with a limited number of forks/branches will not impair legibility.
Since my sketch didn't seem to convey this, here's an elaborated draft how, just generally, I would structure the tech tree (WIP):

[labgnome]

Your work is gorgeous...

So I'll have to get a mock-up together of what I'm thinking for growth (that overlaps with MatGB's thoughts). But basically, there would be separate side-branches for each metabolism type. So instead of getting a tech that let everyone be able to live on hostile planets; you'd get a tech that let organic species inhabit hostile planets, or a tech that let lithic species get hostile planets ect...

So the "wide" growth techs would be metabolism specific, and most of the "tall" growth techs would be general-case (I have a couple of thoughts in this area). Probably with construction branch cross-dependency. More-and-more I want to just move terraforming to construction altogether. This is where cost-influence would come in: each metabolism-specific sub-branch would up the costs for all the others. So you would have to pick a metabolism-type to focus on if you wanted to get all the adaptation bonuses. There would probably be just one theory core still, but the "adaptation" applications would be dependent on the previous ones. Likewise anything metabolism-specific would connect to these side-branches.

[o01eg]

How it should work with cross-categories tech dependencies?

[The Silent One]

But basically, there would be separate side-branches for each metabolism type. So instead of getting a tech that let everyone be able to live on hostile planets; you'd get a tech that let organic species inhabit hostile planets, or a tech that let lithic species get hostile planets ect...
So the "wide" growth techs would be metabolism specific, and most of the "tall" growth techs would be general-case (I have a couple of thoughts in this area). Probably with construction branch cross-dependency.

That sounds good to me. Maybe with one fork or so

This is where cost-influence would come in: each metabolism-specific sub-branch would up the costs for all the others.

That seems a little complicated if you ask me. I'd rather put a time-lock on the different metabolism terraforming like this:

Code: Select all

(theory) environmental adaption   500 RP / 10 Turns
    (app-building) controlled inferno environment 400 RP / 10 Turns
    (app-building) controlled radiated environment 400 RP / 10 Turns
    (app-building) controlled desert environment 400 RP / 10 Turns
..... etc .....

Anyway, I'm looking forward to your mockup and then we can discuss more about the growth category's actual layout. We should start a new thread for that though.

How it should work with cross-categories tech dependencies?

Probably as indicated in my first hand sketch, like I said, the mockup is still WIP.

[The Silent One]

How it should work with cross-categories tech dependencies?

Like this I think. Icons will disappear when the prerequisite has been researched, double-clicking add the tech to the queue.