Themed Tech Tree Fundamentals

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.

Moderator: Oberlus

Message
Author
User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Themed Tech Tree Fundamentals

#1 Post by Vezzra »

The discussion about the fundamental setup of a "themed" tech tree, or what a themed tech tree actually is supposed to be, has been scattered over at least two threads (as far as I can tell), and derailed at least one of them. So, in an effort to focus the discussion and prevent it from going on in multiple parallel threads, as well as relieving the other threads from being clogged by posts related to an actually different topic, I continue this discussion here, in a new, dedicated thread.

Originally I wanted to move all the posts that actually belong to this discussion here to clean up the other threads, but after looking closer at them I gave up on trying to disentangle the posts which would belong here from those that should remain in the original threads. It has become far too convoluted. So I just have to do with continuing here, and aks all participants to also follow up here, instead of the original threads.

Below I follow up on the discussion taking place here and here.

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

Re: Themed Tech Tree Fundamentals

#2 Post by Krikkitone »

My thoughts as to the fundamentals

Goals: Each theme should
1. have a particular "fluff feel" (that also shows up in how it plays)
2. have strategic gameplay weaknesses and strengths
3. be able to be combined with other themes

Those weaknesses and strengths should not be so severe that playing with just one theme is practically impossible (ie no ship weapons in this theme)... but it shouldn't have everything (ie no shields is ok because there are other methods of having survivable ships)

Also the techs need to not combine in game breaking ways (ie two different -50% ship cost techs..resulting in free ships)

to avoid game breaking combos
have them be mutually exclusive (buildings/focuses/policies that can't be applied to the same world at the same time)
OR
have them be duplicates (ie two separate techs in two separate themes give you access to the X building or the Y policy)



To Make sure that tech costs are balanced with both benefits and empire output (so that you can either go deep or wide)

Each tech has a base cost (roughly constant per tier) that is roughly proportional to its benefits

Either per tech or "per tier" total acquired (regardless of which theme the tier is in) there would be an exponential cost multiplier (ie x 1.1 per tech or x 1.5 per tier or so)


I prefer the tier model (since that way each individual tech doesn't really have a cost against your other techs)
...where
N techs researched in a tier (less than the total number in that tier) open up the next tier...and there is also a cheaper repeatable "theory" tech in each tier.. so you can go onto the next tier without getting any benefits of this one.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Themed Tech Tree Fundamentals

#3 Post by Vezzra »

labgnome wrote: Sun Apr 21, 2019 5:53 pm
Vezzra wrote: Sun Apr 21, 2019 3:46 pmI don't, this is the exact thing I want to avoid: that there are things which are in every theme/trunk. No trunk should have everything, and nothing should be in every trunk, IMO that's one of the fundamental principles I hoped we could agree upon.
I don't think that as things stand we will easily come to an agreement on this. I'm not completely opposed, but I think you will need to better detail your own proposal for me to get what you are going for.
Hm, I thought I had already explained a bit more detailed what I'm aiming for... well, I hope what I've already written in conjunction with what I'm going to write here will clear that up. If not, feel free to ask again. :D
20 Themes? I'm gonna give you a nice "whoa there buddy". I know one of the things that has been discussed and gotten a lot of agreement so far is that there should not be excessive scrolling, and this would be a whole lot of scrolling to find the techs in the themes you want. I like the 5 themes because they are nice and concise. A major consideration should be how easy it will be for the player to navigate this thing. Currently the proposed system would be very easy to navigate. Yours not so much.
I disagree - either I still don't completely understand how you intend to actually implement your ideas, or you haven't realized yet what you've gotten yourself into.

Please correct me if I understood something wrong, but unless that's the case, what you want is the following: essentially have a couple (currently 5 or 6) of independent "themed" tech trees. Each tree should contain everything required so you can play a viable strategy by only researching one tree. Each tree will be better at some things, and not so good at others (otherwise it would be pointless to make several different trees), aside from the difference in flavor. This means several implications/consequences:
  • Each tree needs to cover all essential functions/applications: offensive and defensive systems, both for ships and planets, troops and invasions, construction, industry, research, influence, supply, resource distribution, diplomacy, detection and stealth, ship hulls, etc. Which means most techs (or kind of techs, to be more precise) need to exist in different flavors and variations in each tree. E.g. you're going to need 5-6 different weapon lineups. Same goes for armor, shields, etc. That's an awful lot of redundancy.
  • Each tree needs to be large enough so players reach the highest tiers not before late midgame at the earliest, ideally not before late game (in a typical game). I hope we can agree that getting high end late game tier tech already in midgame should absolutely not be possible for a typical empire in a typical game.
  • Each tree needs to be large enough to provide enough stuff to research at least into early late game, ideally longer of course. Again, I hope we can agree that more or less completing the tech tree of your choosing already during midgame should be a no-go (typical game, typical empire).
I don't see how you'll be able to have each tree meet the above requirements without each tree being not much smaller than the entire tech tree we have now. Actually each tree might have to be larger than our entire current tree, considering how the new tech tree mechanics are supposed to work: You usually will have to research only a subset of the techs of a tier, with the option to skip even more of them by focusing on theory techs which are supposed to get you to the next tier more quickly (in exchange of wasting RP for nothing but getting to that next tier). So, in order to prevent players from getting to late game tech too easily/too quickly, the single trees need to be sufficiently large.

And you probably want to have enough stuff in a tree so someone pursuing a single tech tree strategy won't run out of interesting stuff to research too quickly.

Which (IMO) means, even if estimating very liberally, that you won't get away with less than 100 techs per tree. TBH I don't think that will be nearly enough, 150-200 is more what I'd expect (of course, depending on how we price the techs, especially later ones, we might be able to get down to 100-120, I assume the number of techs will be exactly the same for each tree?). That means at least 10 tiers of at least 10 techs/tier. Most likely more.

Which means we are talking about 500-1000 techs in total. I really don't think my idea of a themed tech tree will top that by much, if at all. After all, with my idea, there should be less redundancy. But anyway, if the concern is how extensive the entire tree is going to be, how difficult to display it to the player, how difficult it will be to choose the right research path through it/the right techs from it, do you really think your design is going to fare that much better than mine?
I like the concept of each theme as basically a different tech tree, each representing a mutually alien (to varying degree) approaches to technology. That is why I proposed that the primary means to get technology from other themes being tech trade or stealing through espionage. Basically each tech theme being like a theme for the empire.
So I guess I understood correctly what you want to do, and my above assumptions are probably not too far off the mark.
Mind you I am not proposing that you shouldn't be able to research into multiple themes, just that each theme should be viable on their own.
I'm wondering if things are going to work the way you envision. If each tree is designed so that you can pursue a viable strategy by focusing on one tree alone, and (as has been already discussed and, as far as I can tell, generally agreed upon) that measures need to be put in place that make researching additional trees more costly to discourage a "research them all" approach, consequently trying to get techs from more than one tree will carry a higher cost.

This will probably happen even if no extra measures to ensure that are taken, because you most likely will have to go research some stuff you don't really need/want to get the things you want to "cherry pick" from other trees. Might not be much, but probably not nothing. Which means, more often than not trying to play an optimal game will mean sticking to one tech tree. Because, after all, with the proposed new tech tree mechanics researching stuff that is in your choosen tech tree is an obvious good path, you need to research a certain number of techs to unlock the next tier anyway, and spreading your research effort over multiple trunks raises the number of techs you need to research to unlock the higher tiers, which is a definite disadvantage.

Meaning, making researching into multiple themes will be quite difficult to achieve, at least that's what I'd expect. Not impossible, but I think you won't be able to prevent that going for one tree will in most cases be the easiest choice.
I am not even proposing that each theme should be equally good in every area, in fact I'd like for each theme to have strengths and weaknesses.
Well, I think that goes without saying, and we all absolutely agree with this.
I am not completely opposed to your idea. Though, honestly what you are proposing is so different from what is currently being put together. I will say this feels too much like a departure from what I have already been sold on to jump on-board with any enthusiasm. I do not want to de-rail the work currently being done as there seems to be headway being made and I'd like to see this get worked and not drag out like the stockpile.
As much as I understand that sentiment, I'm afraid all this is going to happen anyway, regardless of what we decide in the end. The problem is, for ongoing development, while not even all of the core mechanics are in place (meaning, essential and important ones still missing which will change the game substatially when finally added), the tech tree in it's current form is actually optimal for adding in new stuff. While a heavily funtionally organized tech tree might not be desirable for some (others like you actually like a functional tree anyway), for ongoing dev work it's ideal.

If we add e.g. Influence and all the stuff that comes along with it, we'll need bunches of new techs related to all that new stuff. With the current tech tree it's easy: just add a couple of additional branches, try to price them roughly reasonably, and we're set.

With the proposed changes to how the tech tree works and the whole tech setup/layout, this will become a lot more work in the future (regardless which variant of "themed" we decide upon). Each addition like Influence and related mechanics will trigger a major revision of the tech tree, and a lot of past work will have to be thrown out of the window. Which is why, at the very beginning of this entire discussion, I asked if we really want to do this now. Efficiency wise it would be a lot smarter to do the tech tree as one of the last things, not before all the core mechanics are in place at least, then we'd spare ourselves a lot of wasted efforts...

The other thing is, concern about throwing out work that has already been done should not influence our decision on what we decide upon in design discussions. If an idea is better/more interesting/preferred/whatever, then it shouldn't be discarded just because a lot of work has already been put into another. Heck, we even ripped out things that haven't just been planned or received much thoughts, but that had been implemented and been part of the game for a long time. Consider all the effort that had been rendered wasted in those cases! But for whatever reason we decided that we didn't want to keep those things, so they needed to go.

So, if we are going to stick with your version of a themed tech tree, then it should be because it is actually better, or we (or the majority) likes it better, or it's easier to implement and my idea, while maybe interesting, much too complicated to do. But not (only!) because we don't want to throw away a couple of pages of design discussion and the hours of work that went into it.

That said, I want to note that I clearly stated the fundamental principles I thought important at the very beginning of this design discussion, when I first posted the idea of a "themed" approach to organize the tech tree. Not having trunks that contain everything essential, not having something in every branch, thus ensuring you'll have to pursue multiple trunks, but also making sure delving into too many trunks won't be viable. So it's not some wrench I decided to throw into your plans after joining the discussion late, just saying ;)

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Themed Tech Tree Fundamentals

#4 Post by Vezzra »

labgnome wrote: Sun Apr 21, 2019 6:30 pmIf you want a run-down of how the themes currently work, form what I can tell it's something like this (someone correct me if I'm wrong):
ThemeInspirationAestheticHullsWeaponsPoliticsOther
BiotechnologyZergLiving TechOrganicMonster PartsHiveBio-Modification
CrystallineKryptonian/Minbari"Crystal Spires and Togas"AsteroidLasersUtopianSubterranean
CyberneticBorgCyberpunkRoboticDrones/FightersCybernetic CollectiveRobots
EnergyProtosAlien/High-techEnergyBeam???Psionics
MechanicalTerranSteampunk/IndustrialMassiveRailgunsOppressiveMegastructures
Ok, I see. Well, that might help me better explain what bothers me/what I dislike about that approach:

My main gripe with it is how much it restricts your choices, compared to my model. I can choose among 5-6 different tech trees, and that pretty much determines my strategies in all key areas: ship armament, politics, hulls, diplomacy and so on. Take the Psionics for example: if I want to pursue them, I have to opt for the Energy tree, which also decides my choice of weapons, my diplomacy and politics, my choices of detection and stealth technologies, etc.

Sure, the system is supposed to offer me at least to also go into another tree, but a third one will already be not a viable option in most cases. But usually (as I pointed out in my post above) the best option will be to stick with one tree. Opting for multiple trees only works for choices that offer actual advantages, but not for ones that are just different, but would nonetheless be interesting to try out - if only it were possible.

This is where my approach is vastly superior (well, IMO, and I can't claim to be totally unbiased ;)): Think of it of kind of a "modularized" version of your approach. Instead of having 5-6 very large trunks/themes (which are "complete sets"), where every theme covers everything essential, you have a dozen or more smaller themes ("modules"), and you compose your very individual tech tree by combining a reasonable selection of all the available trunks (select a subset of the available "modules" and build your custom "set" from them).

So I'm not restricted to energy hulls, beam weapons and whatever the energy tree offers in other areas when I want to play Psionics, with the option to maybe cherry pick one or two things from other trees, but am completely free to for example choose to e.g. play a telepathic Lithic race with asteroid hulls, use a combination of railguns and fighters as my weapons of choice, employ Psionics and Utopian politics (whatever that means).

Or I want to play a research focused human-like race, which use organic ships, lasers, rely heavily on stealth tactics, politics (or better, society) is a collective, and whatever choices I want to make for the other key areas.

Of course I won't be able to just combine everything with everything to my hearts desire. After all, the available trunks are theme-oriented, and by far not every conceivable combination will provide a viable strategy. But I think you'll get a lot more viable choices than with a model where I usually choose one of several complete/self-sufficient tech trees (where I'm only offered already pre-composed "sets" of the theoretically available "modules", without the option the recombine the "modules" to something new).

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Themed Tech Tree Fundamentals

#5 Post by Vezzra »

labgnome wrote: Sun Apr 21, 2019 6:49 pmSo say instead of a weapons and shields category, you would have a physics and chemistry category?
Yes, that's a good analogy. It can't be too "realistic", otherwise you again would need most of the trunks/themes (after all, you can't go all chemistry while completely forgoing physics ;)), but otherwise, yes. Basically inventing some completely unrealistic fields of science/research, where, as in real life, each field can provide applications in many areas (military, industrial, medical etc.), but not all.
I could see that. Although I will say I'd try to keep the numbers down to a reasonable amount to avoid there being too many choices or having to spent too much time hunting down the technology you want.
Of course. But as I pointed out in my post above, that is going to be a challenge for either model.

Two things I guess will be absolutely necessary so players can handle the massive tech trees: you have to be able to hide/display entire trunks in the UI (the same way you can now hide/display the categories), so you can hide the themes/trunks you're not interested in. Because with either model we're going to have tons of techs which the player ususally will only be interested in a small subset of.

The other thing will be a summary description of each tree/trunk/theme, that point out the major features, strength and weaknesses of that particular trunk, so (new) players do not need to analyze the trunk in detail and maybe read the description of a lot of its techs in order to get an idea what they can expect, and how they need to adapt their strategies when selecting that trunk.
If you don't understand high-energy physics well enough how are you even gong to know how to talk to the species of nuclear-matter creatures from the surface of a neutron star?
He he. If you want to go down that road, let me humor you: That's not an application of high-energy physics to diplomacy, but to communications. Sure, the latter is a requirement for the former (you will have difficulties engaging in diplomacy if no one has any idea what the other is talking about ;)), but still two different things. :D
That is pretty close to what I am thinking. With the main difference that your initial, choice doesn't "lock" you into a particular theme, and you can choose one or another. So you can have all sorts of combinations of species and themes. I could play as Egassem with a Mechanical theme or Humans with a Crystalline theme or Scylior with a Cybernetic theme or Cray with a Biotechnology theme, just to throw some possibilities out there.
Well, of course, which is why I said "race/empire" - what you propose isn't that your tech tree gets decided by your choice of starting species, but a choice your empire makes when deciding which tech tree to delve into. Which gives you the possibility to combine each species which each theme (an empire with a lithic starting species does not have to choose the "Crystalline" theme), but IMO that's more of a flavor thing.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Themed Tech Tree Fundamentals

#6 Post by Vezzra »

Oberlus wrote: Mon Apr 22, 2019 1:14 amI can say I can't figure out more than 6 themes right now (if we add psionics to the mix).
I see two reasons for that: first, very obviously, we still lack several core mechanics, and with that all the techs that will come with them. Which, I admit, makes implementing my model at this point far more difficult than a model where you basically have multiple complete tech trees.

The other reason might be because the themes you choose are to broad and general, and include too much. You might need to come up with more narrow themes.
A trunk for the social stuff would be, as I understand it now, quite "functional"
Of course, which is why I don't advocate one "social" themed trunk, but have multiple themes where "social" or "civic" stuff makes sense. Psionics would be an example for a theme that focuses more on these kind of things, and less on weapons, shields, armor, carriers and fighters, or mining/industry.

Although I admit coming up with multiple "social"/"civics" friendly themes poses a challenge. But as long as we disregard any notion of "realism", we should be able to come up with something. ;)
It makes so much sense that different themes have different approaches (if any) to different influence-related stuff.
Of course, absolutely! Psionic and gun-boat diplomacy are going to be fundamentally different...
All themes need weapons and any other combat-related stuff that players will find essential or the themes with no weapons will be less interesting.
Not necessarily. Depends on what you pack into a theme that provides weapons. Although you are right insofar as weapons will have to be in more trunks than e.g. something exotic like the Planetary Starlane Drive or the Black Hole Collapser. In general, very important/essential functions/applications will have to be in more trunks than optional, exotic stuff. And great care has to be taken what you combine with what. If too much important/essential stuff ends up in a certain trunk, that trunk could easily become an obvious/necessary choice.

That said, not all trunks need to be equally important. I imagine several trunks that will only be interesting if you want to pursue a very specific strategy, or have access to certain exotic requirements (like a rare strategic resource). Some should be focused on late game stuff, that no one will be able to afford to even begin to delve into before midgame.

Such trunks could also be smaller than most other trunks (maybe only 4 tiers instead of 6 or 9), and/or have less techs per tier.
Or in other words, the combinations of themes will be something like one or two with weapons and maybe another without weapons.
And e.g. another one which covers maybe stealth, espionage and influence related stuff, etc. Yes, that's exactly what I had in mind when I came up with the idea. Each trunk covers several key areas, but not all. In order to cover everything, you need to combine several trunks. The more of the possible combinations are actually viable, the better. Having all theoretically possible combinations being viable is most likely an unrealistic goal ;)
seems harder to decide where to put each tech, and dunno about gameplay or balancing.
Yes, that's probably the biggest disadvantage of my approach: coming up with reasonable themes that can work and distributing all the techs in a way that things are properly balanced and actually work as we want them to is going to be difficult. Not to speak of having to invent a lot more techs, many of them kind of redundant.

But the last two things are going to be an issue with the 5-6 complete tech trees model too. Total number of techs will be even higher, and you need to come up with enough ideas to cover that.

User avatar
labgnome
Juggernaut
Posts: 833
Joined: Mon Mar 02, 2015 5:57 pm

Re: Themed Tech Tree Fundamentals

#7 Post by labgnome »

I just wanted to say two things.

Firstly, this is mostly Oberlus idea and they are doing most of the work on the project. I am mostly just giving feedback.

Secondly, you are giving me a lot to digest, so this may take a while.
All of my contributions should be considered released under creative commons attribution share-alike license, CC-BY-SA 3.0 for use in, by and with the Free Orion project.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Themed Tech Tree Fundamentals

#8 Post by Vezzra »

One other thing I think you guys should consider:

This whole discussion originated in a debate about revising the tech tree, and about the suggestion to return to the TAR model the tech tree was originally intended to adhere to. That suggestion wasn't universally liked, some objected (the main gripe being having theory techs you need to research but don't give you anything tangible besides unlocking access to other techs). Thus the idea of the tiered tech tree has been proposed: Having the techs organized in tiers, and each tier getting unlocked (that is, the techs assigned to it become researchable) if a certain limit of the techs of the prior tier get researched.

The idea was to get more flexibility/variety when deciding on a research path, and not have to go through so many techs you might not be interested in in order to get to the one you actually want (because those other techs unfortunately are prerequisites for the desired tech). Theories become a means to more quickly gain access to the next tier, but won't give you anything else (tradeoff).

IIRC one of the first concerns brought up was the question if the entire tech tree would become one linear sequence of tiers, as that apparently was immediately perceived as too dull/boring (at least by some, me among them). The answer was, of course not, we are going to organize the tech tree in multiple trunks to make things more interesting and flexible. Oberlus (I think) was the one who made some first suggestion for trunks, naturally functional/application oriented as our current tech tree has been.

It has been at that point that I brought up my idea of making themed trunks instead of functional ones, because the idea of having a weapons trunk, a ship tech trunk, a industry trunk etc. isn't very appealing to me. Instead, have trunks that adhere to a certain theme (e.g. "High Energy Physics"), which would provide techs/applications for several fields: weapons, pop growth, research boosts, etc. In order to get everything you need, you'd have to delve into at least several trunks, etc. - I've explained the concept often enough at this point I guess ;)

Oberlus liked the idea, and began developing ideas for themed trunks. Roughly at that point real life got to me and I couldn't participate in the design discussion anymore.

At no point until then the idea had been to have one trunk represent a whole tech tree, because that would defy the reason why we wanted multiple trunks in the first place. With the concept now suggested, where each theme/trunk is essentially a complete, self sufficient tech tree, we'd basically get what has been origially objected to: a tech tree is represented by one large trunk, a linear sequence of tiers you research your way through. The only difference being you can choose your tech tree out of several alternatives. But after that, if you decide to stick with only one trunk (which is intended to be viable and, as pointed out earlier, probably more often than not the best option), your research pursuit during the rest of the game is more or less the same as if there was only one trunk.

This is one of the things I specifically wanted to avoid when I made the suggestion of themed trunks, because I think it's somewhat dull (which is reflected in the fundamental principles I layed out when presenting my idea). Having to combine several trunks is one of the key elements of the original proposal. Of course you can decide to discard that part, but IMO that will take a lot of what makes the themed trunk idea so intriguing away (well, at least for me...).

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Themed Tech Tree Fundamentals

#9 Post by Vezzra »

labgnome wrote: Mon Apr 22, 2019 2:38 pmFirstly, this is mostly Oberlus idea
Um, correction: the themed trunks idea was mine. However, Oberlus picked it up and fleshed it out, and did most of the following grunt work, which I definitely appreciate very much. Throwing out ideas is easy, actually getting down to the details of the idea and doing actually usable design is the hard part.
and they are doing most of the work on the project.
Well, "most" is probably a bit much, but he is certainly one of the most productive contributors.

I'm not quite sure why you mention this...?
I am mostly just giving feedback.
I'd say you've contributed more than just "feedback". You actively engaged in the design process, contributed a lot of thoughts and ideas to help making an abstract idea (themed tech tree) a reality.

Just because I disagree on some points (even if they are quite fundamental ones), doesn't make your or Oberlus work less valuable. I give my comments here too not only because I want to take the development in the direction I think would be best, but also because I have been specifically asked to do so.

I don't intend to impose my opinion on the rest of you if the general consensus should not turn out in favor of my proposals. Ultimately, the decision on what is going to happen rests with Geoff, he is the Project Lead and Design Lead. If he goes along with what turns out to be general consensus, then that is what's going to happen.

That said, there are some things I'd absolutely object to, and probably would veto. Having themed trunks that are intended to offer everything, so that a player can viably pursue only one, but that runs out of interesting/useful stuff to research already at midgame for a typical empire in a typical game, is one of them. That is not going to happen, if you really decide to stick with your approach, please take that into account.
Secondly, you are giving me a lot to digest, so this may take a while.
Sure, there is no hurry. Take your time. It probably will be a while before I can respond again anyway. :wink:

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Themed Tech Tree Fundamentals

#10 Post by Vezzra »

Krikkitone wrote: Mon Apr 22, 2019 12:10 pmThose weaknesses and strengths should not be so severe that playing with just one theme is practically impossible (ie no ship weapons in this theme)... but it shouldn't have everything (ie no shields is ok because there are other methods of having survivable ships)
What is the issue with having to choose at least two or three trunks in order to cover everything essential? I get where having trunks being more or less complete, self-sufficient tech trees will make the design and balancing easier, but aside from that?

Re shields: well, ok, but while you might not need shields specifically, you will need some kind of defensive system (armor, flak, whatever). Going without any of those will probably not be a viable strategy.

User avatar
Oberlus
Cosmic Dragon
Posts: 5714
Joined: Mon Apr 10, 2017 4:25 pm

Re: Themed Tech Tree Fundamentals

#11 Post by Oberlus »

Vezzra wrote: Mon Apr 22, 2019 4:00 pmWhat is the issue with having to choose at least two or three trunks in order to cover everything essential?
Indeed, I don't know.
I need more insight on the actual consequences of going either few relatively-complete or many incomplete thematic trunks. And to get there I'll explore the idea in deep by trying both. I already had some job done for the fewer themes case, with some formulas to let me know the maximum potential of each theme on each relevant function (e.g. maximum population on an average scenario, maximum pop-based production boost, cost efficiency of weapons against different defences, etc.), that allowed me to grok the general theme idea and the implications when compared to functional tech trees. Now it is the turn for the many, smaller trunks case. I'll try to get just a fraction of it (something like splitting into 5-6 themes what was going to be Mech and Biotec). Then I'll have an actual answer to the quoted question.

User avatar
labgnome
Juggernaut
Posts: 833
Joined: Mon Mar 02, 2015 5:57 pm

Re: Themed Tech Tree Fundamentals

#12 Post by labgnome »

Vezzra wrote: Mon Apr 22, 2019 12:24 pmHm, I thought I had already explained a bit more detailed what I'm aiming for... well, I hope what I've already written in conjunction with what I'm going to write here will clear that up. If not, feel free to ask again. :D
I think I will. Namely, you have gone on to point out a number of potential issues with the current system but you have not elaborated on your own system from what I can tell. The current proposal comes with specific defined themes. It comes with numbers on how many techs and an idea for how thy are organized. By comparison what you are offering seems to be just a vague idea. It might be a better idea, but I certainly don't know enough about it to say if it will be or not. So yes, I want you to explain your idea in more detail. I want to know what sort of themes you envision, what kinds of technologies will go into those themes and hopefully other details about it. You make some good points, but at this stage, I don't know what would actually make your idea better.
Vezzra wrote: Mon Apr 22, 2019 12:24 pmPlease correct me if I understood something wrong, but unless that's the case, what you want is the following: essentially have a couple (currently 5 or 6) of independent "themed" tech trees. Each tree should contain everything required so you can play a viable strategy by only researching one tree. Each tree will be better at some things, and not so good at others (otherwise it would be pointless to make several different trees), aside from the difference in flavor. This means several implications/consequences:
  • Each tree needs to cover all essential functions/applications: offensive and defensive systems, both for ships and planets, troops and invasions, construction, industry, research, influence, supply, resource distribution, diplomacy, detection and stealth, ship hulls, etc. Which means most techs (or kind of techs, to be more precise) need to exist in different flavors and variations in each tree. E.g. you're going to need 5-6 different weapon lineups. Same goes for armor, shields, etc. That's an awful lot of redundancy.
  • Each tree needs to be large enough so players reach the highest tiers not before late midgame at the earliest, ideally not before late game (in a typical game). I hope we can agree that getting high end late game tier tech already in midgame should absolutely not be possible for a typical empire in a typical game.
  • Each tree needs to be large enough to provide enough stuff to research at least into early late game, ideally longer of course. Again, I hope we can agree that more or less completing the tech tree of your choosing already during midgame should be a no-go (typical game, typical empire).
I don't see how you'll be able to have each tree meet the above requirements without each tree being not much smaller than the entire tech tree we have now. Actually each tree might have to be larger than our entire current tree, considering how the new tech tree mechanics are supposed to work: You usually will have to research only a subset of the techs of a tier, with the option to skip even more of them by focusing on theory techs which are supposed to get you to the next tier more quickly (in exchange of wasting RP for nothing but getting to that next tier). So, in order to prevent players from getting to late game tech too easily/too quickly, the single trees need to be sufficiently large.

And you probably want to have enough stuff in a tree so someone pursuing a single tech tree strategy won't run out of interesting stuff to research too quickly.

Which (IMO) means, even if estimating very liberally, that you won't get away with less than 100 techs per tree. TBH I don't think that will be nearly enough, 150-200 is more what I'd expect (of course, depending on how we price the techs, especially later ones, we might be able to get down to 100-120, I assume the number of techs will be exactly the same for each tree?). That means at least 10 tiers of at least 10 techs/tier. Most likely more.

Which means we are talking about 500-1000 techs in total. I really don't think my idea of a themed tech tree will top that by much, if at all. After all, with my idea, there should be less redundancy. But anyway, if the concern is how extensive the entire tree is going to be, how difficult to display it to the player, how difficult it will be to choose the right research path through it/the right techs from it, do you really think your design is going to fare that much better than mine?
The number that Oberlus gives is that the tech tree will have about 300 techs. I don't know if that is enough, and you may be right, but this is something you should talk to them about. I really can't speak too much to this as to weather or not I even agree with your assessment. It also doesn't clarify your proposal at all.
Vezzra wrote: Mon Apr 22, 2019 12:24 pmIf we add e.g. Influence and all the stuff that comes along with it, we'll need bunches of new techs related to all that new stuff. With the current tech tree it's easy: just add a couple of additional branches, try to price them roughly reasonably, and we're set.
Again, this is probably more a question for Oberlus than for me. However my understanding is that adding in new technologies should be easy. If this is wrong then we should definitely go with a different system.
Vezzra wrote: Mon Apr 22, 2019 12:24 pmWith the proposed changes to how the tech tree works and the whole tech setup/layout, this will become a lot more work in the future (regardless which variant of "themed" we decide upon). Each addition like Influence and related mechanics will trigger a major revision of the tech tree, and a lot of past work will have to be thrown out of the window. Which is why, at the very beginning of this entire discussion, I asked if we really want to do this now. Efficiency wise it would be a lot smarter to do the tech tree as one of the last things, not before all the core mechanics are in place at least, then we'd spare ourselves a lot of wasted efforts...
I both agree, and disagree with this part. I agree that things like the influence mechanic should be given priority. However I do not think that the tech tree should be the last thing we work on. Right now it's a mess. It's cumbersome, disorganized and difficult to navigate. It needs to be fixed soon. I actually had some ideas about influence I wanted to throw out there, but I'm addressing all of this right now. Perhaps I'll get to it. That is assuming anyone will want to discuss it at all. I know there is work going on about things like influence, government and policies, and I am interested in ideas about those, but my those ideas have not generated much discussion.
Vezzra wrote: Mon Apr 22, 2019 12:24 pmSo, if we are going to stick with your version of a themed tech tree, then it should be because it is actually better, or we (or the majority) likes it better, or it's easier to implement and my idea, while maybe interesting, much too complicated to do. But not (only!) because we don't want to throw away a couple of pages of design discussion and the hours of work that went into it.
I totally agree. I'm not arguing that we should keep the system as proposed because of the amount of work in and of itself, but because I think I have a better idea of what it actually will be. I think I know what I will be getting and it looks like an interesting idea.
Vezzra wrote: Mon Apr 22, 2019 12:24 pmThat said, I want to note that I clearly stated the fundamental principles I thought important at the very beginning of this design discussion, when I first posted the idea of a "themed" approach to organize the tech tree. Not having trunks that contain everything essential, not having something in every branch, thus ensuring you'll have to pursue multiple trunks, but also making sure delving into too many trunks won't be viable. So it's not some wrench I decided to throw into your plans after joining the discussion late, just saying ;)
I don't think you're trying to throw a wrench into my plans. However I do think you are maybe a bit too hostile to this idea. I think maybe some of your issues are related to the tier, rather than TAR system, which I do like the TAR system better and it would make multi-theme strategy easier under this theme model, so that might address some of your concerns.

I do plan on replying on everything you posted so give me a bit of time.
All of my contributions should be considered released under creative commons attribution share-alike license, CC-BY-SA 3.0 for use in, by and with the Free Orion project.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Themed Tech Tree Fundamentals

#13 Post by Ophiuchus »

edit1: disclaimer - i probably misread Vezzra that he was going for many atomic functional themes instead of lean themes, so read the following with a huge grain of salt ;)
Vezzra wrote: Mon Apr 22, 2019 4:00 pm
Krikkitone wrote: Mon Apr 22, 2019 12:10 pmThose weaknesses and strengths should not be so severe that playing with just one theme is practically impossible (ie no ship weapons in this theme)... but it shouldn't have everything (ie no shields is ok because there are other methods of having survivable ships)
What is the issue with having to choose at least two or three trunks in order to cover everything essential? I get where having trunks being more or less complete, self-sufficient tech trees will make the design and balancing easier, but aside from that?
I am not completely sure what your argument is. Saying "We see no issues with XXX" is not saying "XXX is good". Probably i would hire a "good" person rather than one which "has no issues". Back from logic to tech tree design ;)

Some benefits of self-sufficient composable themes:
  • Newbies will not be completely overwhelmed (they can stick to a single trunk at the beginning and it will be ok-ish) and are able to expand their game later on
  • Themes are more "story-like" and as such they give a better framework to remember the techs for players, and designers
  • Priorities in balancing are better defined
  • Themes are more "story-like" and as such they give a better framework to design new content
  • Balancing is easier (if we give some penalties/restriction for multi-themes) - at least not worse (if there are no penalties/restrictions).
  • Themes are more "story-like" and as such they give a framework to talk about inter-connected techs (as compared to your atomic functional approach)
  • Themes are more "story-like" and may lend themselves to generate story arches in the future (should we add something like this)
On the downside one could say that you neglect thinking about all possible tech combinations. But that complexity comes from the number of techs mostly, not how you organize your branches.
Vezzra wrote: Mon Apr 22, 2019 4:00 pm Re shields: well, ok, but while you might not need shields specifically, you will need some kind of defensive system (armor, flak, whatever). Going without any of those will probably not be a viable strategy.
You missed a lot of discussion. The plan is to have in each of the themes some kind of defensive system. edit1: i think i read your quote out of context

At the state of discussion themes which are self-sufficient and composable are the gold standard I think. If Oberlus is able to pull this off I think it is the best way to go. If not, I think your idea of atomic/orthogonal functional branches is not the best option left. Also i think Oberlus is flexible enough to change course as soon as serious issues with the theme-approach surface.
edit1: One could also go for lean self-sufficient themes (so including some kind of weapons/economy boost/PP distribution.. techs, but not necessarily on covering many levels of these (e.g. only a single +1 supply). So you would be expected at least for mid and end game to research the goodies of the other themes).
Last edited by Ophiuchus on Mon Apr 22, 2019 9:15 pm, edited 2 times in total.
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
em3
Vacuum Dragon
Posts: 630
Joined: Sun Sep 25, 2011 2:51 pm

Re: Themed Tech Tree Fundamentals

#14 Post by em3 »

Ophiuchus wrote: Mon Apr 22, 2019 6:22 pm I think your idea of atomic/orthogonal functional branches is not the best option left.
I'll reply more verbosely, when I have more time, but I think that is a gross oversimplification of what Vezzra suggests.

It all boils down to "you keep using this word. I don't think it means what you think it means".
https://github.com/mmoderau
[...] for Man has earned his right to hold this planet against all comers, by virtue of occasionally producing someone totally batshit insane. - Randall Munroe, title text to xkcd #556

User avatar
labgnome
Juggernaut
Posts: 833
Joined: Mon Mar 02, 2015 5:57 pm

Re: Themed Tech Tree Fundamentals

#15 Post by labgnome »

Vezzra wrote: Mon Apr 22, 2019 1:22 pmMy main gripe with it is how much it restricts your choices, compared to my model. I can choose among 5-6 different tech trees, and that pretty much determines my strategies in all key areas: ship armament, politics, hulls, diplomacy and so on. Take the Psionics for example: if I want to pursue them, I have to opt for the Energy tree, which also decides my choice of weapons, my diplomacy and politics, my choices of detection and stealth technologies, etc.

Sure, the system is supposed to offer me at least to also go into another tree, but a third one will already be not a viable option in most cases. But usually (as I pointed out in my post above) the best option will be to stick with one tree. Opting for multiple trees only works for choices that offer actual advantages, but not for ones that are just different, but would nonetheless be interesting to try out - if only it were possible.
Again, to me this is looking like a problem with the tier system than the current theme arrangement. If this is the case then maybe we should push for a TAR system.
Vezzra wrote: Mon Apr 22, 2019 1:22 pmThis is where my approach is vastly superior (well, IMO, and I can't claim to be totally unbiased ;)): Think of it of kind of a "modularized" version of your approach. Instead of having 5-6 very large trunks/themes (which are "complete sets"), where every theme covers everything essential, you have a dozen or more smaller themes ("modules"), and you compose your very individual tech tree by combining a reasonable selection of all the available trunks (select a subset of the available "modules" and build your custom "set" from them).
So this helps me get what you are going for. I do however think that I would still prefer more broadly encompassing themes than you seem to be going for. I would say that eve if a theme is missing in one or more areas it shouldn't be a one-trick-pony, because in that case we probably should just go functional.
Vezzra wrote: Mon Apr 22, 2019 1:22 pmOf course I won't be able to just combine everything with everything to my hearts desire. After all, the available trunks are theme-oriented, and by far not every conceivable combination will provide a viable strategy. But I think you'll get a lot more viable choices than with a model where I usually choose one of several complete/self-sufficient tech trees (where I'm only offered already pre-composed "sets" of the theoretically available "modules", without the option the recombine the "modules" to something new).
I do't know that we should easily have every possible combination. At least not to the extent that everyone will be easily able to go for every technology. That will just put us back in the same situation we are in now with strategies. I will admit that the approach I have been advocating is the opposite end. Basically picking your strategy in the beginning, but I am certainly open to making it more flexible.
All of my contributions should be considered released under creative commons attribution share-alike license, CC-BY-SA 3.0 for use in, by and with the Free Orion project.

Post Reply