Feedback: Tech tree

[Hypnotoad]

In response to Niki's Analysis:

The Mithril Resource can create quite a production boost. This has made world-project-producing cities for me. So some economic boost to Mithril working?

Perhaps it is just me, but if I want to take territory early on, the Pact of Nilhorn work best, by far. It isn't just the strength 6 units, but also the ability to bombard. I would give this a high military rating. I suppose a 5. On immortal I've taken dozens of cities with them.

What about making the increased movement rate in the horse unit line rather than in construction. That would provide it with another benefit that makes sense.

My strategy is always to go for whatever is going to provide me with commerce quickest -- cottages or plantations, depending on my initial set up. Then I pick up other techs I want while constantly researching to make sure I am ahead of everyone else (or at least neck to neck) on techs. That's the sort of game I like to play and it certainly is rewarded in this mod.

 

[Chandrasekhar]

Alright, Maniac, you said that I should keep "vanilla" FfH conversations in the related thread, so here goes. I read that tech tree article you wrote and referred me to, and I think those priciples of tech tree designing can be applied to FfH as well. (And I hope you don't mind me quoting the bolded point of each paragraph out of context... they speak for themselves.)

For a good tech tree the player has to be constantly presented with multiple interesting choices on how to develop his civ or faction.

This is a no-brainer. In a game that thrives on replay value, this is perhaps the single most important point. We're definitely in agreement here, along with just about everyone here.

Every tech (or at least tech b-line) should be equally attractive.

Depending on playstyle, of course. A builder isn't going to bee-line for Animal Handling, and an early warmonger isn't going to care much about Currency.

The cost of a tech should correspond to the benefits it gives.

Here's a point that I think we essentially agree on, but has some interesting implications.

First, techs don't cost beakers. They cost time. The amount of beakers you're producing determines how much time it takes to research each tech, and the beaker cost of each tech determines what the ratio of beakers/time will be, but that is just the method for determining how much time the tech will cost.

So, let's say you have a choice between researching Horseback Riding and Archery. Which one you research first doesn't have much of an impact on your economy, but in the time you're researching them, cottages will grow, cities will be founded, and resources will be improved. So, in effect, the first one you research is more expensive, and the one you research later is cheaper. This makes sense; having archers at turn 150 is better than having them at turn 200, so it should cost more.

Now, let's say that you chose to take Archery, and now have the choice of researching either Bowyers or other techs on the same tier as Archery. But there's a problem: Bowyers currently has a cost of 115 turns, while the other techs on the same tier as Archery have deflated to a mere 18. This is a case where the tech's cost and its value don't match up. Early longbowmen are almost certainly not worth abandoning other technologies for 115 turns when other techs on the same tier as Archery (Bronze Working, Currency, Code of Laws if memory serves correctly) cost a mere 18. So, going for Bowyers is not an option, at least at this point. It's akin to clicking on Mithril Weapons at the start of the game and waiting for your Spartaiatoi. You'll get there quicker and with more benefits if you choose cheaper techs first.

Take a moment to let that sink in. Before you researched Archery, you had X possible decisions to make about research. As researching Archery didn't open up any more possibilities, you now have X-1 research choices. And if the tech tree just endlessly branches out, then you're going to keep on having X-k research choices until your empire is advanced enough to consider choosing the next tier of techs.

Flash forward. You're now at the point where Bowyers costs 18 turns (a reasonable amount). Archery now costs a little over two turns*. Had you skipped Archery and just continued down the tree, then Bowyers effectively would cost 20 or so turns. So, whether or not you have Archery at this moment has virtually no bearing on the efficiency with which you can research Bowyers.

[size="-2"]Not an accurate representation of actual tech cost. I'm just saying that if Bowyers costs 115 turns when Archery costs 18, then Archery would cost less than 3 turns when Bowyers costs 18.[/size]

So let's look at the situation. You now have Bowyers and comparable techs (Iron Working, Summoning, Priesthood) to choose from. Perhaps you don't have Summoning as a research choice, but as you can pick up Knowledge of the Ether and Elementalism in three or four turns, it might as well be there. It is inevitable that your empire is in this situation sometime, based purely off the amount of beakers you've collected. What you've researched up to this point has virtually no bearing on your situation. You will arrive here every game you play, unless it ends before then.

So, having a tech tree made of nothing but branches, with no weaving at all, just gives the illusion of technological freedom. There is no specialization, just a choice of what bonus in your tier you want to get first, and second, and third before the next tier is "opened up" by your growing empire. I've gone off on a tangent here, but this is the logical progression of why I dislike open tech trees, so maybe it's helped you understand where I'm coming from better. Now back to your regularly scheduled response.

Techs should keep their value even if someone beat you to researching that tech first and already built the wonder/secret project that tech gives access to.

In the context in which you said it, yes, I agree with you. A tech must be valuable even after the wonder/project/hero/GP which its discovery allowed has been used. The only tech I can think of that disobeys this rule in vanilla Civ is Divine Right, and even that allowed an optional path to Nationalism.

[Do] not to give more than one one-time bonus (ie a secret project or in Civ4 also a Great Person) to a single tech.

A good bit of advice. Given the addition of another type of one-time bonus (heroes) and the fact that one person cannot have all of them, it might be more flexible in FfH, but the general idea is still very valid.

Create branches in a tech tree, that allow a certain degree of specialization... all strategies should be equally attractive.

Sort of already covered above, but I should probably add a note to this: the attractiveness of a tech is not just its function. The cost and what it leads to/requires must also be taken into account. Common sense, really.

Include some military benefits in every branch.

...

If you’re creating branches, it’s a good idea to include in every major branch a different yet similar benefit of everything that would be too unbalanced to have in only one branch.

Another thing that the FfH tech tree is sorely lacking. For one thing, it's tough to do this on a tech tree based on specialization. Setting up a branch like Education=>Horseback Riding=>Sorcery=>Divine Right is no good, of course, but Horseback Riding=>Stirrups=>Warhorses=>Domesticate Elephants is really no better. Another point for weaving the tech tree. As long as the branches remain so distinct from each other, this principle can never be applied.

So, to conclude, I think we agree on most of the points, Maniac. The main difference seems to be that you see -or- pathways as desirable to keep some specialization, while I maintain that specialization in that context can be nothing more than an illusion.

 

[Nikis-Knight]

I wonder if it's possible to reduce tech cost if it's one-time benefits have already been taken--the heroes, wonders, and religion foundings. Only really needed for techs that lead elsewhere, of course.

 

[Gamestation]

I wonder if it's possible to reduce tech cost if it's one-time benefits have already been taken--the heroes, wonders, and religion foundings. Only really needed for techs that lead elsewhere, of course.

In a way that occurs already. If you know a civ who has a tech that you don't have, your research of that tech gets increased. If you know more civs with that tech that you don't have, your research is even faster. There is an article on this in the strategy forum of CFC.

 

[Chandrasekhar]

But the effect is very minimal, and it's the same whether or not the tech has a one-time bonus.

 

[Gamestation]

Yeah I know, but the fact remains that the game can keep track of techs other civs have, alter the beaker rate of some civs for some specific techs, and track who got what tech first (sort of). Therefore the answer to Nikis-Knight's ponderings is most likely "yes."

 

[Maniac]

I'll post the "treatise" here too:

***

Personally I think just rebalancing a couple of the vanilla SMAC tech tree issues wouldn’t suffice at all to create a good tech tree for C4AC. Instead I think a new tech tree should be created from the ground up if you want added value from the good new Civ4 features. So I think it’s pointless to comment on specific issues with the original tech tree. That's why I'd like to give some guidelines I used while creating the new tech tree for SMAniaC. To me these seem rather obvious, but I’ve seen few other trees who don’t continuously violate these “principles”, so I guess it has value to write them down.

The basic principle is “one choice = no choice”. For a good tech tree the player has to be constantly presented with multiple interesting choices on how to develop his civ or faction. That’s all you have to keep in mind. All the rest I’m writing here is just a bit of elaboration on how you could do that.

One major difference between Civ4 and SMAC I should mention right away here, because it affects tech tree creation greatly (and actually also makes a lot of what I’m writing here irrelevant :-s ). In Civ4 you can give each tech its own tech cost which isn’t affected by in-game factors. In SMAC the tech cost depends on how many techs you have already researched. This creates a very strict condition for good tech tree creation: every tech (or at least tech b-line) should be equally attractive.

This condition isn’t met at all in the original SMAC tech tree. There it is always best to research towards the Industrial Automation, Environmental Economics and Doctrine: Air Power b-lines. Most of the goodies are concentrated there, so the rest of the tech tree is neglected until the three aforementioned tech goals are reached. People often even refuse to trade for some techs because the tech cost increase outweighs the benefits some tech gives. In SMAniaC I feel this is less the case. There I’m practically always happy to get or trade any tech, despite the tech cost increase. (I could write something about each tech and every change I made from SMAC to SMAniaC as examples of rebalancing, but that would be pointless if one is unfamiliar with SMAniaC.) Tech cost increase due to number of techs doesn’t exist in Civ4, so there the condition is much less strict: the cost of a tech should correspond to the benefits it gives. Just decrease the cost of a less valuable tech. Something else I want to note: if all techs are equally attractive, it becomes impossible for the AI to make stupid research choices => better AI behaviour yay! :b: Unlike original SMAC where the AI researches eg Polymorphic Software early…

An addendum I kept in mind to the guideline of keeping all techs equally attractive: techs should keep their value even if someone beat you to researching that tech first and already built the wonder/secret project that tech gives access to. Put in other words, when deciding if two techs or b-lines are equally attractive do not count wonders as they are one-time bonuses. Ideal would of course be if all b-lines have a wonder as a reward for being the first to research it. More concretely this means that in my SMAniaC tech tree all techs (with the exception of one endgame tech) offer some benefits besides a possible secret project. Vanilla SMAC techs such as Optical Computers or Applied Relativity are big no-no’s IMO.

I also made sure not to give more than one one-time bonus (ie a secret project or in Civ4 also a Great Person) to a single tech. In the original SMAC tech tree this means Mind-Machine Interface and Self-Aware Machines are bad as they both give access to two secret projects. What’s the point of that? One secret project is enough as a reward for being the first to research the tech. It would be better to grant that secret project to another tech and create one more interesting choice and reward elsewhere. Once again this principle can be treated much less strict in Civ4 though, as there you can give wonders multiple tech prerequisites.


Another big recommendation I’d like to make is to create branches in a tech tree, that allow a certain degree of specialization. From the perspective of one game, specialization basically means that it’s a best to closely follow a certain research path. This means once you’ve chosen a certain path somewhere in the beginning of the game, the following choices become fewer and less interesting. This kinda runs counter to what I’ve advocated until now. But of course from the perspective of several games, specialization means that you can pick one faction and play successfully a certain way, and play a second game with another faction, and have the choice to play a successful game in a completely different way. This is largely inexistent in original SMAC where the IndAut-EnvEcon-D:AP research path is more or less the best no matter what faction you play. So in short, in addition to all techs being equally attractive, on a larger level all strategies should be equally attractive. And before creating a tech tree, you should have a clear picture of what different strategies you want to include in your game. In SMAniaC I hope I made fungal terraforming much more viable as an alternative to the usual tree farms and condenser farms. In vanilla Civ4 the goal should be to make a great people focus as attractive as cottage spamming (something which they failed to do IMO).

On the subject of specialization and AI behaviour, clear branches significantly improve AI research strategy. For instance, when you look at the SMAC tech poster, the Explore, Discover, Build, Conquer categories might as well have been assigned randomly. On the other hand, a quick glance on the SMAniaC tech tree should (hopefully ;-) immediately reveal a distinct Explore branch and a Discover/Conquer branch. Here the focuses the AIs has are relevant and my factions follow very well the tech branches I want them to follow.

If you’re creating branches, it’s a good idea to include some military benefits in every branch. Otherwise one would be forced to switch out of your specialized b-line and have to start researching the military branch. In SMAniaC, while most stuff can be gotten from the Discover-Conquer branch and therefore is the best choice if you have conquest ambitions, some conventional weapons and armour also become available in the Explore and Build branches. And of course the Explore branch also offers psionic warfare as an alternative, much earlier and cheaper than in vanilla SMAC. Vanilla Civ4 also does a good job at offering military benefits at about any research path you follow. This is not at all the case in vanilla SMAC. In a game with decent opponents, you have only one choice err I mean no choice but to rush to needlejets and choppers asap.

This of course counts for every must-have benefit without which it is almost impossible to play the game, not only the military. So the above guideline could be generalized to: "If you’re creating branches, it’s a good idea to include in every major branch a different yet similar benefit of everything that would be too unbalanced to have in only one branch." To give some examples: infinite movement in SMAC. I gave magtubes to the Discover-Conquer branch, but Psi Gates to the Explore branch. I gave Paradise Gardens (nut sats in SMAniaC) and Resonance Grids (energ sats) to the Explore branch, and Cloudscoops (min sats) to a Build-Conquer b-line. Of course these are just SMAC-specific examples. In Civ4 magtubes could simply have limited movement like railroads, instead of SMAC's infinite movement.

To show how I implemented the ‘the more choices the better’ principle on social engineering:
In vanilla SMAC there is really only choice on the Values SE line: Wealth. This is because Wealth can be researched fastest, comes coupled with the great benefit of crawlers, and the alternative are higher level techs. In SMAniaC Wealth/Plutocrat stays with Industrial Automation, but that tech is made somewhat less valuable by moving Hab Complexes and the Planetary Transit System to other techs. More importantly, Power/Junta is moved down to a level 2 tech (and only gives -1 Industry), so you have an alternative much faster. And instead of the default setting giving no benefits, you already start the Values/Ruling Elite line with the Technocrat ruling elite (Knowledge basically). So instead of one/no, you have three choices: research nothing, research High Intensity Lasers or research Industrial Automation.


That’s all I can think of to say right now. I hope it’s of some use. Perhaps it’s too general, and it would be better to write down some concrete examples - I don’t know. In any case, my bottom point, personally I feel that we should be willing to do as many gameplay changes as we like in C4:AC. No need to have a clone!

 

[steel.stiletto]

Sorry to possibly sound stupid, but what is SMAC?

 

[Sureshot]

Sid Meiers Alpha Centauri i think

 

[Maniac]

This is a no-brainer. In a game that thrives on replay value, this is perhaps the single most important point. We're definitely in agreement here, along with just about everyone here.

Yeah I know. I just wanted to say it explicitly because I have seen very few mods who apply this principle well.

So, having a tech tree made of nothing but branches, with no weaving at all, just gives the illusion of technological freedom. There is no specialization, just a choice of what bonus in your tier you want to get first, and second, and third before the next tier is "opened up" by your growing empire. I've gone off on a tangent here, but this is the logical progression of why I dislike open tech trees, so maybe it's helped you understand where I'm coming from better. Now back to your regularly scheduled response.

Good point. However I don't understand how your proposal of adding more AND prerequisites to techs will improve on that. The tech tree is very open now, meaning there are indeed many de facto unused choices. However won't adding AND prerequisites cause the opposite problem, the tech tree being too closed? While I of course have never researched a 2000 beakers tech early in the game before I have researched all tier1 techs, I have for instance already researched Iron Working before Festivals. In my current RoM Hippus game, the first tech I researched was Horseback Riding. In your proposal I would already need Exploration (this I researched 4th!) and Cartography for that. Likewise, with the archer now being useful in more situations, I can imagine researching Archery before Festivals. Just giving these examples to show that some people (at least I) do specialize more than you might expect. Besides, while adding more prerequisites would remove the illusion of technological freedom, it doesn't solve the cause of this illusion. It would be at best a cosmetic change.

In the context in which you said it, yes, I agree with you. A tech must be valuable even after the wonder/project/hero/GP which its discovery allowed has been used. The only tech I can think of that disobeys this rule in vanilla Civ is Divine Right, and even that allowed an optional path to Nationalism.

In FfH there are unfortunately many more techs like this.

Another thing that the FfH tech tree is sorely lacking. For one thing, it's tough to do this on a tech tree based on specialization. Setting up a branch like Education=>Horseback Riding=>Sorcery=>Divine Right is no good, of course, but Horseback Riding=>Stirrups=>Warhorses=>Domesticate Elephants is really no better. Another point for weaving the tech tree. As long as the branches remain so distinct from each other, this principle can never be applied.

Really? Personally I think FfH does a pretty good job with this. There are military units, civics, terrain improvements in about every branch. A nice example: in vanilla there is only good way to reduce city defenses, with siege weapons, making warfare practically impossible without them. In FfH you can still get catapults with Construction, but you can also go the religious route for the Ritualist's and Confessor's Ring of Flames, you can go the Sorcery route for Fireballs, you can go the Summoning route for expendable cannon fodder, you can go for the Nilhorn, and in RoM you can go melee for a little bombarding if you don't want to pursue any of the other branches.

Anyway, I agree the tech tree could use change. Horseback Riding=>Stirrups=>Warhorses=>Domesticate Elephants is indeed a very boring design. I just don't feel inspired for something better at the moment. Personally I feel a tech tree should always make sense from a realism point of view. And making cartography a prerequisite for horseback riding doesn't sound real. You can perfectly learn to rise horses without knowing how to make maps.

 

[Kael]

Anyway, I agree the tech tree could use change. Horseback Riding=>Stirrups=>Warhorses=>Domesticate Elephants is indeed a very boring design. I just don't feel inspired for something better at the moment. Personally I feel a tech tree should always make sense from a realism point of view. And making cartography a prerequisite for horseback riding doesn't sound real. You can perfectly learn to rise horses without knowing how to make maps.

Dont get to caught up on the name of the tech. What is more important is what knowledge does it represent. Cartography is more than just the ability to make maps, it implies some distinct scientific and physical knowledge about the world.

Im not saying the current tech tree represents this well (or at all), but just as a general rule when doing design on pieces like the tech tree you have to boil large abstracts down into small bite sized pieces. Just because its more interesting than studying "arcane knowledge #6" or "economic studies #4" we place discrete labels on them and try to attach that label to some specific flavor.

Stuirrups is more than just the invention of the stirrup. Feudalism is more than just a concept about a government system.

So lay out a tech tree that gives worthwile options as Chand mentioned, (do I wan to increase my economy, my military, my infrastructure, etc?) then try to attach as much flavor to each step to make it interesting for the player (and the more flavor on each tech the more interesting ideas you can attach to it).

There is definitly some great examples in the tech tree design, I love how elementalism or necromancy can take you to summoning but which path you pick decides the sort of nodes you can upgrade to. And there are some that are not well devleoped (as you have mentioned the mounted line is lacking).

 

[Maniac]

Dont get to caught up on the name of the tech. What is more important is what knowledge does it represent. Cartography is more than just the ability to make maps, it implies some distinct scientific and physical knowledge about the world.

I know I know. I'm just more perfectionist about that sort of thing than average.

 

[Unser Giftzwerg]

Good point. However I don't understand how your proposal of adding more AND prerequisites to techs will improve on that. The tech tree is very open now, meaning there are indeed many de facto unused choices. However won't adding AND prerequisites cause the opposite problem, the tech tree being too closed?

I suspect there are two general reasons why Chand is thinking of more AND pre-reqs.

One reason is to prod the AI into learning certain key techs even if they have to be dragged kicking and screaming down an under-used tech path. For instance if Archery was a pre-req for all the religions, most AI civs would invent it early. AI civs would therefore tend to build stronger defensive units early on in the game. I nthis case teh AND opre-req is used to drive a desired behavior. (This is just an artifical example off the top of the head and not a proposal.)

The second reason is to slow down progress at certain points on the tech tree. As Chand points out, they key metric for technological progress is turns per invention, not beakers per invention. AND pre-reqs introduce speed bumps by ensureing low-level techs get researched in the 'appropriate era'. For instance, before version 0.16 it was often possible to bypass Agriculture until after perhaps a dozen techs were researched. Ludicrous situations would arise where a nation knew how to snap their fingers and make a ddesert bloom, but sticking a seed in the ground and sprinkling it with water was beyond their grasp. When Agriculture is finaly researched, it might take 2 or 3 turns. Occasionaly an AND pre-req in a key spot is good from that persepective.

Chand can speak better for himself than I, of course. But that's what I gleaned from his earlier posts.

As for me, my opinion is starting to solidify around the notion that FfH can benefit most from a good solid look at its tech tree. There's already massive variety and imagination and new ideas in the mod. The problem seems to be in getting them into play and interacting at the same time. IMO opinion it's sort of nice a new version is months off, because that time can allow some serious though on the tech tree.

 

[Chandrasekhar]

Good point. However I don't understand how your proposal of adding more AND prerequisites to techs will improve on that. The tech tree is very open now, meaning there are indeed many de facto unused choices. However won't adding AND prerequisites cause the opposite problem, the tech tree being too closed? While I of course have never researched a 2000 beakers tech early in the game before I have researched all tier1 techs, I have for instance already researched Iron Working before Festivals. In my current RoM Hippus game, the first tech I researched was Horseback Riding. In your proposal I would already need Exploration (this I researched 4th!) and Cartography for that. Likewise, with the archer now being useful in more situations, I can imagine researching Archery before Festivals. Just giving these examples to show that some people (at least I) do specialize more than you might expect. Besides, while adding more prerequisites would remove the illusion of technological freedom, it doesn't solve the cause of this illusion. It would be at best a cosmetic change.

I was going to type up a big reply to justify my opinion here, but Unser Giftzwerg pretty much nailed the main points that I would make. There's a peripherary point that I should make that helped me reach my opinion, though.

I think you're operating under the assumption that AND requirements inevitably slow down tech progress. This actually doesn't have to be true at all. As the cost to get to a tech is really its cost plus the cost of the prerequisite techs, the cost of the destination tech can be lowered by something like the amount of the extra tech that's required to get to it.

Look at it this way: if you just need AH to get to Horseback Riding, and AH costs 100 while Horseback Riding costs 500, that's 600 beakers to get horsemen. If you need Cartography and AH for Horseback Riding, and Cartography costs 100, too, then Horseback Riding can cost 400, and the appearence of horsemen doesn't have to be delayed. If the tech tree is woven in such a way, then your journey towards a specific goal (horsemen) gets more intermediate features spread along the way (pastures, PotN). It ultimately makes things more interesting.

Having more OR linkages between techs is a bit different. Firstly, the base techs have to be of very similar costs. If tech A or tech B lead to tech C, and tech A costs 800 while tech B costs 1500, then tech C can potentially be 700 beakers cheaper than it would be if the costs was balanced with respect to tech B alone. Also, it raises the problem of turn 700 agriculture: if tech A or tech B lead to tech C, and tech B doesn't lead to anything else, then you can expect it to be researched much, much later than other techs of its cost (unless it has a very useful ability). If every tech that had an OR requirement leading from it also had an AND requirement, then I'd have no problem with it.

Let me also throw in a useful anecdote from vanilla Civ that I'm fond of. When I played vanilla Civ, I prioritized the Apollo Program. I never had a really high producing city, so I needed to get it started early to have a chance in the space race. Now, because I prioritized the Apollo Program, I almost never had access to aluminum (which required Industrialism). If the vanilla tech tree was unwoven like FfH's, I would have had 50 turns of sitting there managing my empire and watching the counter go down... not very fun. But the way that the tech tree actually was set up, I ended up getting all sorts of little benefits that I hadn't prioritized, but were nice to get all the same. Infantry, coal, factories- I ordinarily wouldn't have gone off the beaten path to get these things, but having them sprinkled along the path to aluminum kept the game fresh and interesting.

And in response to everyone else (hopefully this won't take as much computer ink):

Dont get to caught up on the name of the tech. What is more important is what knowledge does it represent. Cartography is more than just the ability to make maps, it implies some distinct scientific and physical knowledge about the world.

Im not saying the current tech tree represents this well (or at all), but just as a general rule when doing design on pieces like the tech tree you have to boil large abstracts down into small bite sized pieces. Just because its more interesting than studying "arcane knowledge #6" or "economic studies #4" we place discrete labels on them and try to attach that label to some specific flavor.

Stuirrups is more than just the invention of the stirrup. Feudalism is more than just a concept about a government system.

So lay out a tech tree that gives worthwile options as Chand mentioned, (do I wan to increase my economy, my military, my infrastructure, etc?) then try to attach as much flavor to each step to make it interesting for the player (and the more flavor on each tech the more interesting ideas you can attach to it).

Very important, of course. We'd all much rather have "Stirrups" than "Tier-3 Light Cavalry Enabling Tech." I guess the biggest hangup with that part is that it might as well be called Tier-3 Light Cavalry Enabling Tech for all the effects it gives you. It's a real tough problem to solve, and tech tree weaving certainly offers no remedy, so I really don't know what the best course of action would be to remedy it. With the large amount of tier-3 and tier-4 units in the game, it's a surprise that so many of them have managed to avoid this problem.

There is definitly some great examples in the tech tree design, I love how elementalism or necromancy can take you to summoning but which path you pick decides the sort of nodes you can upgrade to. And there are some that are not well devleoped (as you have mentioned the mounted line is lacking).

Yes, I like that part very much. It has absolutely no parallel in the vanilla tech tree design, yet it adds so much to that of FfH. Whatever you do to the rest of the tech tree, I hope you leave that part more or less as-is.

As for me, my opinion is starting to solidify around the notion that FfH can benefit most from a good solid look at its tech tree. There's already massive variety and imagination and new ideas in the mod. The problem seems to be in getting them into play and interacting at the same time. IMO opinion it's sort of nice a new version is months off, because that time can allow some serious though on the tech tree.

My thoughts exactly. Fall from Heaven has so many creative and amazing features that would seem impossible to anyone looking at vanilla Civ, and it has been getting more features every month. Perhaps the current tech tree design was adequate when it was introduced (I wouldn't know- I believe I joined the community shortly after the new tech tree design), but the mass of new stuff has ended up clogging some branches while leaving others relatively bare. Since we have so long until the next phase, this seems like an excellent time to reevaluate every aspect of the tree and perfect it while we don't have to deal with new features showing up in the middle of our work.

 

[Chandrasekhar]

Ah, it looks like TheLopez finally released that tech tree editor. That might make this a bit easier.

 

[Unser Giftzwerg]

Here is an outline of a tech tree structure I feel might be useful to FfH. I think it has potential to reinforce the themes of variety and civilization specializations. I think it can help with the timing of military units, that is, fewer instances of Macemen or Hv Crossbowmen running amok over masses of Warriors.

It should offer potential to hammer out some of the pequiliarities of the Agnostic Grigori. It might be possible to craft some agnostic-specific techs.(This aspect just occured to me, so my thoughts are particularly nebulous here. But it seems to have potential.)

And I think it offers potential for a unique funtionality in Civ mod tech trees: It should be possible to cause different civilzations to pay different costs to obtain the same technology. (At least it's unique to my extensive knowledge of Civ mods ... FfH and, um... nevermind.)

The cost of this proposal is added complexity to the overall tech tree.

===Status Quo===

The current tech tree (TT) is wide-open for development by every civilization, with a few isolated exceptions. These exceptions are things like Arete and Guardian of Nature, which depend upon state religion. Anotehr type is the Seagoing tech, known only to the Lanun.

===Proposal (Outline)===

1) Identify Key Civilization-Defining Criteria: Currently this list contains the state religions. There is a special tech and associated civic with each. Civilzations cannot research these techs unless they are declared to the cooresponding religion. The new tree can use the same list. Another set of criteria might be the three alignments. Human/nonhuman could be yet another. Stone/Copper/Iron/Mithril weaponry might be yet another.

2) Set up small tech branches for each criterion. Curently each of the main religions has a single tech associated with it. The idea is to expand those single techs into small offshoot branches each containing perhaps a half-dozen separate techs. All the techs on such a privlidged branch would be open only to those civs that met the criterion.

3) Restrict progress down privlidged branches not by large beaker costs, but by AND/OR linakages to the main TT trunk. That is, tie progress down specialized paths to progress on the unspecialized, main portion of the TT.

I think these three rules sum up the proposal, at least on the level of a rough outline. But I doubt this post makes much sense by itself. So I'll provide a couple examples in the next message or two.

 

[Unser Giftzwerg]

Example One - Discovering a Religion

Currently, the first civ to research a religion gains quite a few rewards for that tech: The religion appears in a city, an acolyte-type unit is gained, the ability to build that religion's Holy City is gained, and all other civs are locked out of gaining these prizes. Furthermore, temples can be built, acolytes can be built, special combat units can be built, and other miscellaneous benefits accrue. Another single tech is allowed when the religion is adopted by the state. These latter benefits can be learned by other civs.

So let's look how this might be modified, using the Runes of Killmorph as our example.

Instead of Way of the Earthmother -> Arete we might have

..........................Way of the Earthmother......................
........................./...............................\.....................
........Call of the Earthmother..................Arete...............
......./...............................\...................|..................
Rally to the Earthmother....Deepshafting....Gem of Killmorph

Way of the Earthmother - Introduces the religion to a city in the civilization (which is starred to allow future creation of the Holy City), creates an Adept-type unit. Allows adoption of Runes of Killmorph as state religion.

Call of the Earthmother Requires Philosphy Introduces the religion to a random city in the civ, allows construction of Temples to Killmorph, allows construction of Adepts (or whatever they're called) allows construction of Soldiers of Killmorph.

Arete Requires Priesthood Allows Arete civic. Adds +1 hammer per mine. Adds +1 hammer per quarry.

Rally to the Earthmother Requires Smelting Allows construction of Paramander units. Allows Guerilla II promotion.

Deepshafting Requires Engineering Allows construction of Deepshaft mines. These are the unpillageable, extra-chance-to-find-resource mines (Dwarven mines) used in the current game.

Gem of Killmorph Requires Mithril Working and Medicine Allows construction of a National Wonder, Runestones of Bambur. RoB functions like Tablets of Bambur in that it helps spread the religion and grants an Earth Mana. It lacks the cultural and GP effects though. Invention also creates three new metal or stone resources within cultural boundries. (Stone, Marble, Gems, Copper, Iron, Mithril.

(The Khazad could be granted knowledge of Deepshafting at game's start. Or it could be known to them alone, like Seafaring.)

Beaker cost: CallotE, RttE, and Deepshafting would have a low beaker cost. For example, if Way of the Earthmother cost 500 beakers to invent, then Call of the Earthmother might cost 125 beakers, or about 25% of the cost of a mainbranch tech of a similar tier.

The left side of the privlidged branch would contain cheap beaker cost techs. These represent quick surges in knowledge that occur when key technologies are learned. Think of them as a limited set of "2 for 1 techs". When a certain main trunk tech is learned, a second, privlidged tech can be learned in only a turn or five ... very rapidly.

In contrast, the right side of the privlidged branch (PB) represents solid advancements in and of themselves. Costs on the right side of the PB would be set to about 75% the cost of a main trunk tech on a similar tier.

===Effects===

In this example Runes of Killmorph has had the status quo benefits rearranged then augmented by a couple new features. It was created with a compromise in mind. That between the "all or nothing" approach to inventing a religion, versus the "remove the risk" approach.

Here the first civ to invent a religion gets it spawned in one city for free, plus it gets an Adept. However it still can't build temples or do much more until the learning quick follow-on tech, Call of the Earthmother. Inventing this will spawn Runes in yet another random civ, even for the first civ to invent Way of the Earthmother. So the first civ to invent a religion will effectively gain teh religion in three civs to begin with. (Or two civs plus a free 'medic' unit.)

Also-ran civs are not screwed forever, however. They can gain the follow-on tech, so the investment in Way of the Earthmother is not wasted. Gaining Call of the Earthmother will spawn the religion in one city and will alllow it to spread. So a desired religion can be gained, but at a significant disadvantage. At the upper end of TT progression, even the Holy City can be duplicated. Allowing such national wonders might have a benefial effect upon Religious Victory strategies.

All practicianors of the religion will have to choose between investing in PB techs or main trunk techs. As a result we might have teh happy situation where it is possible to manage a civ to specialize in religions activty or to just pay lip service. Clever selection of the techs available in PBs will make these choices difficulty (and in some cases, mutually-exclusive) thus enhancing the sense of specialization.

===Disclaimer===

The tree as described above is purely for illustrative purposes. I do not mean to suggest this example is polished or without error.

 

[Unser Giftzwerg]

Example #2 - Military Applications

{Note: A new term has been coined to replace 'Privelidged Branch' ... Frond}

One are in which FfH consitantly underperforms is in getting all the various nifty units to fight one another. All too often higher-tier units appear and begin the process of slaughtering hapless Warriors or Archers. "Recon" units appear early and dominate the battlefield for decades. As the game progresses the disconnections grow larger and linger longer. Macemen are seen often in most every game, but how many Pikemen have you built since teh release of 0.16? Look at the beaker cost to move from Macemen to Pikemen ... why oes it take so damn long to learn how to attach a pointy bit to the end of a long stick?

This problem is a vanilla problem, of course. It is not fair to just dump on FfH. It is a bit more pronounced in FfH though. There are more specialized training buildings, the beaker cost between techs is greater, and unit variety by comparison is beyond compare. Fronds can be used to address this. They need not be restricted to religions only.

Here's how it might work.

Frond of Stone & Sinew

Not really a "Frond". This refers to military capabilities at game's start. that is, al civs can build Warriors and Scouts (or the racial substitutes.) However, one change is reccomended, creation of a third, "proto-Archer" unit. A unit to fill such a bill might be STR 3, an inherent -40% STR on attack penalty, and unable to learn Move promotions. I firmly believe every civ in this game needs acess to a cheap, able city defender at every technological age.

Frond of Copper & Pottery
.......Bronze Working and Horseback Riding........
................./........./..........|...............\............\
Sandcasting..Moving Parts..Assembly..Grommeting...Metalplating

Sandcasting: Allows Axemen, Remove Forest
Moving Parts: Allows Hunters
Assembly: Allows Archers
Grommeting: Allows Horsemen
Metalplating: Allows Copper weapons in the Doviello/Merc weapon quality system.

Secondary pre-requisites may or may not be needed this early into the game. The concept of tying all Tier-2 units to both Bronze Working and Horseback Riding is what holds appeal for me. Doing so requires a substantial investment before moving up to Tier-2, not just a happy side-effect of rushing towards Fellowship of the Leaves. Requiring Horseback Riding is an acknowledgement that - even in a magical mileu - what industry that does exist relies heavilly on animal power.

Frond of Iron & Armor

This would be a more complex Frond than Copper & Pottery to allow construction of Macemen, Pikemen, Longbowmen, Rangers, and the other tier-3non-magical units. There would definitely be a need for secondary pre-reqs as well.

Frond of Mithril & Manufacturing

Ditto for the last Frond, but even more so. Pre-reqs would exist often and usually as OR requirements. Here the concept outlined in Example #3 would find ample use to, for example, delay Elven civs access to heavy infantry units, Dwarven civs from archery units, Lanun civs from cavalry units, and so forth.

Beaker Costs would once again be low for each specific sub-tech. Once a civ is 'over the hump' so to speak, it should be able to rapidly adapt to changing needs on the battlefiled. Each successive Frond offers more options, more adaptions, more choices, more decision-making complexity.

Neccessity is the mother of invention. When you are the proud Bannor, when you know how to and do mine iron, when you have thousands of trained soldiers under your command, each outfitted with steel breastplate and mace, and when you've liven next door to Hippus and have borne their raids for centuries ... well then it should not take you three generations to learn how to pointy-up some sticks and create Pikemen.

Vanilla Civ models changes over 6 millenia. There is a lot of difference between knights and cavalry and tanks. Big, full-sized tech tree steps are appropriate to climb on that time scale. But FfH is meant to model a tiny slice of vanilla's 6 millenia. Compared to Knight vs Tank, there is hardly any difference between Maceman and Pikeman. Moving to a Frond system for military units will allow one civ to produce both these similar units for only a small investment in beakers.

Moving FfH military techs to a Frond system will allow a civ under pressure to rapidly adapt to the threat. There will be a burst of rapid military advancements, such as occured historically during WW2. At the same time a technological leader will be able to enjoy a brief hayday on the battlefield. They just will not be able to count on that advantage lasting for long.

 

[Unser Giftzwerg]

Example #One-Omega - Religious Fronds

I negelcted to mention that a Frond could be created for exploit by civs in the No State Religion condition. Naturally, this would include aspects needed/desireable for the Grigori. But other civs might find it valuable too. I have played a couple games where I did not declare a State Religion until quite late. A No State Religion frond might very well add new possibilities to civs beside just the Grigori.

Anyway, I just wanted to correct that omission.

Example #3 - Same Tech / Different Costs

The simplest way to accomplish this is to let two different techs grant access to the same benefit.

Let's say we want Khazad to be able to build their version of Macemen (Hammerfists) earlier than other civs. This is plausible because Khazad are often able to construct the Mines of Gal-dur wonder. Macemen would be allowable with the dicovery of Iron Working, just as they are now. But Hammerfists (specifically, not the generic Maceman) could also be allowed upon discovery of Arete. Thus if Khazad goes Runes, one of their mini-bonuses would be faster Hammerfists.

I do not know if Civ coding would allow for this. Thankfully, the above method is not the prefered method. The prefered method works this way.

Sticking to our existing examples ... Let's say we want our mechanically ept Khazad to invent Engineering sooner than your mechanically inept Elf.

Currently you must know Mathematics AND Construction to learn Engineering. This could be changed to Mathematics AND ( Construction OR Arete ). Or if you don't like Arete, use Call of the Earthmother. (i.e. a Frond tech, not the current Way ot Earthmother.)

The result of this is, certain techs can be 'fast tracked' if the appropriate exclusive frond is being exploited.

Or let's say we want to buld in more distinction between Order and Veil. For this example we have to pretend there are religions Fronds for each faith. For the sake of example imagine each of them is structured exactly like the example in #1 above.

The change we want to make for Order and Veil is to let Order build Inquisitors real soon but they have to work extra hard to develop High Priests. The exact opposite is to apply to Veil, quick High Priests but late Inquisitors. Early and late being relative terms comparedd to Leaves, Runes, and OLords, of course.

High Priests and Inquisitors are now both given a secondary tech requirement: Religious Law AND ( Arete OR Hidden Paths OR Mind Stapling OR _______ OR _______)

Specificically,

Inquisitors: Religious Law AND ( Arete OR Hidden Paths OR Mind Stapling OR Orders from Heaven OR Dance of the Veil* )

High Priests: Religious Law AND ( Arete OR Hidden Paths OR Mind Stapling OR The Just Following* OR Infernal Pact )

We thus have five religions that all produce the same top-quality units. But they do't all get those units at the same time.


(* 'The Just Following' and 'Dance of the Veil' are placeholder names for theequivalent of 'Gem of the Earthmother' as outlined in example #1. Tat is to say, the 'Religious Frond tech' requiring the greatest investment.)

=========================

That's the upshot of 'Fronding' parts of the TT. It's in very nebulous, genral form. I'm sorry about that, but I gotta do my best to type fast between puppy walks. Speaking of which....

 

[Chandrasekhar]

I don't like to pass any sort of judgment on ideas before they're completely posted, but I have class in 10 minutes, so let me just throw in my preemptive two cents. The system you describe certainly looks interesting, but I'm worried that it will cut down on uniqueness. If I can add my own little bit to it, I'd say that, in your copper and pottery example, you might instead have a base, more expensive tech ("Warfare"), and this tech would be a necessary requirement for all tier-2 unit techs. Each of these tier-2 unit techs would then have their own requirement aside from that (mining for axemen, hunting for archers and hunters, etc.). Off to class I go.