Tech proposal: A tech "tree" and a tech "pool"

This is a proposed modification to the current research management system. 

Problems with current system

Currently, players get to choose from 4 or 5 techs for which they have the prerequisites, and players can "reroll" the random techs for a small penalty to research speed. This system creates a few problems in my opinion:

  • Some technologies unlock critical capabilities which are central needs to almost every playthrough, e.g. building farms or invading core planets. With the random system, there is no way to know when these techs will be made available.
  • I'm assuming that one goal of this system is to avoid Analysis Paralysis (AP), where players have to peruse the entire tech tree to figure out which tech to research next. But it doesn't actually resolve that problem, because even with the limited selection of which tech to research next, we still need to learn the whole tech tree in order to know the techs we need to research whenever they appear in order to reach one that we want.

 

Proposal

My proposal is to have two systems: a core tech tree and a tech pool.

The core tech tree would function exactly as a normal tech tree (like GC3), but it would be limited to the core, critical technologies. This tree would only have ~20 techs in it. Very easy to peruse and understand quickly. These are the techs that unlock basic functionalities, like building farms as described above. (Since the tree would not be super deep, researching techs in this tree would need to be restricted or limited in some way. I have several ideas for this which I'll explain at the end.)

The tech pool would function exactly like the current GC4 system. The only difference is that it doesn't include the core technologies which are on the core tree. Techs from the pool could depend on core tree techs or other pool techs, so you can have as much depth as you want here. Endgame technologies would be great candidates for the pool.

Whenever the player chooses a research, he can select from either the tree or the random 4-5 techs selected from the pool.

 

Why it's better

This solution addresses the problems with the current system. Since the core techs are available in the small tree, they will be easy to find, and the player will always have the ability to see the clear research path needed to reach these techs. This will limit AP and also ensure that the player will be able to get access to these basic capabilities on their desired schedule.

 

How to limit tree research

One challenge of this solution is that, since the tree with the critical technologies is compact, it makes the path to reaching these techs shorter than they are now. Most of the critical techs are early in the dependency chain anyway, but some (e.g. planet invasion) do take a while to get to. So we need to "slow down" access to techs on the main tree. I can think of 3 good solutions to this problem:

  • GC3-style "age", where you can't research techs in the tree until X turns have passed. I don't love this solution, but hey, it worked in GC3.
  • Research times on the core tree could start out very long, i.e. 5x the normal research time. Then, these core techs could have a list of related "desired" techs from the pool which function like soft prereqs: the pool techs are not actually required to research any core techs, but when the "desirement" pool techs are researched, they decrease the time required to research the corresponding core techs. (This is my favorite solution, and I'm happy to explain in more details if folks are interested.)
  • Core techs could be locked until a given number of pool techs from the same family are researched. (e.g. you can't research planet invasion until you've researched 5 military pool techs.) 

 

Please let me know your thoughts!

5,537 views 8 replies
Reply #1 Top

My understanding is that a tech tree will be implemented in future versions but it will act only as a guide - you can look but not touch!

So there needs to be a solution - it is fundamental to the Galciv universe.

I quite like the Core Tech and Tech Pool idea. It's not perfect, but anything is an improvement on the current "blind luck" system. Under "How to limit tree research" I am completely against the first proposal, However the second is much more acceptable and opens up a wealth of possibilities and research strategies - exactly what we want. Extra decisions need to be made (a good thing) and balancing the benefit of the particular pool tech and its effect on the core tech research time will be a challenge. The third solution offered is more simplistic, however...

We have to remember some poor bugger has to program the AI. It's all very well proposing things that involve complex decisions but we all feel let down if the computer opponent does something really dumb because, well, the challenge disappears.

On the other hand, we are contributing to this site because we want to let the devs know what we would like to see, and what we don't. We should ask for the almost-impossible, and then rein it in a bit. Maybe.

So well done for the idea jdschw - I think it has something going for it. Overall, I felt the GalcivIII tech tree system worked OK - not the best, and far from the worst. In IV I'd rather see your system than the one we have been given.

Reply #2 Top

Yeah the current tech setup is by no means final. The big issue is making a "tech tree" that works with the more dynamic way you can get techs and aren't always exclusive or might require multiple prereqs etc.

This two lane system you're proposing is interesting but as always this also comes back to scope creep and if it's worth maintaining two different systems for what is just one aspect of the game.

I'll run it by the team though.

Reply #3 Top

Thanks @SchismNavigator!

I'm a software engineer myself, so very sensitive to the problem of scope creep. My hope is that this dual system I proposed might not be a heavy lift because it basically combines the GC3 and GC4 systems, which both already exist. Of course, integrating them in the way I describe is also a project in its own right.

I proposed this because I think it might be a good way to preserve the "random sampling" system that the dev team likes, while still giving players direct access to those core, critical capabilities that they need to unlock chunks of gameplay.

I'm happy to chat more about finer points of this design if the dev team thinks it would be useful.

Reply #4 Top

I'd like to second the request for tree support, and throw another possible solution into the mix.

I like how the randomness of the current tech setup increases replay ability by making different tech routes easier on different run-throughs. And the reroll ability seems good enough to get to a key technology -- if only I knew which way I need to go.

I'm in a fight in my current game and really need planetary invasions and more logistics. I'm pretty sure planetary invasions is under military. But if I need to research two more techs before I can get it, I have no way to find out which, of any of the ones I'm looking at, is the first one I need. For getting more logistics, I'm not even sure what category I need.

A searchable manual of techs would solve my problem, as long as each tech listed its prerequisites. But it would be kind of clunky.

Something like LifeMap's pan and zoom approach to large trees would be a natural fit for GalCiv, which also uses pan and zoom for the galaxy map:

   https://www.youtube.com/watch?v=vNHfxo77Kes

A visualization like that also has the advantage that it can show me my progress through the tech structure/

Multiple paths to a technology could be handled by adding links across nodes of the tree, as long as the links are mostly confined within branches, so they don't clutter up the whole map.

Long long ago, I worked on a related visualization:

   https://www.youtube.com/watch?v=pwpze3RF55o

It turned out that getting a decent layout was surprisingly easy, only a day or two of coding, and required essentially no run time. Minimizing the length of the cross links makes getting a good layout harder, but a greedy algorithm will probably work pretty well.

Reply #5 Top

My view is straightforward: when in doubt, simplify. In another thread I proposed replacing technology selection with randomized technology awards. I stand by that general idea. I don't know how many techs there are in the tech tree, but it is a lot. That means a lot of clicking, a lot of low-impact decision-making, and a lot of frustration.

That said, my proposal to focus on a single color if you have a Science Minister may be too limiting. Other commenters are right, some technology paths don't really stay on a single color. What if your Science Minister allowed you to select any technology in the tree to emphasize, but the techs are still awarded randomly? Maybe techs on your selected path would be twice as likely to be chosen as the other alternatives. Perhaps that would give a better combination of RNG-based replayability and moon-shotting.

Reply #6 Top

How about adding an Executive Order like the generic "advance research by x points" but one which let you actually pick a specific technology from the Tech Tree to start researching?  Fits into present scope - no new mechanic needed.

Reply #7 Top

I like the idea of having a minister weight certain techs more than others.  It allows for focus within the RNG.

Reply #8 Top

I may be in the minority, but I enjoy navigating the tech tree. Click for click, I find it more enjoyable that driving my ships around, for example.

But a combination of Jeff Yutzler and jdschw's ideas might be the best of both worlds. Suppose that instead of a subtree of core techs, there was a tree of tech directions. It could have entries like:

    ship > attack > beam

    ship > attack > planetary invasion

    ship > speed

    aliens > diplomacy

    aliens > tourism

If a player doesn't like any of their fully random choices, they could choose a research direction from that tree, with maybe a larger penalty the further down the tree they specify. The UI then offers them a few technologies to choose from that advance in that direction, first heading toward any essential technologies in that direction, and then to technologies that improve on them.

The direction tree could be a lot smaller than the full tech tree, since each of the categories above probably have 10 techs behind them. And it would probably work better for the problems I have encountered. If I have lots of antimatter, I can prioritize missile weapon development. If all the aliens hate me, I can prioritize diplomacy. I don't really need to know the entire set of technologies, just that I'm going to make progress in the areas I most need it.

The direction tree can also accommodate there being several paths to a given technology, or their being multiple requirements for one, or a single technology helping in different directions. The system just navigates through the technology structure to any technologies that help with the given direction.

That said, I'd still love some way to get an overview of the tech tree and what techs I've picked up so far. But maybe a summary of what fraction of the techs I've gotten in each direction would be enough.