A better tech tree?
Author
Hey ya'll,
I've been lurking around here for awhile, and made a few posts as an anonomous user (too lazy to sign in) but this is my first real post. Exciting. Whatever.
I've been toying around with a game idea for a while. Think of it as being similar to a typical space based empire building game.
Technology and research. I want to employ something a little different than a typical tech tree, and give the player more of a sense that they are contributing somehow to the development and research that their empire is producing. Rather than just sort of go through a pre-existing tree uncovering what's already there, I want to create a sense that they are almost coming up with ideas on their own, and then implementing them.
So far, what I've come up with is having a list of known "parts" that can be improved such as a ships hull, armor, weapons, scanners, and so on and then a list of known "generic research" such as lasers, biotechnology, radiation, nanotechnology, etc.
The player can pick something from the reseach list and combine it with something from the "parts" list, and then the scientists will get back to the player with what kinds of technology could be developed. The player could then pick one of those things, research it, develop it, and build it. Or try something else if it doesn't suit their fancy.
There are other options and conditions here, but this is the basic idea. So if you combine biotechnology with armor, your scientists might develop a kind of self-healing armor, or they might come up with a kind of living exo-skeleton for your marines, or whatever.
My question is, do you think that this gives the players more flexability and customization for their empire (they could focus heavily on electro-magnetic research and make most of their technology reflect that, if they want, or focus more on optical technology, making holograms and cloak technology...focus on weapons research or on propulsion technology... whatever...)
or!
Would this get annoying as players may end up trying many different combinations until they find something good. (Which, in my opinion, may be a good thing, since real R&D works similarly... try lots of ideas until one works.)
The game isn't spanning hundreds of years or anything, so there doesn't have to be huge leaps and bounds of technology. The game should only last long enough for players to make a few dozen unique enhancements to their forces, and maybe two or three major scientific breakthroughs... building wormholes or or almost impervious shields.
Hope this makes sense. Let me know what you think.
Unless you come up with a very clever way to implement this, it still sounds mostly like a static tech-tree (or tech-something else, but still static). For example, "So if you combine biotechnology with armor, your scientists might develop a kind of self-healing armor". This self-healing armor is almost definitely going to have to be predefined as a combination of biotechnology and armor. You'll still have to program this into the game. In fact, this may make the whole tech-tree even more daunting, because now in addition to all the basic types of tech that you have, you have to specify what the combinations of each type of tech would produce.
But I acknowledge that your desire is one I share. If you could get a very basic but flexible scripting system together, you might be able to get technology to be defined by simple scripts and equations. Then, you may be able to get one technology to manipulate the variables of another tech when combined. And research on a specific tech might tweak the equations in the tech. For example, weapon tech might be defined by a script that takes distance as a parameter, and returns damange. A really sucky, early weapon's script might basically return a/x + b as damage (where x is distance and a and b are some constants). Researching more on this weapon type might increase a or b. It's still roughly the same type of weapon, but it gets better as you research it. And this could change a bit every time you replay the game. Those constants a and b for each level of the weapon could be randomly generated, and thus sometimes you might make major breakthoughs quickly, but other times, research would be slow. And if you research more in weapon tech, you might find a different type of weapon, like one that uses an equation that is just a constant, thus constant damage regardless of distance. Obviously, this is overall a better weapon. But it might have a lower damage than the previously described weapon at a close range. It's a different type of weapon, and thus has different properties, based on the script.
These are just very vague thoughts in my mind right now, though. Maybe it'll spark something in your mind, however.
But I acknowledge that your desire is one I share. If you could get a very basic but flexible scripting system together, you might be able to get technology to be defined by simple scripts and equations. Then, you may be able to get one technology to manipulate the variables of another tech when combined. And research on a specific tech might tweak the equations in the tech. For example, weapon tech might be defined by a script that takes distance as a parameter, and returns damange. A really sucky, early weapon's script might basically return a/x + b as damage (where x is distance and a and b are some constants). Researching more on this weapon type might increase a or b. It's still roughly the same type of weapon, but it gets better as you research it. And this could change a bit every time you replay the game. Those constants a and b for each level of the weapon could be randomly generated, and thus sometimes you might make major breakthoughs quickly, but other times, research would be slow. And if you research more in weapon tech, you might find a different type of weapon, like one that uses an equation that is just a constant, thus constant damage regardless of distance. Obviously, this is overall a better weapon. But it might have a lower damage than the previously described weapon at a close range. It's a different type of weapon, and thus has different properties, based on the script.
These are just very vague thoughts in my mind right now, though. Maybe it'll spark something in your mind, however.
Sounds like a good idea but..
Multiplayer
in start craft i plan for the fastest way to get to a certain technology and i do R&d path just to get to it, with out knowing what will happen when I devolve optical it really limits me and slows down the game play.
but in single player this would be really neat to have it would add a lot of realism to the game.
Multiplayer
in start craft i plan for the fastest way to get to a certain technology and i do R&d path just to get to it, with out knowing what will happen when I devolve optical it really limits me and slows down the game play.
but in single player this would be really neat to have it would add a lot of realism to the game.
I've been working on something similar for my space game. Basically, the only problem you have is that you must be able to define all abilities and equipment which new technology can provide in-game. It may be possible to create a tech engine which occassionally allows for odd abilities to crop up, but the difficulty in coding and designing something like this is daunting.
You're best approach is something like what Agony has proposed. As new tech is developed, it can modify a master equation on that ability.
For Agony's second weapon type example, it would be as easy as setting a=0. Then you'd arrive at the second, constant equation. So as long as you could define some master function for all weapons, like ax^3 + bx^2 + xln y + c + d/x = damage, then a tech tree for weapons could be fairly dynamic depending on the player's research goals, while still remaining simple from the game's standpoint. The only step left is defining some system for modifying all these variables based on R&D, and assigning these variables to various equipment, which all decide damage through this master equation.
A system like this could work for all the other tech tree areas too. For qualitative results, you'd need switches instead of variables. For instance, if you had a tech which allowed oyu to destroy one enemy craft, you'd need a simple boolean variable for the tech which describes if it can do this or not, instead of an integer or float variable telling the degree of its action.
Since I'm trying to put a system like this in my game, I obviously think it's a good idea. So you'll have at least one guy willing to play your game Taolung. [wink]
You're best approach is something like what Agony has proposed. As new tech is developed, it can modify a master equation on that ability.
For Agony's second weapon type example, it would be as easy as setting a=0. Then you'd arrive at the second, constant equation. So as long as you could define some master function for all weapons, like ax^3 + bx^2 + xln y + c + d/x = damage, then a tech tree for weapons could be fairly dynamic depending on the player's research goals, while still remaining simple from the game's standpoint. The only step left is defining some system for modifying all these variables based on R&D, and assigning these variables to various equipment, which all decide damage through this master equation.
A system like this could work for all the other tech tree areas too. For qualitative results, you'd need switches instead of variables. For instance, if you had a tech which allowed oyu to destroy one enemy craft, you'd need a simple boolean variable for the tech which describes if it can do this or not, instead of an integer or float variable telling the degree of its action.
Since I'm trying to put a system like this in my game, I obviously think it's a good idea. So you'll have at least one guy willing to play your game Taolung. [wink]
Author
Thanks for your responses, this is very helpful.
"Unless you come up with a very clever way to implement this, it still sounds mostly like a static tech-tree (or tech-something else, but still static)."
Yes, this is exactly part of the problem I'm running into. Much of the content is already generated, and in a very real sense, the tech tree still exists, but behind the scenes, I guess.
Maybe I'm just trying to find a good way to create the illusion that there is no tech tree, even though there is one.
I'm thinking about some games where you physically see the tech tree, but it's all greyed out until you get to it. I hate the idea of knowing that there's "something" there to research, you just need to get to it. I want to create more a feeling of exploration.
"in start craft i plan for the fastest way to get to a certain technology and i do R&d path just to get to it"
I totally recognize this play style, but what I'm shooting for would actually discourage this. This is much more of an empire simulation kind of game, where research could be bought or stolen from others, and others will be trying to steal yours.
It isn't a race to see who can get to weapon X first, but more of a challenge to see what kind of technology one player develops, and then how the other player develops technology to counter it, etc.
I really like the ideas here that will allow the game to sort of evolve and generate new tech on the fly, and I think that it can be encorporated well into the "tech tree" approach. For example, combining "Lasers" with "Weapons" would clearly result in more powerful laser weaponry. The more time and money spent on this combination, the more powerful they become... Compared to combining "lasers" with "scanners" which would would develop more accurate and longer range lidar. In this example, perhaps the knowledge and experienced gained from developing better laser weapons could be carried over to the lidar, as well. They both use lasers, so when your scientists start to tackle a better lidar, they already have some research to go from.
"Unless you come up with a very clever way to implement this, it still sounds mostly like a static tech-tree (or tech-something else, but still static)."
Yes, this is exactly part of the problem I'm running into. Much of the content is already generated, and in a very real sense, the tech tree still exists, but behind the scenes, I guess.
Maybe I'm just trying to find a good way to create the illusion that there is no tech tree, even though there is one.
I'm thinking about some games where you physically see the tech tree, but it's all greyed out until you get to it. I hate the idea of knowing that there's "something" there to research, you just need to get to it. I want to create more a feeling of exploration.
"in start craft i plan for the fastest way to get to a certain technology and i do R&d path just to get to it"
I totally recognize this play style, but what I'm shooting for would actually discourage this. This is much more of an empire simulation kind of game, where research could be bought or stolen from others, and others will be trying to steal yours.
It isn't a race to see who can get to weapon X first, but more of a challenge to see what kind of technology one player develops, and then how the other player develops technology to counter it, etc.
I really like the ideas here that will allow the game to sort of evolve and generate new tech on the fly, and I think that it can be encorporated well into the "tech tree" approach. For example, combining "Lasers" with "Weapons" would clearly result in more powerful laser weaponry. The more time and money spent on this combination, the more powerful they become... Compared to combining "lasers" with "scanners" which would would develop more accurate and longer range lidar. In this example, perhaps the knowledge and experienced gained from developing better laser weapons could be carried over to the lidar, as well. They both use lasers, so when your scientists start to tackle a better lidar, they already have some research to go from.
How much time do you want to put into this idea?
If you want to make this a central feature you might come up with a very large number of small, incrementable stats that directly impact gameplay in a wide number of areas. It would not be trivial, but the more granular the changes you can make in theory the more combinations you can come up with.
Having said that, though, if you're talking about doing this for an empire game I'm a little concerned about taking control away from the player in this manner. 4X games thrive on long range strategic planning. What is the player to do when they find that they're behind the tech curve, maybe being bombarded by weaponry that they have no defense against? In a planned tech tree you can at least strategize ahead, thinking things like "okay, they have orbital barrage; it's going to take me three turns to finish researching city shields... that means I'm going to lose 4 colonies, but then I'll be okay."
I know an awful lot of players chose the Creative attribute in Master of Orion 2 so that they could get control of their own tech tree choices, btw. It sounds a little like the same effect.
If you don't want the feel that the tech tree is determinant, could you simply use a tree that had modifiers to each leaf as well as a bunch of semi-random situations allowing you to jump branches or levels?
If you want to make this a central feature you might come up with a very large number of small, incrementable stats that directly impact gameplay in a wide number of areas. It would not be trivial, but the more granular the changes you can make in theory the more combinations you can come up with.
Having said that, though, if you're talking about doing this for an empire game I'm a little concerned about taking control away from the player in this manner. 4X games thrive on long range strategic planning. What is the player to do when they find that they're behind the tech curve, maybe being bombarded by weaponry that they have no defense against? In a planned tech tree you can at least strategize ahead, thinking things like "okay, they have orbital barrage; it's going to take me three turns to finish researching city shields... that means I'm going to lose 4 colonies, but then I'll be okay."
I know an awful lot of players chose the Creative attribute in Master of Orion 2 so that they could get control of their own tech tree choices, btw. It sounds a little like the same effect.
If you don't want the feel that the tech tree is determinant, could you simply use a tree that had modifiers to each leaf as well as a bunch of semi-random situations allowing you to jump branches or levels?
My tech tree for my current project is at least a bit different. There is no research per se. Almost everything that units can do is a 'project'. Under the hood, projects are simply a series of function pointers. In gameplay terms, projects are something to spend resources/skills/time/effort to get something in return.
For example, to make a building, you might have the prerequisite that your empire controls the land. Then the project requires a bunch of wood, a bunch of manual labour and a little craftsmanship. When those things are provided, you get a building.
Research is handled the same way. For example, to learn ironworking, you might have the prerequisites of having bronzeworking and some iron. The project requires a bit of iron, a bit of craftsmanship and a bunch of thought. When those are provided, you get the requisite knowledge to make iron weapons [or perhaps only prototypes! ]
Since the setup is so generic, things can be mixed together rather easily.
Some technology cannot be researched though. Some things can only be gained through use. [Would anyone create a longbow without using a shortbow frequently? or similarly, the above prototype comment can be used through this mechanism... To get good guns, you need to use shoddy prototype flintlocks for a while] And some things can only be found. [Who would make boats if they knew no oceans?]
I think mixing up technological advancement like this will be much more interesting than "spend $x research points, get $foo".
For example, to make a building, you might have the prerequisite that your empire controls the land. Then the project requires a bunch of wood, a bunch of manual labour and a little craftsmanship. When those things are provided, you get a building.
Research is handled the same way. For example, to learn ironworking, you might have the prerequisites of having bronzeworking and some iron. The project requires a bit of iron, a bit of craftsmanship and a bunch of thought. When those are provided, you get the requisite knowledge to make iron weapons [or perhaps only prototypes! ]
Since the setup is so generic, things can be mixed together rather easily.
Some technology cannot be researched though. Some things can only be gained through use. [Would anyone create a longbow without using a shortbow frequently? or similarly, the above prototype comment can be used through this mechanism... To get good guns, you need to use shoddy prototype flintlocks for a while] And some things can only be found. [Who would make boats if they knew no oceans?]
I think mixing up technological advancement like this will be much more interesting than "spend $x research points, get $foo".
What it sounds like to me is that either, you're going to end up with an ordinary tech-tree anyway. (You combine biotechnology with armor, and get self-healing armor. Doesn't sound much more flexible than the standard version)
Or alternatively, you're going to end up with people researching a bunch of "grey goo". If you research armor 8 and biotechnology 4, you get self-healing armor 1.
I'm not sure if this makes sense, but I think it's important to make some distinct items that the player can latch on to. If you see a suit of exo-skeleton armor, with unique graphics, a proper description and a name, that is much more satisfying to research than simply "armor 14", which is what you got after researching armor 13 (which, in another game, might be called Powered Armor", and reactors 7 (which could have been called "Portable Fusion Reactor".
Players need to be able to see the difference between having a tech and not having it. They need to know which branch promises the most interesting inventions.
Anyway, you should definitely give the player some kind of pointers on what he should expect to get if he researches something. The scientists should have a pretty clear idea of what they're attempting to do, so they can give you some brief report on what potential improvements you could get from each.
Then again, maybe I've misunderstood part of your idea. In either case, something better than the standard tech tree is a great idea. :D
Or alternatively, you're going to end up with people researching a bunch of "grey goo". If you research armor 8 and biotechnology 4, you get self-healing armor 1.
I'm not sure if this makes sense, but I think it's important to make some distinct items that the player can latch on to. If you see a suit of exo-skeleton armor, with unique graphics, a proper description and a name, that is much more satisfying to research than simply "armor 14", which is what you got after researching armor 13 (which, in another game, might be called Powered Armor", and reactors 7 (which could have been called "Portable Fusion Reactor".
Players need to be able to see the difference between having a tech and not having it. They need to know which branch promises the most interesting inventions.
Anyway, you should definitely give the player some kind of pointers on what he should expect to get if he researches something. The scientists should have a pretty clear idea of what they're attempting to do, so they can give you some brief report on what potential improvements you could get from each.
Then again, maybe I've misunderstood part of your idea. In either case, something better than the standard tech tree is a great idea. :D
Author
What I'm trying to avoid is a tech tree that just moves from "Laser turret 5" to "laser turret 6" and so on. I definately want to add more flavor and distinction to research, even if it is to an extent an illusion. What I mean, is that at some level, players are going to develop a better laser turret. It's slightly more powerful, has a longer range. Logically, you could just add a number to it to denote that it's better than a previous version, but I feel like this gets boring. Even if it just gets a spiffy name and brief description just to set it apart, rather than a number, I'll be happier.
I'm rethinking my approach, though.
My intention was to create a system that would allow a player to decide what they wanted, and find a way to create it. For example, a player may decide that he wants to improve his ships speed, deciding that this will greatly improve his tactical advantages.
He selects "engine" as the item or "thing" he wants improved. He could then match it with "engine" again to indicate that he wants to conduct general research into engines. This will most likely simply improve his existing engines. They'll get faster, use less fuel, whatever. His scientists will report back with what kinds of improvements they're making, or other enhancements that they see as being possible. Through a random event, they might stumble upon something really unique and revolutionary.
Conversely, the player could combine "engine" with "graviton research." His scientists come back with a design that would form a small gravity well just in front of the ship, allowing it to "fall" faster and faster in any desired direction. They give him figures as to how much energy something like this would require, how much time and money it'll take to develop, how big the engine will be, and so on.
Maybe that isn't what the player is looking for, so he combines "engine" with "string research". His scientists come back with a design that uses interdimensional strings to pull the ship along.
What's the difference? Despite differences in power requirements, time, size, and other statistics, there should be a strategic difference. For example, perhaps the other player has already developed a graviton scanner, which detects changes in gravity wells to locate, track, and identify ships. Creating a graviton engine would be like sending up a flare, so the player elects to develop the string engines, which will move much more silently.
What I want to try to do is enable a system where the player gets to identify a specific need, create the technology that meets that need without having to develop other, unimportant technologies first (the ability to skip around the tech tree, to an extent, as desired), and that allows opportunities for various strategic advantages and struggles. I want to see a kind of chess match where players try to come up with a technologies that will outsmart, not just outgun the enemy. One clever technology, used just the right way, could trump the massive armies that the other player has developed.
I'm rethinking my approach, though.
My intention was to create a system that would allow a player to decide what they wanted, and find a way to create it. For example, a player may decide that he wants to improve his ships speed, deciding that this will greatly improve his tactical advantages.
He selects "engine" as the item or "thing" he wants improved. He could then match it with "engine" again to indicate that he wants to conduct general research into engines. This will most likely simply improve his existing engines. They'll get faster, use less fuel, whatever. His scientists will report back with what kinds of improvements they're making, or other enhancements that they see as being possible. Through a random event, they might stumble upon something really unique and revolutionary.
Conversely, the player could combine "engine" with "graviton research." His scientists come back with a design that would form a small gravity well just in front of the ship, allowing it to "fall" faster and faster in any desired direction. They give him figures as to how much energy something like this would require, how much time and money it'll take to develop, how big the engine will be, and so on.
Maybe that isn't what the player is looking for, so he combines "engine" with "string research". His scientists come back with a design that uses interdimensional strings to pull the ship along.
What's the difference? Despite differences in power requirements, time, size, and other statistics, there should be a strategic difference. For example, perhaps the other player has already developed a graviton scanner, which detects changes in gravity wells to locate, track, and identify ships. Creating a graviton engine would be like sending up a flare, so the player elects to develop the string engines, which will move much more silently.
What I want to try to do is enable a system where the player gets to identify a specific need, create the technology that meets that need without having to develop other, unimportant technologies first (the ability to skip around the tech tree, to an extent, as desired), and that allows opportunities for various strategic advantages and struggles. I want to see a kind of chess match where players try to come up with a technologies that will outsmart, not just outgun the enemy. One clever technology, used just the right way, could trump the massive armies that the other player has developed.
Perhaps what you really want is to allow the players to reasearch improvements and counter measures. So if one player develops a metal eating microbe that they use to destroy enemy ships. Then the player on reciving end, after learning of this weapon can begin to research a counter measure for it. Perhaps their scientists come up with a couple of diffrent possible solutions and the player choose the one they prefer.
The other player now discovering that their enemy has a defense against their main weapon has a choice, they can start developing an alternative weapon or they can research an improvement to the microbe weapon. The player decides to continue with the microbe weapon and their scientists give them a couple of option: from a new strain which is fastest to reasearch but also easiest to counter, to transforming the mircobes into a swarm of nanites. The diffrent options all have their own trade offers in terms of time, resources and strategy. As well depending on the opponents technology effectiveness of an improvements/counter-measures wil change.
The other player now discovering that their enemy has a defense against their main weapon has a choice, they can start developing an alternative weapon or they can research an improvement to the microbe weapon. The player decides to continue with the microbe weapon and their scientists give them a couple of option: from a new strain which is fastest to reasearch but also easiest to counter, to transforming the mircobes into a swarm of nanites. The diffrent options all have their own trade offers in terms of time, resources and strategy. As well depending on the opponents technology effectiveness of an improvements/counter-measures wil change.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
