Jump to content
  • Advertisement
Sign in to follow this  
Portugal Stew

A Quick Thought on Research, or, Why Tech Trees Make No Sense

This topic is 2757 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

From Supreme Commander to Civilization, there's always been something a wee bit off about tech trees. In the case of RTSs, it's generally a matter of presentation; it doesn't make a lick of sense for a small battalion of soldiers to have to "research" new battlefield technology. Build the infrastructure to be able to utilize said technology, sure. Prove the battle crucial enough to afford advanced weaponry, sure. But research? The idea of "researching" the technology for cloaked psychic warriors in a minute and a half is preposterous, even if you argue there's a timelapse where every minute represents a day or something.
Then, in the case of Civilization, it's not necessarily that the tech tree is poorly described (after all, your society is researching, in a literal sense, because the technology simple doesn't exist yet or has not been revealed), it's just bizarre that a paleolithic tribe has the foresight to understand that someday, in the distant future, deciding whether to make pottery or spears will better prepare them for the nuclear arms race.

So my question is, how would you fix the research/tech tree system? Personally, in the case of a long-term Civ-type game, I had been thinking about if you had a system that procedurally develops technologies with progressively better, albeit sometimes unpredictable stats, where the outcome is statistically proportionate to what you put in but without guaranteed results. That is, you can invest money into developing a technology the improve upon or counter an existing weapon, but you have no way of being certain what you'll ultimately come out with or if it will ultimately be viable. You see a lot of this in real life warfare, such as during the War Between the States with the Hunley submarine and during WWII with all sorts of crazy giant tanks Hitler tried to build. I'm sure some players would hate the luck of the draw aspect, but I think random results would spice up the game.

So, any other thoughts?

Share this post


Link to post
Share on other sites
Advertisement
Comment:

I think in a well-balanced game, each faction should only be able to achieve a fraction of the possible researches. So at any moment, the player still needs to decide what researches are most beneficial for the situation.

In the game you describe, the strategic landscape is too monotonic. If a player is planning what to do in the nuclear age while the game is still in stone age, the game isn't giving the player enough concern that the game could end well before the nuclear age.

In this perspective, the problem is not about the tech tree, but the threat being unrealistic: The players are playing too defensively in the early ages.

In terms of a multiplayer game, one model that may fix this is to present the ages as a series of matches in a tournament. In such a system, when a player starts a game in stone age, he plays against a player in stone age on a small map. Because the two villages are so close, one of them must be eliminated (perhaps as the victory condition). The winner then continues in a different match with his original village, and plays against the winner of another match, using the same territory tiles they used in the previous matches as their starting points, inside a bigger tile.

Topology:

Match 1: Both players on a 1x1 tile. Winner takes the tile.
Match 2: Both players start on a 1x1 tile on the opposite corners of a 3x3 map. Winner takes the 3x3 tile.
Match 3: Both players start on a 3x3 tile on opposite corners of a 9x9 map, winner takes the 9x9 map.
Match 4: and so on...

Share this post


Link to post
Share on other sites
Could you make it so that technology was not the most dominant force in determining victory? I can see technology being varied as you imagine only so long as you could know that you didn't lose a long term game just because of a bad "hand." But if you have to invest huge amounts of time but can't know your status versus others I don't see that working well.

I can see a more short term, casual strategy game benefiting from lots of randomization on the other hand. There's less to invest in and the bad feeling you get from a crappy hand can be ameliorated by quickly being able to play again.

Share this post


Link to post
Share on other sites
Quote:
Original post by Wai
I think in a well-balanced game, each faction should only be able to achieve a fraction of the possible researches. So at any moment, the player still needs to decide what researches are most beneficial for the situation.

That sounds like an arguable design philosophy, but probably a generally agreeable one. I do hate getting to that point of a game in, say Supreme Commander 2 when everything's upgraded to maximum capacity and that entire aspect of gameplay is lost. But then again, SC2 was an absolutely dismal game overall. So good point.
Quote:
Ibid.
In the game you describe, the strategic landscape is too monotonic. If a player is planning what to do in the nuclear age while the game is still in stone age, the game isn't giving the player enough concern that the game could end well before the nuclear age.

In this perspective, the problem is not about the tech tree, but the threat being unrealistic: The players are playing too defensively in the early ages.

In that particular example I was commenting on the Civilization series' model, which technically works, and works well, from a gameplay perspective, but is transparent to the point of omniscience. To be fair, my experience with Civ IV has never involved rushing to nuclear arms, except when I was getting started and wanted to watch some cities explode.
Quote:
Ibid.
In terms of a multiplayer game, one model that may fix this is to present the ages as a series of matches in a tournament. In such a system, when a player starts a game in stone age, he plays against a player in stone age on a small map. Because the two villages are so close, one of them must be eliminated (perhaps as the victory condition). The winner then continues in a different match with his original village, and plays against the winner of another match, using the same territory tiles they used in the previous matches as their starting points, inside a bigger tile.

blah blah blah blah

I wasn't thinking of a specifically historically-oriented game divided into ages, but that's a neat concept, except that it allows for complete player annihilation at every stage, which seems like pretty high stakes. At least in Civ you could almost always scrape by, even if you lose land and influence.

Quote:
Original post by Wavinator
Could you make it so that technology was not the most dominant force in determining victory? I can see technology being varied as you imagine only so long as you could know that you didn't lose a long term game just because of a bad "hand." But if you have to invest huge amounts of time but can't know your status versus others I don't see that working well.

I can see a more short term, casual strategy game benefiting from lots of randomization on the other hand. There's less to invest in and the bad feeling you get from a crappy hand can be ameliorated by quickly being able to play again.

That is more or less what I had in mind (when I spoke of "long-term" I was thinking more in terms of in-game chronological scale than actual play time; I guess I was vague on that point), but now that you bring it up, there's plenty of room for balancing things out.

For example, keep in mind the fact that once a technology's been discovered, it can be duplicated. Trenches, for example, quickly became the dominant strategy for every country in both world wars and several others, even though they were only "discovered" by a few commanders (granted, digging holes is about as old a technology as fire, but in the case of industrial/modern age warfare it's value was something else entirely). But for better examples, tanks, airplanes, and atomic bombs were all swiftly adopted by everyone withing years or months of their introductions.

So there could be perks to learning a given technology first, perhaps simply by design for it being there first, but your application of said technology to a competent foe could make developing similar, if not identical, technology incredibly inexpensive. And vice versa.

Or something else. There's a billion and a half ways you can balance bad luck with good play. Short skirmishes are the most practical and most effective, but they are by no means the absolute unique solution.

Share this post


Link to post
Share on other sites
I have thought about some different ways to handle technology in games.

The first way is to have technology as an ephemeral object. That is one that can be lost.

With this, technology, once discovered, could be lost by that player. This requiers it to be stored somewhere that other player could attack and destroy (destroying the tech along with it).

This would also allow tech to be traded, copied or stolen as an object. Thus a spie might be sent in to steal a copy of a tech, then destroy the original.

Also, if the player was limited in how much technology they could store, then they would have to make a choice when they develop new tech as to what old tech they can afford to loose (and willingly destroy the unwanted tech).

Strategies would develop in protecting their technologies and destroying and stealing their opponents, making technology a much more dynamic gameplay element. And what could be more fun that litterally blowing your oopponent back to the stone age... :D


Another idea was that tech research is uncertain (and this could be combined with the first idea). Although there might be tech trees, just researching a branch does not mean that you will ever develop a particular tech, or that you will develop it at a specific level of research/time.

As you research a particular branch, you get a greater chance of discovering a technology, but the amount needed and what that tech is is not certain. However, some advancements will increase the chances of discovering certain technologies.


The last idea is more about how technology effects a game. In most cases, when a tech advance is researched, it gives makes the player more powerful in a general sense. That is the new units/weapons/etc are more powerful than all other previous technologies.

A good example is in RTS games where new units are always better than existing units.

In terms of strategies and tactics, the new units dominate the older units.

My idea is to turn this upside down a bit. What if the new technologies, instead of making units that are better across the board, are more of a specialist. That is they are better against certain enemy types, but are less effective than the old units against other types.

For example:

Lets say you have 3 units in an RTS: Conscript, Rider, Slinger

Conscripts are good against Riders but not so good against slinger. However they can attack and do significant damage against slingers. The same applies to the other unit types (but I am just going to foccus on one unit type for this example).

However, the player can research Pikes, and then produce (or turn the conscript into) Pikemen.

Now, Pikemen are excelent against Riders 9and other units based on the riders such as cavalry), however, the bukly pikes make them less effective against slingers as pikes won't allow them to move as quickly as the slingers.

This means that if you are uncertain what your enemy is going to throw at you, then a more general unit will be better, such as the conscripts. Howeverm if you know what your enemy is going to attack with, then the higher tech unit (the pikemen vs riders) will be a far better option.

Technology advances from the generalists to the specialists. But, because the specialists are good against any unit, the generalists are still useful as the core part of your army.

Share this post


Link to post
Share on other sites
Quote:
Original post by Edtharan
I have thought about some different ways to handle technology in games.

The first way is to have technology as an ephemeral object. That is one that can be lost.

With this, technology, once discovered, could be lost by that player. This requiers it to be stored somewhere that other player could attack and destroy (destroying the tech along with it).

This would also allow tech to be traded, copied or stolen as an object. Thus a spie might be sent in to steal a copy of a tech, then destroy the original.

Also, if the player was limited in how much technology they could store, then they would have to make a choice when they develop new tech as to what old tech they can afford to loose (and willingly destroy the unwanted tech).

Strategies would develop in protecting their technologies and destroying and stealing their opponents, making technology a much more dynamic gameplay element. And what could be more fun that litterally blowing your oopponent back to the stone age... :D


I believe you would love my idea about ship-design-driven-research over here: http://www.gamedev.net/community/forums/topic.asp?topic_id=589945 it could do all of those. (How do you make a link at these forums, and are there some guidelines somewhere which I missed?)


Quote:
Another idea was that tech research is uncertain (and this could be combined with the first idea). Although there might be tech trees, just researching a branch does not mean that you will ever develop a particular tech, or that you will develop it at a specific level of research/time.

As you research a particular branch, you get a greater chance of discovering a technology, but the amount needed and what that tech is is not certain. However, some advancements will increase the chances of discovering certain technologies.


The risk here is that if the effort for player trying to get something is high, but he doesn't, it is taken in very negative way, especially if the player is still exploring the game. In SotS this is somewhat implemented, but it is sometimes very frustrating when you have built a fleet based on the assumption that you will get certain tech maxed and you don't. It is quite challenging to start maxing some other tech late on the game if your opponents manage to do it first. The tech transfer is not too expensive in the gameplay, but it is mentally very frustrating.

Dead-ends should not make temporary dead-ends in your gameplay, but they perhaps should force you to go somewhere you didn't expect to go at.

Quote:
The last idea is more about how technology effects a game. In most cases, when a tech advance is researched, it gives makes the player more powerful in a general sense. That is the new units/weapons/etc are more powerful than all other previous technologies.

A good example is in RTS games where new units are always better than existing units.

In terms of strategies and tactics, the new units dominate the older units.

My idea is to turn this upside down a bit. What if the new technologies, instead of making units that are better across the board, are more of a specialist. That is they are better against certain enemy types, but are less effective than the old units against other types.

For example:

Lets say you have 3 units in an RTS: Conscript, Rider, Slinger

Conscripts are good against Riders but not so good against slinger. However they can attack and do significant damage against slingers. The same applies to the other unit types (but I am just going to foccus on one unit type for this example).

However, the player can research Pikes, and then produce (or turn the conscript into) Pikemen.

Now, Pikemen are excelent against Riders 9and other units based on the riders such as cavalry), however, the bukly pikes make them less effective against slingers as pikes won't allow them to move as quickly as the slingers.

This means that if you are uncertain what your enemy is going to throw at you, then a more general unit will be better, such as the conscripts. Howeverm if you know what your enemy is going to attack with, then the higher tech unit (the pikemen vs riders) will be a far better option.

Technology advances from the generalists to the specialists. But, because the specialists are good against any unit, the generalists are still useful as the core part of your army.


My idea solves this (for 4x space strategy games) by making the technology prototypes vulnerable against basic-level anti-tech weapons (idea being that new tech is vulnerable because of unstability, but only building and using tech-prototypes in combat will advance tech quick). In general, new tech prototype is expensive and vulnerable, but are also the only way to improve your tech quickly.


EDIT: Oh, you actually already commented it, though check my new post.

Share this post


Link to post
Share on other sites
How about instead of researching techs individually you can choose an area to research in. That way results are semi-random but you can control if you want Military research or trade research etc.

As a failsafe you could make it that if, after 10 turns or something, you still haven't unlocked anything then something is unlocked automatically.

Share this post


Link to post
Share on other sites
While I agree with the outline of the idea in the OP, I think that it's a problem that is largely inescapable. There can be a lot of different approaches to reducing the prescience of a commander or leader in a game (hidden future tech on trees, for example), but once a player has played through a game a time or two he or she will already have knowledge of what technologies lead to others regardless of whether or not an explicit tree is presented in-game for their review.

If you ever played Pax Imperia 2, it had a system in which you researched different areas simultaneously and controlled the rate by adjusting funding levels among them. You would have several options of what to research in a given area at a time, and some technology was labelled as "pure research", which would unlock new options rather than producing a new, usable technology. It prevents game-spanning forsight from the start the first few matches you play through, but after that tech hiding isn't very useful. Master of Orion II had a similar system, with a similar problem.

Randomizing stats of research presents its own problems, as noted above. The narrower the range of variation you allow in tech stats, the less impact your idea will have on research decisions as it will still be fairly predictable. A wider range will reduce more of the game to luck, which can be frustrating for players.

What I might do to add a little variety to research would be to research categories rather than specific techs ("firearms" rather than "M-4 carbine rifle"), with a behind-the-scenes bonus applied to specific research paths. For example, you may want a better rifle, and so you research "firearms", but your research produces a machine gun instead. You can keep researching firearms and hope to develop a new rifle, but the machine gun will tweak research probability towards the next step in the machine gun path, rather than developing an entirely different tech. That way there's still an unpredictable element, but the specific results you get just give you a specialized focus that you can use in your strategy rather than risk getting a dud technology.

You could also include scientists as a resource, with some being better at researching one thing than another. That way you can still work towards more specific technologies, but only indirectly by adjusting who works for you and how much funding they get, etc.

Share this post


Link to post
Share on other sites
Technology could be developed on need.
If you use archers a lot your people would develop longbowmen, and then crossbows.

That way the player controls what is going to get but not directly, like "research longbowmen"

The player controls research founding, wich determines the speed of advancements.

Some techs could be dependant on infrastructure.
For example, to build tanks you need an assembly plant, for the assembly plant you need a factory, for that you need a foundry, for that you need a workshop, for that you need a blacksmith.
So if you destroy all production buildings of a civ, you sent'em back to stone age.

In SC happens the same, if you are in tier 3 and your enemy blows all your high tech buildng, you fall back to tier 1. Now change "tier" for "age".

Share this post


Link to post
Share on other sites
There is something to be said for the difference between infrastructure and technology, though. For example, although much of Africa lacks the industrial infrastructure to produce semiautomatic weapons, the existence of the technology means it can be acquired, which still makes militants even competitive against much better-funded, more technologically advanced nations (remember Black Hawk Down?)

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!