But that is because of weapons being smuggled from ex-USSR and China.
Are you planning to include technology leaking and weapons/tools commerce in a strategy game?
A Quick Thought on Research, or, Why Tech Trees Make No Sense
Quote:Original post by klefebz
But that is because of weapons being smuggled from ex-USSR and China.
Are you planning to include technology leaking and weapons/tools commerce in a strategy game?
There is still a big difference between technology and infrastructure. If you take a blacksmith who knows how to produce steel in a reliable fashion, and you burned his shop to the ground then he could still quickly rebuild and go back to producing decent steel for tools/weapons.
Now compare the time it takes the smith to rebuild his shop to that of another smith to develop the technology independently. The first smith can be up and running again in a few weeks or months, (Or days if there are other people he can call on to provide the materials he needs) but other smiths may take generations to develop the same technology.
Also consider that there are gun smiths in very poor parts of Africa who can make the parts needed to build something like an AK. They've seen the weapon, have torn it apart, and rebuilt hundreds of the things. It might take them all year to get all the materials needed to build one from scratch, and usually would scrounge parts from other guns, but they skill have the ability. It is the infrastructure and industrial base that allowed Soviet States to produce hundreds of thousands of the weapons.
We also have examples of people producing armored vehicles in their home garages. Groups producing high powered rocket engines on shoe string budgets, and many other projects that go against the idea that you need a large industrial base devoted to producing something.
Quote:Original post by Portugal StewTrue!
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.
Quote:So my question is, how would you fix the research/tech tree system?Who says it's broken?
Quote:If you take a blacksmith who knows how to produce steel in a reliable fashion, and you burned his shop to the ground then he could still quickly rebuild and go back to producing decent steel for tools/weapons.
And what if he was inside his shop while i burned it? Bah, we are just rambling about this.
We should reduce this to the abstraction of a game.
Say player A had a lot of tanks and lost'em in battle. Player B bombed all the factories.
Since factory was needed to make tanks, or at least make them at a reasonable speed, player A can no longer count on those weapons.
Now lets say player B detonated a EMP weapon over A's territory, eliminating electrical infrastructure. This means A can no longer use electrical illumination (production down because of less workable hours), food refrigeration (famine and disease), telecommunications (efficiency down), and many other 20th and 21st century techs, wich means A is practically dragged 150 years back.
Technology is based on knowledge and knowledge is a property of society, not infrastructure, however, infrastructure is essential when you need to utilize technology in a larger scale. Though one thing about games is that it is ok to get aspiration from the real world, but a game that models the real world perfectly would be really boring. Games are about game mechanics that adds entertaining value to the game.
The problem is that tech trees are boring and VERY old way of portraying technological advancement. But what are the greatest innovations on technological advancement in the history of games? That is the question we should be asking.
The X-Com series was very funny since you had to capture enemy artifacts, bodies or ships in order to advance in tech.
MoO2 was nice since you had to use diplomatics for trading tech in order to complete the tech tree (unless you were like me and always chose the racial trait that enabled all techs for each tech level)
Natural Selection (Half-Life mod) was awesome FPS/RTS game where the tech advancement was very well done considering it was an FPS for all players but the human commander.
Space Empires were nice since you didn't have to research tech tree a one tech at a time but could specify how you spend on each area of technology.
Besides that I really can't recall too many positive tech innovations.
I think this trap of tech-tree is a dinosaur from the ages when software engineers felt that the safest way of implementing tech advancement in a game would be to model it in simple ontology, because that is the safest way to do it. Today the computers are able to model much more complex ideas, though are the engineers up to the task in game balancing sense?
The problem is that tech trees are boring and VERY old way of portraying technological advancement. But what are the greatest innovations on technological advancement in the history of games? That is the question we should be asking.
The X-Com series was very funny since you had to capture enemy artifacts, bodies or ships in order to advance in tech.
MoO2 was nice since you had to use diplomatics for trading tech in order to complete the tech tree (unless you were like me and always chose the racial trait that enabled all techs for each tech level)
Natural Selection (Half-Life mod) was awesome FPS/RTS game where the tech advancement was very well done considering it was an FPS for all players but the human commander.
Space Empires were nice since you didn't have to research tech tree a one tech at a time but could specify how you spend on each area of technology.
Besides that I really can't recall too many positive tech innovations.
I think this trap of tech-tree is a dinosaur from the ages when software engineers felt that the safest way of implementing tech advancement in a game would be to model it in simple ontology, because that is the safest way to do it. Today the computers are able to model much more complex ideas, though are the engineers up to the task in game balancing sense?
Quote:Original post by SneftelQuote:Original post by Portugal StewTrue!
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.Quote:So my question is, how would you fix the research/tech tree system?Who says it's broken?
The tech tree works much better for RTS games, because it makes the game more interesting and the game has much focus outside the technological advancement. It is just a one strategy to go for high tech, with the risk of getting rushed by earlier tech. Makes perfect gaming sense.
For more heavy strategy games the tech tree is more central part of the game, but it is BORING. While everything else in the game might be different every time you play it, the tech tree is the same, and if you want to try different strategies you always have to get to high enough tech level to try it out. Even though games offer the chance of starting from an age where certain level of tech is already acquired, we gamers tend to want to see the affects of the strategy from the start.
I've also noticed that starting the games over is much more fun than finishing them and I believe the tech tree gets part of the blame. In many games it advances more quickly early on and is more significant, later in the game it is just about finishing the opponents off with maxed out tech.
Tech tree is broken in many ways. Parallel research makes the tech tree more interesting. Tech advancement that requires more than just the previous tech as prequisites (like in X-Com) makes it more interesting. Taking out the complete control of the tech advancement from single player makes things more interesting. Certainly there are more ways and I believe pointing out those ones would very benefical, especially from the perspective of already published games, since you can always make abstract ideas, but when someone has published a game with one, it is more concrete and at least implementable.
Another idea I recently had for tech advancments (I posted it off the top of my head in another thread, but have since though more on it) is that of a more maleable research system.
In curent tech trees, there are nodes that give you the tech advancment. They are hard coded into the game, end even if you only have access to certain ones of them in some playthroughs and not other, they are still hard points.
My idea was to have a system where the player describes the goal of the research and the game calculates how much effort (time and resources) the player has to put into it to reach it.
Some advancments would give a speed boost to the research for certain technologies (for example "Theory of Relitivity" might increase the speed of researching "GPS navigation systems"). This way you could still have prerequsites, but they wouold not be a hard prerequsite.
As an example of this sytem, you might want a laser weapons that does 100 points of damage and fires 10 shots per second.
So to get that, the game would work out that you need to research "laser weapons" to get the base technology, then research it long enough in "damage output" to reach 100 damage and also "fireing speed" to 10 shots per second. It would then give the player an indication of the time and costs involved in doing this.
IF this was too long a time, they might choose toinstead first research a laser weapon that only did 10 damage each hit and fired 5 shots per second and later on research further to get their 100 damage and 10 shots per second.
If you made it so that there was a geometrical increase as time goes on (eg it would take 10 minuts to research a 50 damage weapon, but 30 minutes to research it straight to 100 damage), then the players would be encouraged to make small advancments at a time (it would be due to the uncertainly in the tech needed to achieve the desired result).
This system would allow players to design the tech nodes that they want and give more strategic variation to the game (eg: a player might in one game focus on fire rate of weapons, but in another game they might focus on raw damage power depending on the tactical situations that occur).
In curent tech trees, there are nodes that give you the tech advancment. They are hard coded into the game, end even if you only have access to certain ones of them in some playthroughs and not other, they are still hard points.
My idea was to have a system where the player describes the goal of the research and the game calculates how much effort (time and resources) the player has to put into it to reach it.
Some advancments would give a speed boost to the research for certain technologies (for example "Theory of Relitivity" might increase the speed of researching "GPS navigation systems"). This way you could still have prerequsites, but they wouold not be a hard prerequsite.
As an example of this sytem, you might want a laser weapons that does 100 points of damage and fires 10 shots per second.
So to get that, the game would work out that you need to research "laser weapons" to get the base technology, then research it long enough in "damage output" to reach 100 damage and also "fireing speed" to 10 shots per second. It would then give the player an indication of the time and costs involved in doing this.
IF this was too long a time, they might choose toinstead first research a laser weapon that only did 10 damage each hit and fired 5 shots per second and later on research further to get their 100 damage and 10 shots per second.
If you made it so that there was a geometrical increase as time goes on (eg it would take 10 minuts to research a 50 damage weapon, but 30 minutes to research it straight to 100 damage), then the players would be encouraged to make small advancments at a time (it would be due to the uncertainly in the tech needed to achieve the desired result).
This system would allow players to design the tech nodes that they want and give more strategic variation to the game (eg: a player might in one game focus on fire rate of weapons, but in another game they might focus on raw damage power depending on the tactical situations that occur).
Quote:Original post by Edtharan
As an example of this sytem, you might want a laser weapons that does 100 points of damage and fires 10 shots per second.
So to get that, the game would work out that you need to research "laser weapons" to get the base technology, then research it long enough in "damage output" to reach 100 damage and also "fireing speed" to 10 shots per second. It would then give the player an indication of the time and costs involved in doing this.
I don't think this is applicable in many games. Any realtime game should have too much going to allow you to play with sliders and calculate which setting is optimal for your current situation.
Even in turn based games like civilization I would not research every tech in this way, since I really like fixed tech with a short note on historical data, nice pictures and eventually a short film-sequence as a reward for researching some technology.
A much better approach could be to build tech-labs and assign them parts of the tech tree to work on (like economy, military etc - or general for small advancements in all fields). If you assign the same branch to several labs they give dimishing returns. Also already researched tech should get cheaper for everyone else, eventually leaking completely after when it is in use for some time.
Finelly for designing specific units you could to it similar to alpha centaury. For example if I have tech for armored vehicles, infantry armor, vehicle laser guns and automatic infantry weapons I could design some mobile infantery unit driving around in their vehicle and jumping out for combat. This specific combination of used tech is made into a research project which can be assigned to a military tech-lab.
But the general teching becomes too much of hazzle if you have to set a dozend parameters every third turn.
Most 4x games I've played have some randomness in research, I suggest you play these games first.
I've always wondered how a 4x game with randomized conditions equal for and displayed to all players would work out.
Imagine that on the loading screen, the game displays a graph showing a tech tree, building requirements, unit requirements and abilities, etc, and each time you play the options available are randomly chosen - the game could have a kind of behind-the-scenes "point buy" system that it uses to build random units and cost them out - advantages like '5x base health' would raise "points" and 'requires tier 2 building' or 'costs double gold' or 'takes extra time to build' or 'takes 2 food' could reduce "points", with all units generated to have a given point value (randomizing this per game could be interesting - some games you're playing superhuman ubersoldat with 50 point units and the next you're playing a peasant uprising with -10 point units).
Of course, such a game could not be as complex per play as other games are because the randomization would take a lot of work to understand and take advantage of, but averaged over the long term, such a system might still lead to a kind of deep gameplay.
On the other hand, having a 'build your own units' with the same 'point buy' system used for randomization could provide an interesting alternate game mode - instead of seeing the random data on the loading screen, you'd see the choices your opponent made for their units/tech/etc and they would see your choices. Everybody would have to create a diverse set of units to allow them adapt to different opponent choices (all the pointing being done before game, you have to have a counter for at least a few strategies built into your set of units/etc)
Imagine that on the loading screen, the game displays a graph showing a tech tree, building requirements, unit requirements and abilities, etc, and each time you play the options available are randomly chosen - the game could have a kind of behind-the-scenes "point buy" system that it uses to build random units and cost them out - advantages like '5x base health' would raise "points" and 'requires tier 2 building' or 'costs double gold' or 'takes extra time to build' or 'takes 2 food' could reduce "points", with all units generated to have a given point value (randomizing this per game could be interesting - some games you're playing superhuman ubersoldat with 50 point units and the next you're playing a peasant uprising with -10 point units).
Of course, such a game could not be as complex per play as other games are because the randomization would take a lot of work to understand and take advantage of, but averaged over the long term, such a system might still lead to a kind of deep gameplay.
On the other hand, having a 'build your own units' with the same 'point buy' system used for randomization could provide an interesting alternate game mode - instead of seeing the random data on the loading screen, you'd see the choices your opponent made for their units/tech/etc and they would see your choices. Everybody would have to create a diverse set of units to allow them adapt to different opponent choices (all the pointing being done before game, you have to have a counter for at least a few strategies built into your set of units/etc)
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement