Jump to content

Scientific Research


Heresiarch

Recommended Posts

So I have not heard anything about in game research yet.
Now I am not saying that research is a needed feature for this game, it is not.
But then again, this game tries to emulate our universe in a computer, and give us the means to build up a new civilisation.
In this grand context, and especially in a science fiction game, I think that research would add a lot of depth, believability, and gameplay options and goals.

It is also something I have spend a good bit of time thinking on over the years - how you could represent it in a game with maximal flexibility (the way this game handles Organisations for example).

So let me try to describe the framework I would utilize. I can always explain certain aspects in more detail, if people are interested, and I would very much like to hear about weak points in the concept! :-)

For the purpose of this concept, I am going to assume that functional elements in this game are crafted gathering the needed materials, and then assembling them (how else would you do it?).

My concept:

The buildingplan of any functional element (weapon, cockpit, shields, armor, electronic warfare, ...) has a component called "build pattern", which gives the building machine the precise instructions on how to construct the result.

==Science machine==
Now we add a new machine to the game, lets call it "science machine".
The science machine takes a "build pattern" (see above) as input. As it's a big bulky supercomputer, it also needs a lot of energy, and possibly other ressources to run.
What it does, is that with a very very low probability, it throws out an improved version of the build pattern (think on average once every 10 years). It just runs continously until that happens or it is turned off.
That improved version has one of the stats of the previous version improved by a little bit (1% or 2% maybe) compared to the base values of the input build pattern. (I think it is very very important that this improvement is multiplicative - many games make the mistake of making them additive. If you make them additive, later improvements will matter way less than earlier ones, which means that improving everything equally will be the obvious best strategy, killing variety.)
There is also a chance that a different stat has gotten worse be a small amount.

==Represenation of science developement==
Now there should not be very many possibilities for new versions from any given base "build pattern" - you could for example have 1 of 5 different metrics which could improve, and 1 of 5 different metrics that could get worse as a trade off -> 25 possibilities.
If we make sure the possibility space stays about that small, we can express it with a single symbol (letter/number)!
That means, that on close examination of the build pattern / the functional element build with it, we can show that string of symbols to the player (if he has the right skill).
If that player compares two such strings, he can even see at what point in the development process the two components diverged! That is a pretty cool feature. B-)
That compactness should also make the server side storage very manageable, while still keeping the door open for future possibilities like merging of different technologies, or reverse engineering of stolen hardware.

If you want to make this system even more complex(!), you can even let every advancement modify the building cost of the component. I would not randomize that though, but only make it dependent on the current improvement, how complex the technology already is, and possibly also which improvements came before (e.g. how often you improved into the same direction). That way, you keep the number of possibilites the same, and can still express this more complex variant with one symbol per improvement. Increasing ressource usage can also be a good way to balance the improved stats of the functional elements a bit. :-)

 

By the way, getting the same symbol when building upon the same base building plan (same symbol string), literally means that you just made an identical improvement. Which means that different players can by chance make the same improvement to a technology - which I think is a very cool thing. :-)

You can think of that symbol string as a very compact form of representing the DNA of the tech in question.

==Scientific progress==
Another thing this representation is very useful for, is guiding the science machine by the game server.
I said earlier that it will only succeed very rarely (you just let it run endlessly). Obviously it would succeed more often the lower level the base tech is (=symbol string of improvements is short).
But the server could also change the probabilites of different improvements based on the history of improvements already made.

So if it takes so very long for the machine to have a success, how do we as players get our hands on new technology before 10 years have passed?
Parallelisation! A big alliance might make a big building with 100 science machines, which all search in parallel for an improvement on the best tech the alliance could get its hand on as base technology.
And every time an improvement is found, you'll have to weight the pros and cons of using it as the new base technology for further research: Yes, you just made the tech better, but you also just made it more complicated (symbol string got longer), so any further improvement will be just a little bit harder.
That might not make a lot of difference for a single improvement (the work should only become a few % harder for each improvement), but it does compound, and at some point you might just have made a technological dead end which can't compete against a different technological strain whcih has been kept more efficient. :-)

==Effects in game==
So now that I have outlined (hopefully halfway understandibly) how I would handle scientific research, what effects would that have in game? We (hypothetically) just spend a good amount of work to incorporate a fairly complex mechanic, so there should be a payoff ingame for that, right!?

- Not all components are the same! Without research, all lasers of a certain size in the game would be functionally identical (maybe with some small bonuses for being exceptionally well crafted). With research, you might buy lasers that fire 10% faster, or instead lasers that do 10% more damage per shot, if both variants have been researched.
- Not everybody has access to the same stuff! Without research, as long as you crank your skills up high enough, and gather all the ressources, you are able to build all the functional elements in the game. Now with research, everybody can still build a laser, but if you don't have the build patterns for the variant with +10% damage, then you can't build that variant. And maybe the alliance that developed it is jelously guarding it and only using it for its own ships? Now you need spies to steal it! :-D
- There is a sense of progress! If you are in a current state of the art warship, and you fight against a similar ship that has not been upgraded in the last 3 years, you will probably come out ahead - Your shields are better, your weapons do a bit more damage! This gives players more to do, and feels very scifi-like to me...
- It opens up additional game play paths and goals like building a research center, collecting a scientific library, having a monopol on certain advanced tech, stealing tech, trading tech, and more.

Stuff not touched upon in this post:

- How you could implement reverse engineering and merging of tech (I'll just state that its possible to do within the framework outlined).
- Pseudoscience to explain all of this in game (think simulation/evolution, but I'll gladly go into detail if there is interest)

- Effects on markets. How can you have a market for lasers of a certain size if those lasers have different tech levels? I don't have an answer for that one yet, but would love to hear more details about how markets are going to be implemented. I have a hunch that this kind of problem has already occured elsewhere in developing them (e.g. if you think about lasers getting bonuses due to high crafting skill).

I am really curious what people think about all of this. Please poke holes into it!
I'll update this concept post as I get suggestions to improve it (if I decide to take them ;-)).

Link to comment
Share on other sites

I don't think research should be a in game thing, it would make more sense as a metagame feature, like an alloy composition, molecule etc. Having a linear stat based crafting system would be required, as it would not be possible with a standard crafting system with fixed recipes.

Link to comment
Share on other sites

I don't think research should be a in game thing, it would make more sense as a metagame feature, like an alloy composition, molecule etc. Having a linear stat based crafting system would be required, as it would not be possible with a standard crafting system with fixed recipes.

 

You mean kinda like how the skill system is projected to be? i.e. you learn the skill you selected kinda passively over time till it's complete. At least, that's my understanding of the skill system.

Link to comment
Share on other sites

You mean kinda like how the skill system is projected to be? i.e. you learn the skill you selected kinda passively over time till it's complete. At least, that's my understanding of the skill system.

No with metagame i mean the research is not like a tree where you do stuff to unlock features, you already have all features, you just have to discover them

Link to comment
Share on other sites

I don't think research should be a in game thing, it would make more sense as a metagame feature, like an alloy composition, molecule etc. Having a linear stat based crafting system would be required, as it would not be possible with a standard crafting system with fixed recipes.

Well, having research only in the sense that the developers add in new items, is of course a possibility. I already outlined the advantages I see in having research in the game, but in the end, this is just the idea box forum, right? ;-)

 

As for the changing recipes: It depends on the crafting system.

If you immediately go from ressources to finished product, it is no problem:

Say a laser weapon takes 100 iron ore and 20 unobtanium to make. You can then have the rule that every damage time your tech increases the damage of the weapon, you increase the iron ore requirement by 1%. So after the 1. increase the recipe would cost 101 iron ore, after the 2. increase 102.01 (rounded down to 102).

The fun thing is, the player does not need to know these rules, so they can be way more complex. All he sees is the new builing plan for the improved laser weapon, which lists the ressources it needs.

 

Now what if a a laser weapon recipe instead needs an emitter, a weapon base, a ball bearing, a tracking computer and 2 cans of oil to be build?

In that case I would not make the laser weapon recipe itself improvable by research at all. Instead you could tech up the recipe for the emitter (increasing firing speed, damage, and energy draw), and the tracking computer (improving tracking speed and electronic warfare resistance).

 

I think there should be a prototype machine where you try to make an item from the "DNA" so you would guess the meaning of the symbols.

.

Almost, but not quite. ;)

 

If somebody just texted you the symbols of the tech in the game chat - that just won't be enough information to do anything with (well not quite, it might still be usefull in reverse engineering, see below).

 

If you got your hands on a building plan (either by buying or by stealing it), then you have everything you need to build the item yourself - these are the detailed instructions after all!

 

But what if you only captured an item? That is reverse engineering, which I did not explain in my first post, because that is an addon to the basic concept (additional work to implement, and not needed for the framework to function).

The way I imagine reverse engineering might work, is that you have a scanner you use to analyse your item as well as possible. Depending on your own skill, the quality of your scanner and maybe the amount of time you spend, you would get a data package of a certain quality out of that (the data package is an item). So you don't even have to aquire the item - A spy with enough time alone with the item could smuggle that analysed data out!

 

You would put that data package into a reverse engineering machine (ain't I grand at naming things? :P), together with a building plan for a different variant of the same item (so still a laser, but with +10% firing speed instead of +10% damage).

Now the engineering machine goes to work (endlessly), and if you are lucky, at some point it will spit out the building plan for the item your data package came from. (Optional: Even if it has not managed to give you the building plan, after some time it will start to figure out the symbols of the tech used to make the item whose data you are analysing (Not all at once, one symbol at a time), allowing you to look for a closer technological match to feed the machine.)

 

What does the difficulty for the reverse engineering machine depend on?

-The amount of commonality between the tech you are analysing and the comparison item you gave it, as determined by how long the tech DNA strings stay identical before diverging (this should be the deciding factor - if the analysed tech is only a minor variation of tech you already know, it will be way easier to figure out compared to if development forked of a long time ago).

-The tech level of the item it is analysing, the higher the harder

-The tech of the reverse engineering machine ;)

Link to comment
Share on other sites

@Heisach my point was there is no fixed recipe, you can make a gun with x material but you have to find the optimal one for damage, accuracy, ammo contained, or a balanced one, those middle points would be hidden, and would have to be researched. but if you already know them, you can make them again.

I would even go to the point where the recipe doesn't necessarily use one material but a family: for instance halogens, or radioactive isotopes, and each one adds complexity in the path to find the optimal one, and then you can sell the recipe on the market (patent wise or as is (equivalent to be able to generate a small % of sales or have a big price up-front))

Link to comment
Share on other sites

What wicpar essentially means is a meta puzzle game with unknown rules so that the players actually have to try stuff (you know like experiments in real science)

 

I'm in favor of this but I dunno about the patent part.

Very quickly a wiki will be established for something like this which could dampen the fun.

Link to comment
Share on other sites

Thats the problem I see with any "just figure it out in game" schemes - thanks to information on the internet, the best solutions will be known very quickly, and from that point onwards, very little experimentation will happen.

You definitely would not be able to make money ingame for the recipe, if just knowing the information is enough.

Link to comment
Share on other sites

What wicpar essentially means is a meta puzzle game with unknown rules so that the players actually have to try stuff (you know like experiments in real science)

 

I'm in favor of this but I dunno about the patent part.

Very quickly a wiki will be established for something like this which could dampen the fun.

with the patent thing i was just saying that the person would get a % of the buy price of the recipe resulting item, in every transaction, this would allow to get a lot of money if you have a heavily traded and produced recipe result. one can only discover a recipe once, others that find it won't have that bonus.

 

The progression formulas wouldn't be lust like one best, and then tend to a single best value, it would be a random value map more like, that has no limit but the bigger a stat is the rarer it is. This would prevent from ever finding the best one, just finding a best one. The recipe would be the seed of the results random stats.

Link to comment
Share on other sites

with the patent thing i was just saying that the person would get a % of the buy price of the recipe resulting item, in every transaction, this would allow to get a lot of money if you have a heavily traded and produced recipe result. one can only discover a recipe once, others that find it won't have that bonus.

 

The progression formulas wouldn't be lust like one best, and then tend to a single best value, it would be a random value map more like, that has no limit but the bigger a stat is the rarer it is. This would prevent from ever finding the best one, just finding a best one. The recipe would be the seed of the results random stats.

I suppose game mechanics wise that might work out fine. I just don't think it would add much fun to the game. And it would totally break both immersion and the spirit of the game:

We are supposed to build up a new society by our own imagination, the way we see fit. The developers are giving us all the tools for that, e.g. minecraft like capabilities to make the architecture we want, and generic "Organisations" which we can sculpt from anything from democracy to dictatorship, and into other stuff we don't even have names for yet.

Why the hell should there be a game mechanic which enforces a patent office? Which physical law keeps me from rediscovering something another guy or gal figured out before me!?

 

I am sorry, but very strongly opposed.

Link to comment
Share on other sites

Thats the problem I see with any "just figure it out in game" schemes - thanks to information on the internet, the best solutions will be known very quickly, and from that point onwards, very little experimentation will happen.

You definitely would not be able to make money ingame for the recipe, if just knowing the information is enough.

One way to fix is that Novaquark would continuously shift the stats, so that 1 really high quality iron is now just regular iron and vice versa.

Link to comment
Share on other sites

I suppose game mechanics wise that might work out fine. I just don't think it would add much fun to the game. And it would totally break both immersion and the spirit of the game:

We are supposed to build up a new society by our own imagination, the way we see fit. The developers are giving us all the tools for that, e.g. minecraft like capabilities to make the architecture we want, and generic "Organisations" which we can sculpt from anything from democracy to dictatorship, and into other stuff we don't even have names for yet.

Why the hell should there be a game mechanic which enforces a patent office? Which physical law keeps me from rediscovering something another guy or gal figured out before me!?

 

I am sorry, but very strongly opposed.

Yes i see, the patent thing would be a way to make it worth while... but isn't necessary, there is a simple alternative: do nothing. Keep it a company secret or sell it once to a big corp or give it openly to everyone, i think you will agree with that mechanic instead of the % thing, as you said: it isn't in the spirit of the game.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...