An idea for restructiring research

[lastpick]

So we have somewhat realistic production chains, however it seems as if we are just using them to produce fancier research bottles. I offer an alternative approach that would be somewhat more realistic and in the end I think more fun. What if research labs lived up to their name? Here below is a quick outline of what I am proposing.

1 Labs accumulate knowledge slowly over time and more quickly as they are being used, this is used to unlock higher tiers of research labs which are faster at producing resultant technology and as an efficiency modifier for the lab itself.
2 Remove the research tree, or rather replace it with a map showing what has been researched and what paths are still open.
3 Start with an initial Lab that accepts two inputs, allow the research to be guided by what the is placed into the research lab for research.
4 Once an item has been created in the lab (i.e. a prototype) it is then unlocked for manufacturing.
5 Continued research into an existing item by itself should lead to improvements in said item.
6 Each item has an associated knowledge value and minimum requirement for manufacturing. Knowledge can be lost when labs are destroyed (an immediate penalty to all knowledge values *Items being manufactured cannot fall below the minimum
knowledge value for that item) or when an item is not manufactured for a long period (this is how we retire items... note a direct example if tractors have replaced plows then the knowledge of making plows would disappear over time) which would lead to items no longer being available.
7 Lastly, manufacturing itself adds a small bonus to the knowledge value of an item which over time would unlock higher versions of the item.

An example,...
Hmmm I wonder what would happen if I place iron, copper, and sulphuric acid in this level three research lab....sometime later... Eureka our scientists have discovered fuel cells.
Or hmmm I have all this copper and iron, I wonder what the research lab can make with that... sometime later.... COOKWARE.... ummm not what I was expecting keep researching.... and then they discover Iron gears.
One last one lets add iron, copper, and coal to the lab and see what happens.... after awhile .... Graphite ... and so on.

With such a system we bring back discovery into the game and also add some additional risk to the player by losing knowledge. In such a system knowledge becomes a commodity itself. I also think it would help with keeping the list of manufacturing items manageable, for instance why show me wooden crates if I am no longer making them anymore? Or maybe a player just likes wood crates and we end up with smart crates made of wood and electronics. It would be interesting to see how manufacturing lists would evolve over time with different players.

[ssilk]

I'm not against this. You know, this game is about automation. How is it possible to automate this kind of research?


OT: I think something into this direction is useful, for a parallel way of research. Like "I need lasers now, but for the long term I need speed modules."
Example: I use the gun now so often to kill biters, that I "learned" the piercing bullets. Some ideas from DyTech mod. Or: I have a basic research, which enables to research everything which needs only red and green potion and the current research, which enables to research the rest. Having them in "parallel research" seem to be a good idea for me. This is somehow the same problem as with the basic inserter, which needs advanced assembly.

[lastpick]

I agree the game is about automation which is why I'm suggesting that the way research is handled right now really breaks down when we consider that the ever increasing production chains all revolve around creating potions. If knowledge were treated like a commodity in the game I think it would allow the focus to shift back to building things that further the story of planetary domination not one of creating magical potions. Think of it in another way almost all progress in the real world happens iteratively which results in the occasional major breakthrough that then opens up new avenues of research. By letting the player guide the progress of research through their actual usage or the labs it would allow the focus to shift back to player driven goals and not centered on creating say a blue potion or a green potion just so that a better gun can be created to deal with an immediate threat.

[lastpick]

Okay,

Now for a more technical game design discussion then of how we would make this work. First what I'm suggesting would mean that the current research oriented production chains would be obsolete, which would be a major change. So I do understand that it may be to late in the design of the game to move the focus away from these production chains that all revolve around creating potions as the end product. Perhaps, if I had discovered the game several months ago it might have been more feasible to sway the developers to a different path.

As for how this system I'm proposing would work, well I think the labs would need supply chains to them much like the steam generators or chemical plants. However, once the resources were available research itself would happen based on what was put into the lab. So lets do a run through. I as a new player build an initial lab which only has room for 2 items in it. So to start off I notice that on my map I have very little coal and a lot of trees. I'd like to use them for something so I put some tree logs in the lab and wait to see what happens. Behind the scenes unknown to the player, there is a value assigned to each tech, which when reached would unlock that piece of technology. Also behind the scenes is a value representing knowledge which is a resource that the lab is providing (think of it like how oil is represented in the game now). Over time the created knowledge get's applied to the items being researched. In keeping with our example let's take an item say a wood chest, that behind the scenes would require a knowledge value of 10 in wood to be unlocked. Our lab in this example is creating 3 knowledge per minute with nearly a non-existent research bonus because it has no experience. That would mean the wood in the lab being researched would in roughly 4 minutes have accumulated enough research to unlock the wood chest. Now what would happen is that the player would get notified that they had unlocked this tech, at this point the wood chest in this example would have a knowledge value of 12. This is important because at this point the item a wood chest is unlocked and has a knowledge value assigned that is higher than the minimum required to unlock it. However, if the player then didn't manufacture a chest that value would slowly start to drop, albeit slowly. If the knowledge value dropped below 10 then that tech would disappear from the game until re-discovered. Continuing with our example, let's assume the player got distracted because they found a nest of biters which put the urgency on staying alive and not building a chest to store stuff. In this case let's say the knowledge value for the wood chest has dropped to 9, something the player doesn't see but does realize that chest's are no longer available. So he put's the wood back in the lab and in less than a minute the tech would again be unlocked.

One neat consequence of this would be that initial techs could be granted at the beginning of the game to boost initial game-play, however if the player doesn't keep the knowledge up on those techs then the ability to manufacture them would drop. Kind of like the idea that if society collapsed tomorrow I might have my car and know how to use it but there's no way I'm going to be building my own anytime in the near future because it would take a lot of knowledge I don't currently have. So in the short term I'd still be able to use my car until it broke down but afterwards well...

Lastly, I think that with a system like this it would be easier to manage the dependency problems we have now, because with the idea of a dynamic knowledge value assigned to each item there would be more flexibility to allow for recovering from a missing tech. The hard part would be in setting up the required knowledge recipes for each item definition as well as the decay rate for knowledge on items.

[ssilk]

I would say even one year ago it wouldn't be possible to bring this in as suggestion. This is a major change.

The most irritating thing to me is, that this kind of research doesn't need any big amounts of resources. That would make everything ridiculous easy - as I understand it - and that undermines the sense of the game, which "is a game about mining resources, planning and building factories, automating production and fighting alien enemies."

Maybe that can change over time, but the second is: You don't see anymore, how much you already researched. That is also a major break in the game, cause everything in factorio can be seen anywhere. I think this is one of the most basic game elements: Instead of hiding mysterious game mechanics it show the mechanics on screen, every stupid item can be followed and every item has it's own way through the system until it vanishes, either for energy, for defence, fight or in a lab. But even then it accomplishes a sense: energy is produced, we killed some biter or we researched a new tech.

I think this has some way into factorio, I think the DyTech way is really cool, and I could imagine a comination of this with that. For example the research for higher levels of modular armor.

There have been already made suggestions into that direction and you brought me to the point to search for it: https://forums.factorio.com/forum/vie ... hine#p6671