Jump to content

moruvis

Member
 View Profile  See their activity

  • Posts

    6

  • Joined

  • Last visited

 Content Type 

Posts posted by moruvis

    Automated Mining

    in General Discussions

    Posted

    I think maybe everyone is underestimating how difficult it will be to write this autonomous miner. Getting it to do what you want, easy, getting it to not do what you don't want, very hard. A medium level of complexity might get something that could go off and start mining but would require some manual intervention every few min to stop it getting caught in a loop, or suck in a hole or whatever. If you are afk then the bot is vulnerable to other players interfering with it.

     

    If scripting complexity carries an in-game cost to running it (eg. power requirement) then the economics of running it break down at a point. A higher complexity system that was economical to run would require smaller specialized units for specific tasks and then the game becomes commanding the fleet of your automated bots. It would be more productive overall, i.e. extract a greater volume of resource, but it would become a much more intensive process where there is a genuine skill in coordination (and not the eve-style sit reading a book and click the screen once a minute, in my opinion this actually makes mining interesting)

     

     

    Also, why macro when you can just script?

    DevBlog: LUA Scripting and Distributed Processing Units (DPUs)

    in DevBlog Feedback

    Posted

    I would guess to safely run untrusted code they would need scan the code before it is run to make sure it doesn't try to reference any system objects that have references to the executing environment that sort of thing. If we are going to be able to script some actual AI or drones then we might get some official libraries that have thrust calculations and path finding algorithms already made, those would be able to be implemented in optimized c++ that is called from the Lua script to save on cycle time on the server end.

     

    I have a couple of questions:

    1. Data storage: Already been asked and I agree, if we are going to be able to make anything more than a fly-by-wire interface for ships then there will probably be need for some long term persistent data storage. At its most basic this would allow remote sensor probes, surveying mining spots, collecting data for tech research.
    2. Sensors: I don't think it was explicitly mentioned but we would need access to various different types of sensors, eg. gravimetrics, positioning, surveying, friend-or-foe etc. Would those be separate modules you add and then access through a specific interface?
    3. Communication: Autonomous drones would need some communication between each other and back to the base ship to propagate orders, will there be a messaging framework for this that would need to be coded or would work outside the scope of the scripting?
    4. Libraries: Will the code only exist on the unit that has it programmed or will there be an independent repository of code that each player has / can access that will then be deployed for execution? Similarly, will we have some sort of version control? I wouldn't be able to write much without an in-game git before it all became a mess.
    5. Testing: Could we have a sandbox within the sandbox to test code? If you were to write a drone AI it might decide to start attacking everyone and hurt your standing if you got some code wrong.
    6. Collaboration: Will groups of players be able to work on the same codebase to produce something that a single player wouldn't have the time or skill to make themselves?

    I know all of this would start adding quite a lot of complexity to something that might not need to be (and maybe it is because I don't know that much about Lua), but the whole idea excites me with the possibility of lots of interesting and rich gameplay ideas. 'Scripter' could be a legitimate in-game profession if scripts and be written, sold and distributed independently of actually having something that runs it. Writing a decent drone AI might be outside the skillset of almost all players, but selling a library to manufacturers or having a scripting division of a corporation adds a whole extra level to the game. The ingame skill is a real skill rather than an eve-like wait-20-hours-and-you-can-do-it skill.

     

    If communication is handled within the script then writing a communication protocol for drones would allow hacking to be a legitimate battle tack-tick. Finding a flaw in the protocol, deliberately confusing the sensors, running a DDoS attack on the baseship to block communications, destroy the communication array on the base ship and then the drones can't receive orders (without some mind bogglingly difficult code to resolve a new ship to receive new orders from without specifically having been instructed to do so), firing an EMP and wiping the installed scripts before it reboots, and so much more than I can even think of right now would all become possible.

     

    The idea that most intrigues me is the idea of boarding a ship or space station and trying to take control of it. If everything is controlled by a control panel that is just an interface for calling procedures in the installed scripts then who controls say a space station or that sub system is the person standing in front of it. Does that mean that on a public space station there needs to be some locked doors to prevent people that shouldn't have access from entering the control room? Is that locked door controlled with a password? Can I get that password from a traitor or spy? Can I blow a hole in the side of the station to get me into the controlled area? Can I cut through the door? Can I lead a mutant with some other players to take over the station? Is there some subsystem outside the controlled area that I can use with a hole in the physical logic of the station to start gaining access? What if I own it and I've accidentally locked myself out? So many possibilities that would lead to some amazing organic gameplay that a game maker would never be able to design.

     

     

    I'm getting too excited by this...

    DevBlog: From Barter to Market Economy

    in DevBlog Feedback

    Posted

    With multiple currencies, it is just an issue of supply and demand, anyone could create their own currency and just say that they own some amount of it, but unless it can actually facilitate the purpose of trade, i.e. someone else will accept it in exchange for goods then you wouldn't sell your own goods in that currency. Multiple currencies are exchanged just like other commodities, so if there were multiple currencies you could chain a transaction with the exchange rate to quote in any currency you want. The exchange rate depends on the relative value of the currency, and the bid-offer spread will depend on how many people actively use it, so if no one else used you wouldn't be able to exchange it anyway.

     

    While multiple currencies would be interesting and probably lead to some interesting gameplay I would be far more interested in market exchanges that offer futures, options and swap contracts on commodities as well as bond and equity markets. This seems to be the point where most game economies stop, since there is no explicit support for it there is the issue of credit risk, but it sounds even without explicit support some scripting could be done to lock away funds in margin accounts. 

×
×
  • Create New...