Skip to content
This repository was archived by the owner on May 22, 2025. It is now read-only.

Conversation

@Bibby0110
Copy link
Contributor

@Bibby0110 Bibby0110 commented Aug 10, 2022

Document the changes in your pull request

Instead of AI data cores always being connected you'll now have to wire them up using ethernet cables.
The "home" network of the AI is the one where its current core is. If the current core fails it will connect to a different core in the network.
You can link 2 remote networks with networking devices (sprites pending, currently solar panels in the screenshot). In this case the AI will also attempt to relocate to a remote core, and switch it's home network. This is done recursively so you can nest networks if you wanted that for some reason.
If you have a data core in a separate network, aka in a network that isn't the AIs main network, you'll need a server cabinet with 2Thz of CPU and 3TB of RAM for the core to be "active"

Explanatory picture (Not actual roundstart setup):
image
Network outlined in red contains the core and a server cabinet. Also contains a networking device(solar panel)
Network outlines in blue contains only a server cabinet and networking device.
The two networking devices are connected, connecting the red and blue networks and allowing them to share processing power. Connection signified by black line. (Note that solar panels point towards each other, allowing for simple tracking of clusters)

Why?

Why this change you may ask? Mainly for the following 3 reasons (There are more that I have forgotten):

  1. Removal of the AI control console.
    Making AIs confined to physical networks instead of all data cores means it's now possible to model hacking into the AIs network physically, without being able to build a new core and doing it there. This way you'll need physical access to download the AI like before.
  2. Separation of networks
    Wanna run 2 AIs with separate resources? You can do that now.
    Wanna build an AI on lavaland? You can do that now (Reenables AIs for ghost roles)
  3. Making computing networks independent
    Having computer networks be physical like this allows us to use them for things that aren't just AI related. This way the Netmin will be able to use the processing clusters for bitcoin mining, research and [REDACTED, COMING SOON]

TODO

The things I am yet to complete

  • Require local computing resources for data cores to be active
  • Actually consider local resources for AIs
  • Resource allocation console moved
  • Remove the control console
  • UI for networking devices
  • Allow plugging into individual AI networks
  • Allow uploading of AIs after plugging in
  • Allow downloading of AIs after plugging in
  • Allow individual control of networks through laptops. Allow for bitcoin mining on networks with no AI/cores present
  • Allow control of networking devices from console/laptop
  • Netmin prioritised when AI is present
  • Server overview + removal of old consoles
  • A sprite for networking machines maybe (orca working on it)
  • Map in networking machines on each map
  • (Expand box sci-maint to create a proper netmin area?)
  • (Decouple netmin from NTSL?)

Changelog

🆑
experimental: AI networks are no longer based on bluespace wireless technology. You'll now need to wire up individual servers with data cores for them to work
experimental: Each separate network with data cores need 2 Thz of CPU and 4TB of RAM to function
experimental: You can link networks together using networking machines. The allowed size of the network you're connecting from depends on the upgrade level of the networking device. Connecting two networks does not remove the above criteria for data cores
experimental: AI uploading and downloading is no longer done through the AI control console. You will now need a modular laptop to hook into the network by gaining physical control. This process can be done through a connected network if you so desire
/:cl:

@Bibby0110 Bibby0110 marked this pull request as ready for review September 1, 2022 15:41
@Bibby0110 Bibby0110 requested review from a team and monster860 as code owners September 1, 2022 15:41
@Bibby0110 Bibby0110 closed this Sep 10, 2022
@Bibby0110 Bibby0110 reopened this Sep 15, 2022
Yogbot-13 added a commit that referenced this pull request Sep 15, 2022
@Bibby0110 Bibby0110 requested a review from a team as a code owner September 15, 2022 17:40
Yogbot-13 added a commit that referenced this pull request Sep 15, 2022

added_access = list(ACCESS_ENGINE, ACCESS_ENGINE_EQUIP, ACCESS_MAINT_TUNNELS)
base_access = list(ACCESS_TCOMSAT, ACCESS_TCOM_ADMIN, ACCESS_TECH_STORAGE, ACCESS_RC_ANNOUNCE, ACCESS_CONSTRUCTION, ACCESS_MECH_ENGINE, ACCESS_NETWORK, ACCESS_RESEARCH, ACCESS_MINISAT)
base_access = list(ACCESS_TCOM_ADMIN, ACCESS_TECH_STORAGE, ACCESS_RC_ANNOUNCE, ACCESS_CONSTRUCTION, ACCESS_MECH_ENGINE, ACCESS_NETWORK, ACCESS_RESEARCH, ACCESS_MINISAT, ACCESS_TOX)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fix access

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

Admin This PR affects administrators Mapping Will result in a change to a map.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants