Factorio TAS planning
3 months ago
New Zealand

Speedrun planning can be incredibly complicated and Factorio doubles down on this by making every previous decision continually impactful for the duration of the run. Tool assisted speedruns (TAS) become even more specific & precise. Theis maintains a Factorio TAS helper mod, already helpful in building Zaspar's Any% TAS and Ameateye's Steelaxe TAS.

Now Omni has undertaken a TAS project for the significantly more complicated Factorio speedrun category of Default Settings. (Not a completely random map seed, but hey.) My goal here is to centralise planning & speculate on high level milestones, timings, build sizes, etc, to help determine in wide vague terms what this DS TAS might develop into.

At best this is a collection of descriptions so we don't have to continually scroll up through Discord. The most up to date information on the actual gameplay of the Factorio DS TAS will be either the TAS channel on the Factorio speedrun Discord, or Omni's Twitch channel.

What this post is not

This is not a specification of/for the DS TAS build, nor is it an immutable reference, nor is it necessarily up to date. It's also not community agreement of any preferred plan, though I will try make updates where/if such convergence appears.

It's also not really about TAS-specific tech or tricks like belt manipulation for boosting movement speed. That stuff is cool and important and maybe turns up in replies, but I'm intending here to summarise size, speed, layout & timings of the constructed base. Wide routing concerns like what order to visit which mining patches.

New Zealand

Guessing at numbers

State of plan: vague, tentative, speculative

These are numbers based on doing handwavingly better than the Nefrum's DS run of 2h 19m 12s, the WR at the time of planning.

  • TAS rocket launch: 1h 45m target because why not, improves RTA by 35m?
  • Chemical science: 16x assembler 2 before 50m(?) (Nefrums gets 14x @ 57m)
  • Utility science: 18x assembler 2 with 2x prodmod1 before 1h 29m (Nefrums: 15x @ 1h 52m)
  • Production science: 16x assembler 2 with 2x prodmod1 before (Nefrums gets 15x @ 1h 52m)

The intent is to work backwards from these goals to help estimate build scale, component requirements, then mall sizing and design patterns. Expect all of this to change as we figure what we're doing.

The TAS gets to skip construction robotics - that's 350 blue science saved - with the build happening so early, a mild increase in blue science can still confer substantial overall time savings.

Key intermediates:

  • Green chips: several builds, complicated & messy timings
  • Red chips: ??? (Nefrums: 24x assem2 @54m + 40x @1h 43m, none with prodmods)
  • Blue chips: ??? (Nefrums: 20x assem2 with 2xP1 @54m)
  • Steel: ??? (Nefrums: mini steel + 1 ore belt @41, +1 ore belt @1h 25m)
  • Petrochem: ??? (Nefrums: 9 refineries @50m, advanced recipe @ 1h 3m, +4 refineries later)

Those are all empty spaces waiting for us to get a better idea / agreement of build scale, likely also highly map-specifc.

New Zealand

Patch allocation

Allocation state: weird, handwaving?

Seed 2417168439 has been selected for the TAS DS speedrun, the map of the 1h 28m 36s DS multiplayer run (WR at time of planning). Asserting resource requirements might inform which patches can be used in which order for which inputs / products. Omni has annotated a map with miner counts which can also be unambiguous patch names.

  • starter patches (called "1p"): infrastructure, mall, no science. TAS implemented.
  • decent close west coal (unindexed): early powergrid. TAS implemented.
  • tiny close northwest coal (unindexed): belt it north for smelting fuel?
  • coal 16 (north): belt it north for smelting fuel?
  • iron 87 (north): likely first iron expansion, also bugs, now just within TAS planning horizon
  • copper 137(north): mall & science soon after 1p
  • iron 123 (northeast): pretty nearby, but also bugs
  • oil 3018%(northwest): first oil? good defensive position on the shore
  • oil 1862% (northeast): quite a safe position
  • iron 136 & 164 (east): close enough together they might be analysed the same
  • copper 109 (northwest): it's big and not too far, in desert but no close nests
  • copper 104 (west): also big, also bugs
  • coal 120 (west): powergrid? mixed build smelting fuel?
  • copper 107 (north): it's a long way, perhaps the last copper connection?
  • stone 68 (east): awkward, but also the most viable non-starter stone
  • iron 84 (east): has serious bug problems
  • stone 83 (east): also serious bug problems
  • unindexed south coal: pretty far away, no number assigned, not sure anyone cares

Contributions

Calculating backwards from how much we need and knowing which patches will supply, perhaps different speedrunners contribute blueprint designs for different parts? These aren't guaranteed to be implemented by the TAS but might be helpful resources.

Game stats
Followers
1,020
Runs
2,908
Players
759
Latest news
Latest threads
Posted 2 months ago
12 replies
Posted 2 months ago
0 replies
Posted 2 months ago
2 replies
Posted 2 months ago
20 replies