Factorio is two games: pre and post robots.

[blazespinnaker]

I've found lately when I play factorio, I pretty much stop once I get robots, or start a new game at robots.

The problem of the race to get to robots can be done in different ways, and especially so in deathworld. And then with mines and robots, the game really changes. For example, basic deathworld is no longer a challenge and it's just a regular robot game, with the bots replenishing mines when they get hit by incoming biters.

In DW, there is some minor hassle in clearing out spawning biter bases, but that is just a tedious task. I'd actually be curious what the most efficient way is to do this. Right now I do it with mines for defense, flamers and rockets, but I'm convinced there is a better way. Obviously artillery on islands with no path for biters is best, but getting to artillery takes time and a few bases need to be cleared before then.

Admittedly, railworld + deathworld type settings are more problematic and trains become an interesting part of game play. Definitely a very fun option for sure. I'd also love to see a hard game play option where biters actively go for poles/tracks/belts/pipes - everything really. EMF/fluid sensing where they travel long distances to take out power and pipes. That would be an interesting challenge. (viewtopic.php?f=6&t=92012)

The reason why the game is different once you're at robots is that you have different options for parallel building activities that can be exponential because placement time no longer is constrained by your walking / keyboard speed, but rather just constrained by resources in the game (which are theoretically infinite). See Dave's math here:

viewtopic.php?p=521707#p521707

Robots really change the game. I've done some experiments with imported blue prints, and when optimized for robot layout you can get compound building efficiencies via JIT manufacturing. I explain the ideas behind the math here with concrete examples from wave defense:

viewtopic.php?f=5&t=90012

I think if recursive blueprints were added to the game, this would bring even greater clarity to the game regarding post robot play and would make it a more pure play. I've wondered why they haven't added this. Could be because they don't want to make the game too programmy.

That all said, there is still the problem of getting to robots. On monty hall maps, Steelaxe and GOTLAP speedruns I think are the right way to get to robots, where they do large amounts of intense and optimal handfeeding.

https://www.speedrun.com/factorio/run/zgw15ljy

GOTLAP is almost the same as getting to oil.

Macros really is an expert at this, and I've seen some of his reasoning. He's on the right track I think for the most optimal speedrun game to Rocket launch, and the other runners that build up infra are over weighting their infra setups. I sort of understand why they do it, what Macros does is very intense and requires a sustained effort that is a bit dull compared to building cool factories.

But if you extrapolate from GOTLAP (24 minutes) it is reasonable to say it wouldn't take much more to get to Robots. Just need the plastic for red circuits. Engines you can start early and are not a huge bottleneck though you do need sufficient iron ore production. My amateurish experiments show it takes me about 10-15 minutes to get to bots after oil on Any%, but I imagine the pros can do it in less.

You need about ~20K iron ore, and about 10K copper to get to bots from the beginning.

If you can get to bots in 35 minutes, than from there I believe you can get to rocket launch in 20-25 minutes with the right map on Single Player speedruns. Again, I refer back to Dave's math, which on a good Any% monty hall map isn't that improbable to get really high numbers of production. With flexible and optimal atomic designs easy to hand craft and taking into account robot travelling speeds I believe you can copy/paste your way to massive over production quite quickly.

One particular game style I'm interested in is massive multiplayer with blueprint import speedruns. With this format, players can initially pool their burners/furnaces and get out massive ore production very quickly. From there I think they can likely get to robots even faster than single player, and then with blueprint import, I believe rocket launch is much shorter than what you can do without it. There is no need to do hand crafting of memorized blueprints and blueprints can have arbitrary complexity not optimized for keyboard dexterity.

JD has gotten this down to 1 hour 20 minutes, but if I had to guess, you get it down to sub 40 minutes or so. 20 minutes for robots and 10-15 for rocket launch as your minions can all be massively hand feeding to build robots.

JD unfortunately doesn't go down the bots route, which I think is a mistake. https://www.speedrun.com/factorio/run/ylvvlxrm

Imported blueprints and bots are error free and what I saw in wave defense was pretty amazing in terms of bootstrapping out to massive production. And as I mentioned, they can be arbitrarily complex and micro managing blueprints is much easier when they are imported.

One last comment here, getting to bots on DW can be done in a hacky/cheaty style via mining ore and deconstructing it before your pollution cloud invites biter attacks. Once you get mines, you can settle in, defend and really mine ore to get to bots. Getting oil isn't that hard, unless every patch is overrun with biters and that gets a little tricky, though with mines you have pretty safe defensive/offensive power. In that situation, it's best to get the oil first before biters have really evolved into medium/big versions.

For those who try to build bases and defend early, I wonder how they do it successfully. I find trying to build bases in the beginning and defending against biters a pain as they evolve quickly and the resources you end up using to defend is rather extensive. You spend a lot of time repairing/replacing. Some maps on DW can be easier than others tho.

You can also set up a DW type map with even more challenging options that make this strategy less effective. Eg, railworld+super hard DW (desert, no trees, few resources, etc) You can probably make maps which make even defending with bots/mines challenging - which I think I'll probably try next.

[blazespinnaker]

As a follow up on the last strategy, I would use miners to grab the ore, and then smelt it in some other place. That way you can minimize the pollution cloud you build up while getting the ore and can max that resource grab out as much as possible. Watching the pollution production tab gives a good warning when your cloud hits a biter base and has always given me enough time to deconstruct my mining operation.

Likely if you're good at clearing out nearby bases, that might work as well early on. I'm not particularly good at it and I find I get killed off rather frequently, especially if the base is has several spawners. It also means you're accelerating the evolution early on, so other bases or spawned bases will be a problem until you get to mines.

[Yoyobuae]

As a follow up on the last strategy, I would use miners to grab the ore, and then smelt it in some other place. That way you can minimize the pollution cloud you build up while getting the ore and can max that resource grab out as much as possible. Watching the pollution production tab gives a good warning when your cloud hits a biter base and has always given me enough time to deconstruct my mining operation.

You can cheese out 10k ore out of a modest size starting patch before even the biter attack comes, even with a biter base just two~three chunks away. And when the biters come you deconstruct. Once the buildings are deconstructed it doesn't matter if you die even, just retrieve the items from corpse.

I use that kind of strategies in my pacifist deathworld runs.

[blazespinnaker]

I like to play hard core and restart if I die

But yeah, it sort of depends on the map. Sometime a biter base will be like right next to the ore patch.

How do you do it on your non pacifist maps? Clear out nearby bases? Even with grenades and lots of fish I tend to die. It's always touch and go.

I'm lousy on the keyboard tho. I suppose I could practice some keystrokes.

[blazespinnaker]

Wait, what, 10K ore with biter base 2 chunks away? Hmm, is that really possible?

Screenshot 2020-12-02 153419.png (2.1 MiB) Viewed 9289 times

On the first DW map I created, this is what I got. You can really get 10K ore from that?

[blazespinnaker]

Hmm, most I could get was 5K raw ore + 800 (used for miners, power) and that was leaving it until they overwhelmed my miners. Usually much more cautious than that.

Still, 5K is enough to get to close to oil/mines. a couple more pulls and you should be good to go.

[Yoyobuae]

Hmm, most I could get was 5K raw ore + 800 (used for miners, power) and that was leaving it until they overwhelmed my miners. Usually much more cautious than that.

Still, 5K is enough to get to close to oil/mines. a couple more pulls and you should be good to go.

Firstly, I'm talking about electric miners. Burner miners are just terrible (more pollution, half the speed and very energy intensive).

Also there's some degree of randomness to how early biters will attack and it also depends on the ore patch size. But yeah, getting 5k~10k per pull is about what I would expect.

If you manage to pick up and run you can go onto another resource patch and repeat the operation.

[blazespinnaker]

Yep, burners are not ideal for this scenario, though they work in others. Their production per sq m is better than miners I believe. Plus they don’t need copper.

Are Pollution inspired biter attacks random? How was that confirmed?

One thing I’d like to play with is wind direction and pollution. Not sure how much of a factor that is tho

[blazespinnaker]

Thinking about it, you might be able to cut the power when wind direction flows toward a biter base. Maybe leave a furnace going as a windsock of sorts.

[Yoyobuae]

Yep, burners are not ideal for this scenario, though they work in others. Their production per sq m is better than miners I believe. Plus they don’t need copper.

The thing that's really annoying about burners is they produce almost double the amount of pollution and consume 3x as much coal to run. The pollution is specially annoying because biter bases will continue randomly spawning attack groups for as long as pollution continues spreading into them.

So yes, you might get a few more resources, but then get penalized steeply with far longer waiting for pollution to dissipate enough to hit the same resource patch again (without getting immediately attacked).

Are Pollution inspired biter attacks random? How was that confirmed?

https://wiki.factorio.com/Enemies#Defense

One thing I’d like to play with is wind direction and pollution. Not sure how much of a factor that is tho

Wind plays zero role in the spread of pollution. Pollution spread depends only on the amount of pollution in a chunk (https://wiki.factorio.com/Pollution#Pollution_spread).

Thinking about it, you might be able to cut the power when wind direction flows toward a biter base. Maybe leave a furnace going as a windsock of sorts.

Biters don't care if the machines are running or not. They just head search up the pollution gradient towards areas of higher pollution, find some entities and then head there. Once they get there they can attack pretty much everything, whether it pollutes or not.

[blazespinnaker]

Yeaah, I saw that, but also this - viewtopic.php?t=27627

Not clear how the wiki is sourced, though it seems relatively authoritative.

[Koub]

Not clear how the wiki is sourced, though it seems relatively authoritative.

Mostly directly by the dev team, at least in the English version. More specifically Bilka.

[Yoyobuae]

Yep, burners are not ideal for this scenario, though they work in others. Their production per sq m is better than miners I believe. Plus they don’t need copper.

I gave burner miners a second chance I did find them a bit useful during the first few hours of a pacifist deathworld run with desert biome starting area. Their main benefit is their ease of setup specially with on-site smelting.

On-site smelting is useful because ore is harder to carry manually in large amounts. Plates can just be "stored" in handcrafting queue. I think once one can run belts for ore that benefit might not be as significant.

[blazespinnaker]

Yeah, my DW games are now pretty much until I get 10k ore and then I stop. After that, it’s fairly obvious how it will play out (research off site, get heavy armor, more pulls until oil/mines).

15 minutes for 10K is best I’ve done so far but it was an overly ideal map with a large time gap before biter attack.

One thing I am playing with is baiting biters with offsite burners polluting to give myself more time to do e-mining. That and hand mining the resources for the power/miners to avoid using burners at all.

I tried and I’ve found I am not a huge fan of deathworld marathon. It’s not really harder, but rather with expensive resource costs it drags everything out and slows the pace of the game. In some ways it dials down the biter difficulty as well. I got to mines on DW marathon and called it quits as I was getting bored. Once I am at mines the biters become an afterthought.

Maybe I’ll revisit marathon some other time. There’s a good marathon DW exchange string that was posted in the forums that I’ll look for and repost. The map was pretty challenging, though I’d switch off the marathon for now I think.

[blazespinnaker]

Btw, you are right about the on-site smelting, though I still try to do it in the midst of trees if I can. Trees provide absorption and power.

You can also buy some time by placing them further away from nests.

[Yoyobuae]

I got to mines on DW marathon and called it quits as I was getting bored. Once I am at mines the biters become an afterthought.

Yeah, land mines are kinda ridiculous powerful for how cheap they are. Can just spam them everywhere they kill all biters (at least before big ones).

[jodokus31]

I tried and I’ve found I am not a huge fan of deathworld marathon. It’s not really harder, but rather with expensive resource costs it drags everything out and slows the pace of the game. In some ways it dials down the biter difficulty as well. I got to mines on DW marathon and called it quits as I was getting bored. Once I am at mines the biters become an afterthought.

Personally, i see DW marathon as a slow Motion race against relativly even faster evolving biters. You are not so limited by placing speed and can react a bit faster. Also, you have to concentrate on really important researches. F.e. Heavy armor, fast inserter, etc. might not be that important.
I think it could be interesting to take the same biter evo speed to Progress speed relation as DW marathon and make it a bit faster. F.e. Only Double research and non expensive recipes, but more time and pollution evo and less attack consumption rate.

[jodokus31]

If you can get to bots in 35 minutes, than from there I believe you can get to rocket launch in 20-25 minutes with the right map on Single Player speedruns. Again, I refer back to Dave's math, which on a good Any% monty hall map isn't that improbable to get really high numbers of production. With flexible and optimal atomic designs easy to hand craft and taking into account robot travelling speeds I believe you can copy/paste your way to massive over production quite quickly.

You may not forget, that any% needs prepartion for late game, which GOTLAP does not need. some kind of mall. Also buffers of plates, gears, circuits, steel and stone for later.
Then bots take a long time to be crafted and until they are useful (chests, roboports...)

[jodokus31]

For those who try to build bases and defend early, I wonder how they do it successfully. I find trying to build bases in the beginning and defending against biters a pain as they evolve quickly and the resources you end up using to defend is rather extensive. You spend a lot of time repairing/replacing. Some maps on DW can be easier than others tho.

My strategy:
It's important to build compact and fast. Electric miners asap before science. Collect all the stone for furnaces via Rocks. Walls are expensive so pipes 2-3 layers are recommended around 4-6 turret camps. Rush grenades for fast clearing of big attacks. Heavy armor + grenades + shotgun + fish for fast and unexpensive base clearing, if base is to near. Physical damage 3 and stronger explosives 1 are helpful. Physical damage 2 has no advantages against Small biters, though. Stay on yellow ammo. Use assembler2 and steel furnace to reduce pollution. Rush flamers with walls before spitters show up, bc Spitters are quite deadly for gun turrets.

You can also set up a DW type map with even more challenging options that make this strategy less effective. Eg, railworld+super hard DW (desert, no trees, few resources, etc) You can probably make maps which make even defending with bots/mines challenging - which I think I'll probably try next.

Imo once you have bots/mines, nothing is really challenging as long you have oil and iron. Also flamers are quite useful against everything. But It's tougher to get there.
Maybe you try mods with tougher biters or declare self-limitations to dont use mines, flamers, efficiency modules, etc.

[blazespinnaker]

If you can get to bots in 35 minutes, than from there I believe you can get to rocket launch in 20-25 minutes with the right map on Single Player speedruns. Again, I refer back to Dave's math, which on a good Any% monty hall map isn't that improbable to get really high numbers of production. With flexible and optimal atomic designs easy to hand craft and taking into account robot travelling speeds I believe you can copy/paste your way to massive over production quite quickly.

You may not forget, that any% needs prepartion for late game, which GOTLAP does not need. some kind of mall. Also buffers of plates, gears, circuits, steel and stone for later.
Then bots take a long time to be crafted and until they are useful (chests, roboports...)

Nah, that prep IMHO is a waste of time. All that walking around and placement time is very unproductive and limiting, as it can only be done in a serial manner. Bots are the key, because they can remove walking around / placement by multitasking. Plus the belts, poles and inserters, is a total waste of copper and iron that could be used for research. Finally, handfeed is super fast and reduces inserter lag significantly. On top of this, assemblers can be hand crafted in 0.5s.

I'll prove this all out eventually, I guess. Getting world records isn't my thing, but the math backs it up. The speedrunners are too focused on keyboarding and memorizing sequences to do the math unfortunately.

It will require creating some clever blueprints, though, especially ones that can bootstrap up to over productive very quickly. Direct insertion at first to reduce latency between smelting and production, and then a more productive assembly line but still direct insertion, and then mass production (or maybe even another stage of direct insertion).

And the key is to create blueprints which can be easily re-used via recursive copy/past and minimize any custom building. Ie, an atomic blueprint would want to be the building block of larger, more meta blueprints. On top of this, you need to keep robot flying distances in mind and order of layout (belts, power first).

You also need the right math of what order to bootstrap with, miners / assemblers / bots / ?

It's a very interesting problem to solve and I think really goes to the soul of factorio and what it represents. I'll do it with blueprint import first and likely start at GOTLAP WR resources as a baseline (though I'll have oil instead). Maybe I'll learn the keyboard sequences if folks are still resistant to learn the new way, but I'd rather not. I'm more interested in the theory tbh

Plus, I have a life, so it'll probably take a few months before I get around to all of this. But nice to know there is a lot of interesting game play ahead!