Speedrunning the game

[Calebrj]

Hi. My name's Caleb, and I've been playing this game for a few weeks now. How is it that people speedrun the game in 3 or 4 hours, when I have not even finished my first map in 3 or 4 weeks? Some sort of method? Or do they just know what they are doing, and they do it really well?

[Ohz]

Play Zelda Ocarina of Time on N64 on your own, and then watch the speed run. Same ratio.

[wahming]

Some sort of method? Or do they just know what they are doing, and they do it really well?

What's the difference?

[Calebrj]

The difference is that a method a beginner can use; whereas the other option takes experience and time. IS there a method to speedrunning the game?

[DaveMcW]

The simplest method is to beat the game normally and take a picture of your final factory. Then restart, and try to rebuild your factory as fast as possible. You should be able to get under 15 hours using that method.

More optimizations include: no trains, no roboports, no electric furnaces, no solar panels, build bigger mines and smelting lines, turn on peaceful mode, and rig the map generation to have tons of resources.

[Calebrj]

Ok. Thanks!


Does anyone have any tips as to defeating the aliens? I don't have it set to peaceful (stupid) and every once in a while, I practically rebuild my base. I'm also unfortunate to have a HUGE alien base practically right next to me. I take a single step in the wrong direction, and... well, yeah. How can I get this alien base downsized?

[DaveMcW]

Turret creep. Build a line of turrets towards the base, leapfrogging the last turret up to the front.

20 Gun turrets + 1000 ammo
or
20 Laser turrets + 50 steam engines

[Calebrj]

Ok. I'll give it a try. Repost in about 10 minutes some of the results.


Results:

Got steamrolled. Again.

Anything else?

P.S. : I used about 40 laser turrets, with full speed and damage upgrades researched. Made a dent. But then they all came at me, and I couldn't withstand that.

[DaveMcW]

Is your electric network producing the 96 MW that 40 laser turrets require?

If you have finished the tech tree, you can also use destroyer capsules.

[Calebrj]

I'm producing about double that. (I do have a HUGE wind farm) and no, I don't have destroyer capsules yet. Right now I'm trying to get productivity module 3 so I can research the rocket pad.

[Mr. Tact]

How did the turret creep fail? Even at full evolution that method is pretty tried and true....

[Mr. Tact]

Check this out - go to about 9:30 in this video: https://www.youtube.com/watch?v=3YmtgGpbYac

Now, Zisteau is using personal roboports to place the turrets but that isn't necessary -- just convenient. Even with the crazy increased alien settings Zisteau has used, turret creep works. Granted, he is using large amounts of turrets (as he needs to)..

[Koub]

The things you need to keep in mind while speedrunning :
- Take into account how much ressource you need to do some research and build what the research allows you to. If you can make it without it, and un less time, then don't research it. The only things you should research and build are those that will make you arrive to the end of the game faster, INCLUDING the time you spent getting them.
- At one time, you have to find the sweet spot between building more to get things done faster and building less because each building is more ressources to invest into basically not launching the rocket
- Do a lot of planning. I mean a LOT of planning. The idea is you need to get to a point where you have a build order from beginning to winning. You must know how much of everything you'll be needing at any point of your speedrun.
- Speedrunners have permanently (or close to) manual crafting queue full of stuff so that every second of gameplay, they are always crafting something.
- Most speedrunners I've seen play in peaceful, and do their turret creep very early to get the number of alien artifacts they'll need before pollution has made aliens evolve too much.
- Speedrunning is a very specialized kind of game practice. Most people won't take pleasure in it. If it's not your thing, you'll probably get bored pretty fast.

[huwp]

To kill the aliens you need to take out the spawners effectively. Pushing ahead with turrets alone will only get you so far. You can use the tank with regular-shells to destroy spawners in two shots and drive around picking them off, when your health gets too low then it's time to retreat to the turrets and repair for the next drive by.

Biggest threat are the worms, which can be killed with a bunch of stacked poison capsules. Distraction capsules can give you valuable seconds to take out more spawners.

Any weapon works, as long as each time you can get out alive with less spawners than there were before. Make sure to have enough ammo at hand to finish the job, take too long and they'll grow back again.

[KatherineOfSky]

To prepare for my speedrun, I've been watching a few of those done by AntiElitz. Lots of careful planning.

https://www.youtube.com/watch?v=IL-kB3gdVIs

If he can do it in 2 hours, I reckon I can make it in just under 8 for the achievement

[daniel34]

To prepare for my speedrun, I've been watching a few of those done by AntiElitz. Lots of careful planning.

https://www.youtube.com/watch?v=IL-kB3gdVIs

If he can do it in 2 hours, I reckon I can make it in just under 8 for the achievement

I also made a speedrun in 0.12 as preparation for the 0.13 8h achievement. I managed to do it in 5h 38min on a peaceful random map with custom settings (low frequency, high size, high richness).

I've made a lot of mistakes, like only realizing I'd need 200 electric engines right before crafting the rocket silo, and only researching and crafting solar panels and accumulators when it was required by the satellite.
Given this new information and the fact that my copper/iron smelting was lacking in the late game I think I can do it in 4:30h, but only my next run in 0.13 can tell.

As for power: my factory required 40 MW, so extending the steam setup to 8 x 10 steam engines whenever you have time is recommended.
For technology: Only research what you'll actually need. Don't research walls/gates if you don't need them, nor bullet speed/damage if you're taking them down with laser turrets.

tl;dr If you know what you'll need for the endgame/rocket and plan ahead, <5 hours should be not that difficult to achieve.

[Nova]

He said wind farm, so I guess he plays with mods. Normal rules don't apply then.

Turret creep always works in the normal game with enough energy, even if the enemy has huge bases on the highest evolution factor.

[tuplex]

I was curious to see how quickly I could get to a rocket launch on a randomly-generated map. Without mods, my quickest was 6h30m, and without having to rush things or make a spaghetti base. I used a main bus like I would with any normal base. What I found to be essential to doing it quickly is to know in advance what you need in terms of production. When I started my game, I already knew that I needed 64 iron and copper furnaces, 48 steel, 12 green circuit assemblers, 16 red, etc. The Foreman app is great for this - I set it up with 20 blue assembler machines for each of the rocket parts, and then worked backwards to find out what I needed to support that.

TLDR - If you know from the beginning what capacity you will need, it makes building up your base much more efficient.

[purdueme91]

Any other general tips that anyone can provide? My fastest right now is 10.5 hours and I dissected my last run so I can try to get faster. I was always running out of steel and power so I'm going to extend those lines. My science seemed slow once I hit purple. I thought 10 labs would be enough. Should I shoot for 20? I've been researching some lab efficiency tech (I think up to 3) or is that a waste of time? My next run will be strictly steam engines (with full belts of coal this time too ). Too many times with brownouts when solar shut off and accumulators went dry. Thanks for any help

[BlakeMW]

I think that early on about 20 labs and later about 40 is a good target to aim for (as a general rule the end game researches take about twice as much time per research cycle as the early game researches, so you need to double the number of labs even if you don't increase science pack production). Bear in mind all the resources going to science pack production will later be diverted to rocket parts production, so it is ideal to do research on a really grand scale, you can't really put too much resources into research.