By lhaxqlhaxq Last updated
This is a guide to Steelaxe% for all skill levels. Feel free to ask me any questions on discord or when I'm streaming at https://www.twitch.tv/lhaxq
I'll partition the run into two parts, the early game and the late game. The early game will be everything before the automation research and the late game everything after.
The early game has a fixed list of items that need to be (hand) crafted. For the build of this guide, this includes all mining drills and furnaces, 2 labs, 10 science packs and all the necessities for power. From this, we can derive the key fact of the early game: Assuming you are fast enough to always keep your crafting queue busy (after some point), the only way to save time over an existing run is to reach the point where the queue is always busy faster, or to reorder the queue. Concretely, the only time save is to make a change to the first ~70s or to move an item past the labs in the queue. The second option means that automation will be completed faster but less time will be available to craft intermediates for the assemblers. So:
If you are fast enough to always keep your queue busy (and build the labs and feed them as soon as possible), the only time save in the early game is in the first 70 seconds!
There are two small exceptions to this: First, extra stone furnace crafted before the queue is fully busy will save half a second (so good stone RNG saves time) and building burners faster will result in more resources for the late game, but this is a very small benefit.
This principle also applies to other categories but it will be less important there.
The late game again consists of two parts, researching steel and steel axe. The steel research consumes science packs much faster, so in general you will be bound by the speed of the assemblers producing science packs when researching steel, and by lab speed when researching steel axe. This means that it is important to start the assemblers for science packs as soon as possible and to always keep them running during the steel research, but it is ok to let them buffer in the assemblers for a bit. On the other hand, afterwards it's ok to have some of the the science pack assemblers not running some times, as long as all the labs are always researching. Having as many labs as possible is also crucial.
The assemblers for intermediates should almost always be running for the entirety of the late game. It's almost impossible to have too many intermediates but not enough plates. Gears should usually go first into the science pack assemblers and any leftovers can be turned into labs.
All runs for this guide use the same build except for minor variations. The burner setup was taken from Phoenix's setup. Compared to the setup used by macros, it has the advantage of only having two rows of burner pairs, making it less micro intensive to refuel and it is easier to walk to all the important places in the setup, as almost everything can be reached by only walking left/right between the burners and assemblers. In a setup where you need to walk in 4 directions it's easy to get stuck walking against something. It is also not too difficult to add mining drills to this setup.
You can of course also come up with your own build. If you use macros map, most of this guide should still apply. I would recommend not building less miners than I describe here though. If you've gotten fast enough that saving 4 seconds is a big time save, you already need to build this big, so you might as well do it from the beginning.
Beginner/intermediate mining/building order
Generally, place and fuel the miners and furnaces as soon as possible. Also smelt the ores as soon as possible by putting them into furnaces that don't have a miner yet. Turn all stone from the first 2 rocks into furnaces as soon as possible and place many furnaces at once.
Don't try to emulate the exact order of the WR run yet. Instead, do the following easier order (this order applies to every build I've seen on macros map, so even if you do your own build, try this order):
- Place 1 miner on iron, mine a tree and put 2 wood in the miner and 1 in the furnace
- Mine 1 coal rock
- Take the iron from the spaceship and furnace and craft a miner (iron)
- Mine 1 coal rock
- Mine 1 copper while crafting the next miner (iron)
- Mine 2 iron, craft 1 miner (copper) and 1 science pack
- Mine 1 more iron
- Craft 1 miner (iron) & 3 science packs
From here, I'll just give you the details of how to intersperse mining coal rocks with placing miners. Whenever you go to mine a rock, take all the resources, fill your queue with miners and science packs as necessary and go. Don't try to queue all the mining drills you need to place when you return, just do as many as you can & fill your queue with science packs and furnaces whenever it would be empty otherwise.
- Mine 1 coal rock
- Just take resources
- Mine 1 coal rock
- Place 2 miners on iron
- Mine 1 coal rock
- Place 3 miners on iron
- Mine 2 coal rocks
- Place 4 miners on iron, 4 on copper and 6 on coal, in that order
This run executes that building order, apart from a slight variation at the end, which you shouldn't do, except if you have to:
Make sure to have hotkeys for both types of plates, gears and science packs. Try to always keep your assemblers busy, but don't buffer too many items in the science pack assemblers. Anything above 4 (2 if you're more advanced) is overkill. It's usually best to first make sure the science pack assemblers are stocked, then take the intermediates and queue a lab and then throw all your remaining plates in the assemblers for the intermediates.
Beginner (Goal: Sub 11)
Your first goal as a beginner is to be sure to know the setup and the order it's being built in properly.
Don't worry too much about the exact order of the crafting queue or stone RNG, just try to always keep your queue busy and your burners running. I can't stress the importance of the queue enough. I watched some runs on the leaderboard that took 11-13 minutes and many of them could have easily saved over 30 seconds just by prioritizing their queue. After ~70s, consider every second you're not crafting as a second of time loss (this is of course not exactly true because the burners generate resources, but still). Remember that the 10 science packs take 55s to craft, so fitting them in your downtime is almost a minute of time save.
Make sure to have a hotkey for coal and drop twice as many pieces in the miners as in the furnaces. Always drop single pieces (using the drop item hotkey) but be generous, especially if you often see your burners being out of fuel.
If you don't know what to craft, go for stone furnaces, science packs or copper wire/circuits, just try to not craft too many furnaces or science packs (but don't worry if you craft a furnace or two too many!). For now, try to have exactly 10 science packs before you start crafting the labs (but if a few are missing when you could craft the labs, craft them after the labs). Make sure to place the labs and feed them as soon as possible, the first lab gets 6 and the second 4 packs. Apart from the miners, craft the least items necessary for running the labs before the labs, so only one steam engine and one pair of power poles. Anything you craft extra before the 2 labs is a time loss.
Intermediate (Goal: Sub 9:30)
Your next goal will be to build a routine that you can execute consistently.
Make sure you're familiar with the details of the route. You should know exactly how many pieces of ore you should mine and roughly when to take plates from the burners and when to refuel them. Your crafting queue should be the same every run and make sure to never drop it once you've got enough resources. Replace your last 2 science pack with crafting copper wires and green circuits, and craft the pack right after the labs. With this, your target for finishing automation should be sub 5.
If you're getting close to a 9:30 you might want to start resetting if your stone RNG is below average.
Make sure to start your assemblers as early as possible and keep everything fed while also keeping the buffer low enough to build labs with the remaining resources. It's important to find a rhythm for all those tasks and to practice the late game separately.
Advanced (Goal: Sub 9?)
Now it's probably time to switch to the build the WR uses.
If you want to improve substantially from here (without a new map), remember that you can only save time by reworking the queue and by optimizing the first 70s. Grind the early game out until you can reach the late game without dropping the queue. This will likely take a while but once you've got it down, every run that reaches the late game will be very close to WR pace with the only variables being stone RNG and the first 70s.
If you want to improve on this, see if you can find a science pack that can be moved until after the labs are crafted without losing time elsewhere. Moving a mining drill after the labs also saves time, but it also means it will produce less resources, but it might still be worth it.
If you are fast enough, you will hit the point where you will not be limited by the assembler speed anymore during the steel research, but instead by the lab speed. From there, you want to start optimizing on building labs during the steel research as well.
It's very difficult for me to say anything here about the steel axe research part, as this is my weak spot and it's much less optimized than I'd like. Just craft as many labs as you can, and don't forget placing and feeding them.
In general, the late game is full of trade-offs with unclear benefits. At some point in the future the late game will probably be as carefully planned out as the early game, but that's not yet the case (and it's unclear how much time could be saved by doing it).
- Can we find a better map? Might save 10s or more.
- More assemblers? More mining drills?
- Is moving a mining drill after the labs worth it? Lets us start research 4s earlier each, but costs over 10 plates that mining drill would have produced otherwise.
Rocket rush category
List of non-made categories.
Resources - New Save
Speedrunning with expensive recipes and 80% evo
New GOTLAP record.