[0.17] There is no spoon!

[mward]

I just completed "There is no spoon" with 6 minutes to spare!

Here are some hints and ideas I came up with along the way:

(1) Obviously you will want to max out all resources: apart from Uranium, which you won't need.

(2) Turning off the biters will disable the achievement: but there is a new terrain map setting for "a large island in an endless ocean". If you set the island size to 75% and set the starting area (the area with no biters) to max (600%) then you might find that there are no biters at all on your island. The island will have plenty of space and resources.

(3) Construction bots are essential with 0.17: the higher tier sciences require a lot of resources, so you will need a large base. Rush to get the bots as quickly as possible, but leave space for a main bus and for lots of miners and smelting arrays.

(4) You don't need military research (as I eventually realised... )

(5) Once you have construction bots, you can slow down and take your time(!) Save the game, carefully lay out the next section of the base: either adding to the main bus or adding more miners and smelting arrays. Make sure you don't change any part of the existing base. When you are happy with the layout, take a blueprint of it (including some of the existing base). Then re-load your save and place the blueprint for the bots to build. When most of it has been built, take another save and start laying out the next section.

(6) When you get productivity modules, start making four level 3 modules for the rocket silo. (I forgot to do this, so had to make do without!)

(7) Stockpile red and blue circuits, rocket fuel and low density structures as soon as you start making them.

(8) Level 3 assemblers are too expensive, as are modules in anything except the rocket silo. Most of my base was yellow belts with the odd stretch of red belt where space was tight.

(9) You don't need to launch a satellite! (I was frantically hunting for the satellite recipe as the last few minutes ticked down: then found that it required Space Science research, which needed 2,000 each of the other science packs! So I launched without it and still won the game )

Post your experiences here please

[torne]

(6) When you get productivity modules, start making four level 3 modules for the rocket silo. (I forgot to do this, so had to make do without!)

(8) Level 3 assemblers are too expensive, as are modules in anything except the rocket silo. Most of my base was yellow belts with the odd stretch of red belt where space was tight.

Helpful tip from past versions' spoon threads which still applies in 0.17: As soon as you get prod 1 research done, make 80 prod 1 modules (the number you will eventually need to make the four level 3 ones for the rocket silo) and then use them in the recipes that have the best payoffs (expensive science, research itself, blue circuits, etc), and only convert them into the prod 3 modules for the rocket silo once you actually build and start supplying the silo

It doesn't cost any extra resources since you were going to make them anyway, and it can save you a large amount.

[mward]

As soon as you get prod 1 research done, make 80 prod 1 modules (the number you will eventually need to make the four level 3 ones for the rocket silo) and then use them in the recipes that have the best payoffs (expensive science, research itself, blue circuits, etc), and only convert them into the prod 3 modules for the rocket silo once you actually build and start supplying the silo

That's a good idea. Note that the "best payoffs" might not necessarily be the most expensive items, but rather the ones you need more of and for which you can't easily scale up production. For example: I had plenty of blue circuits (they had their own two red belts of green circuits), but started running out of red circuits.

[leadraven]

I' made it with a margin of 23 minutes the other day. No bots - you need only one setup, nothing to copy-paste.
I widely applied Prod 1 : in labs ASAP, yellow and purple science, red and blue circuits, expensive intermediates.
Red circuits were the most dangerous bottleneck. They need at least doubling after all-science setup done.

[mward]

I made it with a margin of 23 minutes the other day. No bots

I take my hat off to your planning and placing abilities! I would be interested to see the final result.

One of the nice things about Factorio is that there are many ways to play: and you can still get the no spoon achievement even if you don't like speed running

[leadraven]

Here are my smelter array, main bus with science factory, and oil refinery. Power is 80 steam engines.
I don't like speedrunning, just wanted to test my skills. If I ever want to repeat this, I definitely will reserve place for second red circuits setup.
Productivity modules 1 pay back quickly enough to use them in 8h run.
Welcome to the elite club of 1.5%

[torne]

Note that the "best payoffs" might not necessarily be the most expensive items, but rather the ones you need more of and for which you can't easily scale up production. For example: I had plenty of blue circuits (they had their own two red belts of green circuits), but started running out of red circuits.

Since the modules themselves are "free" in this scenario and can be moved around any time, and you're using just productivity modules with no speed, I'm not sure I follow your logic here.

You have to build more factories to make the same amount of something if you put prod modules in it, because of the speed penalty - you just don't have to make more of all the factories for its inputs. So, the thing that is the most important is, I think, the same in a speed run as it is in any other game: which things save the most input resources all the way down the production chain.

So, in your case: if you had plenty of blue circuits, you could have built a couple of extra blue circuit assemblers and then put prod modules in all of them, and it would consume fewer red circuits while producing the same amount of blue, and you said you were short of red circuits so that extra production would come in handy

For example, if you had 10 blue circuit assembler 2s making 45 per minute at full speed, you could replace that with 14 of them with two prod 1 modules in each running at *almost* full speed for the same 45/minute (you only need 13.3 to keep up), and that would leave you with an extra 6.67 red circuits per minute that you could route elsewhere where you needed them. (also 80 green per minute - 1.33 factories' worth that you could use for something else or just not bother building at all).

Putting productivity modules in the red circuit assemblers would be much less effective: making that same 6.67 red circuits per minute takes ~0.89 assembler 2s normally, but needs ~1.18 with prod 1 modules due to the speed penalty, so your red circuit factory would need to be even bigger than it already was.

Productivity modules save you from building more *inputs* for a given recipe, not from building more of that recipe itself, so it's best to use them in things whose inputs are expensive, even in a speed run. The inputs for red circuits are pretty cheap - the only reason you need so many assemblers making them is because the recipe is slow, and productivity modules just make it even slower

[mward]

I'm not sure I follow your logic here.

The logic is that blue circuits only use a fraction of the red circuits.

My factory was producing about 60 blue circuits a minute: so the blue circuits were using 120 red circuits per minute. Putting productivity modules in blue circuits would save 8% of the red circuits: so about 10 per minute (plus a load of green circuits: but my factory is swimming with green circuits!). I was producing around 450 red circuits per minute and the productivity modules would give about a 5% bonus (too many assemblers for them all to be filled). This would give 22.5 extra red circuits per minute.

[torne]

I'm not sure I follow your logic here.

The logic is that blue circuits only use a fraction of the red circuits.

My factory was producing about 60 blue circuits a minute: so the blue circuits were using 120 red circuits per minute. Putting productivity modules in blue circuits would save 8% of the red circuits: so about 10 per minute (plus a load of green circuits: but my factory is swimming with green circuits!). I was producing around 450 red circuits per minute and the productivity modules would give about a 5% bonus (too many assemblers for them all to be filled). This would give 22.5 extra red circuits per minute.

But it wouldn't though - if you were making 450/minute that's 60 assembler 2's, and if you put two prod 1 modules in all of them that would *reduce* your red circuit output to 340.2/minute due to the speed penalty: you're producing 8% extra output for free from each assembler, but they are running at only 70% speed, so the total output is .7 * 1.08 = 0.756. You would have to build an extra 20 red circuit assemblers just to get back to 450/minute.

As you say there aren't enough modules to actually fill 60 assemblers with two each, so the overall penalty would be a bit less, but it's never going to be positive unless you are also using speed modules to counteract the speed penalty.

[leadraven]

With Prod modules you are free to balance production the way you want :

1. More output with same input
2. Same output with fewer input

When I had shortage of red circuits, I made [1] for red circuits and [2] for everything that consumes them. Sure, [1] has priority as it applies all red circuits produced.

[torne]

Right, but my point is that's oversimplifying it and the real options (for assembler 2s with 2x prod 1):

1. 108% output with the same input, if you increase the number of assemblers by ~43%
2. Same output with 93% of the input, if you increase the number of assemblers by ~32%
3. 76% output with the same input, if you don't increase the number of assemblers at all

If you have a shortage of red circuits, adding the prod modules to the red circuit assemblers is a bad choice here: you will have to build a ton more red circuit assemblers to avoid your output rate going *down*, which will mean you need to use up more of your limited supply of modules for them, which means you can't use them anywhere else.

If you're using a fixed number of modules for a speedrun (the 80 you need anyway) and the machines with the modules in are always running, then the cost/benefit analysis always comes down to the same basic things: the benefit is just how many input resources are saved by the module being in a particular machine at a particular time, and the cost is just the *opportunity cost of not putting the module in a different machine* - which is, in turn, just determined by how many input resources you would save if you put it there instead.

If the machines are *not* always running, either because output is backed up or input is too short, the modules are less effective there and might be better someplace else, but you can either solve this by moving the modules around at different times, or by building more efficiently in the first place such that your machines don't back up (which is wasted time in a speedrun; you either will need more of that thing later in which case you should be buffering it into a chest, or you won't in which case you should be cutting off the input to the machines entirely and stripping their modules out to move someplace else).

Edit: also, that's enough of me posting this here and hijacking the thread. The main point is: you have 80 prod module 1s for free and you should use them rather than only making the 4 prod 3's, to speed up your run. If people want to discuss the exact best places to use them that would probably be better in another dedicated thread

[mward]

If the machines are *not* always running, ... you can either solve this ... or by building more efficiently in the first place

Some people like to design the most efficient layouts and ratios.

Others just like solving problems: so we play the game in a way which creates more problems I guess that the first group will never know the joy of threading a belt of science packs through a messy factory because they didn't leave space for it in the first place

If you haven't created the most efficient factory in the first place, then an otherwise sub-optimal placement of modules could still help out.

The point of this thread is that you can still get the achievement without a lot of careful planning

De gustibus non est disputandum

[leadraven]

If you have a shortage of red circuits, adding the prod modules to the red circuit assemblers is a bad choice here: you will have to build a ton more red circuit assemblers to avoid your output rate going *down*, which will mean you need to use up more of your limited supply of modules for them, which means you can't use them anywhere else.

Modules 1 are cheap and plentiful. More input means more logistics, and it's a major issue. More assemblers isn't a problem, more belts is. I need to get maximum from limited input.

[mrvn]

(5) Once you have construction bots, you can slow down and take your time(!) Save the game, carefully lay out the next section of the base: either adding to the main bus or adding more miners and smelting arrays. Make sure you don't change any part of the existing base. When you are happy with the layout, take a blueprint of it (including some of the existing base). Then re-load your save and place the blueprint for the bots to build. When most of it has been built, take another save and start laying out the next section.

That's basically cheating.

[mward]

That's basically cheating.

Who is being cheated?

Is using blueprints cheating? Why then are they in the game?

Is using blueprints that you created in a different save cheating? Why then do they remain available?

[Koub]

(5) Once you have construction bots, you can slow down and take your time(!) Save the game, carefully lay out the next section of the base: either adding to the main bus or adding more miners and smelting arrays. Make sure you don't change any part of the existing base. When you are happy with the layout, take a blueprint of it (including some of the existing base). Then re-load your save and place the blueprint for the bots to build. When most of it has been built, take another save and start laying out the next section.

That's basically cheating.

What's the difference between that, and people who develop builds in creative/sandbox with god mode, and just plonk their blueprints on their other games ?

Or maybe I might just have missed the irony in your remark .

[mrvn]

(5) Once you have construction bots, you can slow down and take your time(!) Save the game, carefully lay out the next section of the base: either adding to the main bus or adding more miners and smelting arrays. Make sure you don't change any part of the existing base. When you are happy with the layout, take a blueprint of it (including some of the existing base). Then re-load your save and place the blueprint for the bots to build. When most of it has been built, take another save and start laying out the next section.

That's basically cheating.

What's the difference between that, and people who develop builds in creative/sandbox with god mode, and just plonk their blueprints on their other games ?

Or maybe I might just have missed the irony in your remark .

The difference is that you are playing against the clock and constantly turning it back.

[mrvn]

That's basically cheating.

Who is being cheated?

Is using blueprints cheating? Why then are they in the game?

Is using blueprints that you created in a different save cheating? Why then do they remain available?

You are cheating yourself.

There is nothing wrong with using blueprints for a normal game. But in a game where the goal is to design and build your base against the clock you are taking out half the challenge by constantly turning back the clock to let you try out designs till they are perfect.

[E-37]

I don't consider myself terribly good at the game and I just got the achievement 3 days ago on my first try. (I also got raining bullets and steam all the way by accident) I got it on default settings with no cliffs and I didn't have a very hard time. The biters were a pain since I just came from my 1k SPM megabase where I had spent the last 200 hours with my artillery basically removing biters. Other than that there wasn't too much of a challenge. I was able to get to the final 3 technologies in the first 5 hours and then got kinda stuck on purple science since I didn't plan on the rocket silo requiring it.

That's basically cheating.

Who is being cheated?

Is using blueprints cheating? Why then are they in the game?

Is using blueprints that you created in a different save cheating? Why then do they remain available?

You are cheating yourself.

There is nothing wrong with using blueprints for a normal game. But in a game where the goal is to design and build your base against the clock you are taking out half the challenge by constantly turning back the clock to let you try out designs till they are perfect.

If I can do it on default world settings without doing any tricks, I don't think that with rich resources and no biters that anyone would have trouble with it. I guess it is a challenge for yourself. You need to decide what rules you view as fair and then complete it based on your own rules. It is a singleplayer sandbox game. You get to define the rules for yourself. Just because someone got it by doing something that you think is cheating doesn't reduce your achievement.

[Koub]

I just discovered a recent Factorio speedrunning guide by Nefrums (the current record holder for Factorio speedrun in 2h28mn with 0.17.x) :
https://www.youtube.com/watch?v=h0HDZnYFfY4

I found it extremely helpful, and as speedrunning is the most extreme version of the spoon challenge, I think others might take advantage of the advice Nefrums gives to speedrunner wannabes.