Factorio Forums

2k barrels is 4x chests, so if I have 4 for loading and 4 for unloading, it ensures the loading ones are always full, and unloading ones just about empty. Anyway, I've played around a bit and added a couple more items into the circuit, I'm just letting it run for a while with 2 trains in parallel to make sure it all keeps in sync - http://i.imgur.com/uOWD0dd.jpg

I think I found a way to do the 'a AND (b OR c)', and I don't believe combinators can be used since a and c are relating to item counts on the train, not in the chests. It needs a bit of testing, but I think 'a AND b or a AND c' does the same thing.

Edit: Actually, it doesn't appear to work properly, when c was false, but b was true, it didn't leave the station, which is a bit annoying

just for comparison, you only need a dash under over 10k crude per rocket if you use PM3 all the way up your oil chain, in comparison to your (correct) calculation of 44k crude with no productivity modules. You also only need about 40k ores compared to 200k without. Definitely PM3 is the way to go if you are targetting a serious rocket output.

My little calcs (using Solver):

edit: correction (double counted final rocket-parts stage):

I don't know how good you are with Solver, but an extra challenge for you might be to add the amount of rockets you need to produce in order to amortize the cost of the Prod3 modules (you may assume no speed beacons are used at all).

siggboy wrote:I don't know how good you are with Solver, but an extra challenge for you might be to add the amount of rockets you need to produce in order to amortize the cost of the Prod3 modules (you may assume no speed beacons are used at all).

that is no problem if you assume the PM3 are made either with all PM3 or all without PM3. The tricky part is how in reality productivity (1, 2 and 3) are used to 'bootstrap' themselves as the player ramps up his base. The further tricky part is that speed beacons massively increase the 'profitability' of PM3 so you would definitely want to use them with it.

In my current factory plan (construction still in progress ), I am using a smart forge with 8 speed beacons effecting each furnace, 8 per green circuits, 8 per steel (0.13 fucked up the steel part of my old smart forge, hence the seperate block) and about 4 for every other production stage.

Also please note I messed up in my posted image and double counted the final rocket parts productivity . Corrected image at http://imgur.com/a/K03kj

Well for simplicity let's just assume that the cost of making the modules is a flat 100% (the effect from ramping up the module production is not interesting here, I want to know the amortization effect of the rocket production) and that max speed beacons are used everywhere (8 effect sources from the speed beacons).

siggboy wrote:Well for simplicity let's just assume that the cost of making the modules is a flat 100% (the effect from ramping up the module production is not interesting here, I want to know the amortization effect of the rocket production) and that max speed beacons are used everywhere (8 effect sources from the speed beacons).

bootstrapping is highly relevant when talking of amortisation because everything after breakeven is pure profit. 1 PM3 = about 1.2k ores, 270 crude when built 100% PM3s, or 3.2k ores, 660 crude without

But anyway, rough calcs reading off my calculator:

1 PM3 (built without PM3) = c. 3k ores, 660 crude
1 rocket without PM3 = 35k ores, 8k crude
1 rocket with PM3 everywhere = 8.7k ores, 2.2k crude

Thus PM3 saving per rocket (basic rocket - PM3 rocket) ~= 26k ores, 5.8k crude

Therefors the savings on each launch are about equal to 26/3 = 9 PM3 modules (and FYI, 1 PM3 is about equal to 1.2 SM3).

The scale of your factory (i.e. speed of rocket production) determines how many PM3 and SM3 you will need and so how many launches before breakeven in absolute terms, but you can divide the rough number of modules required by 9 to decide from a ratio perspective of modules 'bought' per launch

This only applies to modules throughout the production chain. You can easily reduce the breakeven point by only using PM3 in high throughput and high value areas (the silo being number 1, number 2 green circuits)

Thank you for this question - it was interesting to run my 5 min rocket factory plan into the same breakeven calcs and found: I will need about 900 modules (I being conservative as I hope to squeeze a bit more beacon overlap out but oh well...), which as 90% of them were built with a PM3 bootstrapping factory cost on average about 1.5k. Therefore breakeven = 900 * 26 / 1.5k = about 50 launches

Thanks for working this out. The exact time required does not really matter, as long as the time frame is reasonable. If it took 10,000 launches to amortize it would be monkey business to try to "save" resources that way.

Let me point out though that a complete amortization calculation also needs to take into account the additional energy spent.