The Productivity Cap and why the FFF #375 recycler shouldn't be introduced

[Epb7304]

TL;DR

The recycler shouldn't be introduced since it forces a cap on productivity bonuses, which go against the very nature of "the factory must grow"

There is also another option to fix this

What ?

The recycler (shown in FFF #375) shouldn't be introduced because it necessitates a cap on the amount of productivity any given machine could have.

To make sure that there are not tons of useless low-quality items floating about a new machine should be introduced that takes an item of lower quality and attempts to make it higher quality if it fails, it dumps the same quality as the input out.

Why ?

In FFF #376 (the very next FFF after quality) We are introduced to quality research, take that and the boost from max level quality modules and we would very quickly reach the incredibly low cap of 300% productivity, this would stunt the growth of the factory, which is clearly unacceptable. so short of giving each item a custom return value (based on how much productivity was used to make it) there is no other method of removing the cap except through the removal of the recycler.

One other option

There is one other option

We could have the recyclers return less of a percentage of the ingredients from a recipe the further its productivity research is done (each level does 10% so at research level 10 the recycler would return only 12.5% of ingredients (25% base)).

This solution would also allow the removal of the productivity cap as well

[Illiander42]

> a new machine should be introduced that takes an item of lower quality and attempts to make it higher quality if it fails, it dumps the same quality as the input out.

That completely trivialises the quality mechanics. There would be no point in doing anything else but that machine.

[Epb7304]

That completely trivialises the quality mechanics. There would be no point in doing anything else but that machine.

I am not a game designer, I was just proposing a potential solution, not necessarily the best one, I'm curious, assuming the recycler were not in play, how would you choose to fix the issue described?

[mrvn]

You could have a machine that can update the quality of an item. But unlike your proposal when it fails it destroys the item. Or it has a chance to output a better item, the same item or nothing. The point though would be that there is a risk of loosing. Updating the quality costs material, not just time (and power).

How you would rationalize how the machine actually improves the quality would be interesting. Maybe say it's quality testing. The items actually have the better quality, they just aren't recognized as such and the test reveals the actual quality. Sometimes the test is destructive though.

Note: there still wouldn't be any reason to use anything but this machine but it would preserve the fact that upgrading quality is something you only do for special cases, at least at first. Too costly to do for everything.

[evanrinehart]

> a new machine should be introduced that takes an item of lower quality and attempts to make it higher quality if it fails, it dumps the same quality as the input out.

Dyson sphere program has a building like this. The deuterium fractionator. Loop its output back to input to get what's basically a hydrogen to deuterium factory. It's not much more interesting than a building which just turns hydrogen to deuterium very slowly. Which makes no sense either way.

[Illiander42]

How you would rationalize how the machine actually improves the quality would be interesting. Maybe say it's quality testing. The items actually have the better quality, they just aren't recognized as such and the test reveals the actual quality. Sometimes the test is destructive though.

Amusingly, that's the actual process for computer chips.

[computeraddict]

we would very quickly reach the incredibly low cap of 300% productivity

I would hardly call 20-30 levels of infinite research per recipe "very quickly"

[DarkShadow44]

Sorry, but I think that's a terrible idea.

The recycler is a great concept and long awaited by a bunch of people. I'd rather have that than potentially infinite productivity. I'd rather have the recycler, even if it means having a limit (that I probably won't reach anyways).

To make sure that there are not tons of useless low-quality items floating about a new machine should be introduced that takes an item of lower quality and attempts to make it higher quality if it fails, it dumps the same quality as the input out.

As others noted, this would lead to simple loops like in Dyson Sphere Program, hardly interesting.

We could have the recyclers return less of a percentage of the ingredients from a recipe the further its productivity research is done (each level does 10% so at research level 10 the recycler would return only 12.5% of ingredients (25% base)).

You mean when I research productivity I get punished with lower recycling yields? I don't think that makes any sense.

[Ranakastrasz]

I would flat out prefer having the recycler over having any productivity at all, if I had to choose between the two.

[mrvn]

We could have the recyclers return less of a percentage of the ingredients from a recipe the further its productivity research is done (each level does 10% so at research level 10 the recycler would return only 12.5% of ingredients (25% base)).

You mean when I research productivity I get punished with lower recycling yields? I don't think that makes any sense.

Having the recycler use a negative productivity kind of makes sense.

If you recycle an existing item you get less resources. But the fewer resources would recreate the item again because the higher productivity creates the item cheaper. If you recycle a new item you get less resources, but you also spend less to create the item. So yeah, for old items you don't get back what you put into. But what you get back is worth more than it was before the research.


Where I see this failing is productivity modules. That still would allow you to produce the item with a ton of productivity modules and then recycle it without, because why would you put (negative) productivity in a recycler?

So the 300% cap for modules would still be needed but research based productivity could be unlimited and would be balanced by a negative productivity on the recycler.

[coppercoil]

Updating the quality costs material, not just time (and power).

I agree with that. So let's make a special expensive recipe with x2.718 cost and producing time factor.

[Ranakastrasz]

Updating the quality costs material, not just time (and power).

I agree with that. So let's make a special expensive recipe with x2.718 cost and producing time factor.

Are you suggesting an... e^xponentally increasing cost?
I mean, that has to be intended as a joke, right?

[coppercoil]

Are you suggesting an... e^xponentally increasing cost?
I mean, that has to be intended as a joke, right?

By saying "x" aka "*" aka "multiply" I mean that an appropriate cost factor may be set to match the original idea. Setting up a splitter filter and recycler in the loop is not er... challenging enough. Not interesting. Despite the recycler may be useful in other area, producing quality items may be "linear", just more expensive.

[Ranakastrasz]

Are you suggesting an... e^xponentally increasing cost?
I mean, that has to be intended as a joke, right?

By saying "x" aka "*" aka "multiply" I mean that an appropriate cost factor may be set to match the original idea. Setting up a splitter filter and recycler in the loop is not er... challenging enough. Not interesting. Despite the recycler may be useful in other area, producing quality items may be "linear", just more expensive.

I meant more the usage of 2.718, I.e. e, the exponential constant.

[coppercoil]

I meant more the usage of 2.718, I.e. e, the exponential constant.

I think I should have written some another interesting number . It means just quality expenses. I have no real numbers, so I wrote random number I like.

[EustaceCS]

TL;DR

The recycler shouldn't be introduced since it forces a cap on productivity bonuses, which go against the very nature of "the factory must grow"

I'm under impression that Recycler introduction is not the most pressing matter to which Productivity cap attends to...
We don't know enough to justify it yet.

What we know to date:
- Recycler is planned but where in Tech Tree it would gonna appear is currently unknown
- Global Productivity boosts for particular Intermediates are planned but where in Tech Tree these would gonna appear is currently unknown
- Productivity bonus cap is planned to be introduced

We also know that when Productivity cap is reached, we can swap redundant Productivity modules with Speed modules. . .

[Illiander42]

- Recycler is planned but where in Tech Tree it would gonna appear is currently unknown

The implication is that it's up near beacons and blue chests.

- Global Productivity boosts for particular Intermediates are planned but where in Tech Tree these would gonna appear is currently unknown

The implication is that they'll be white science after the first couple of levels.

And they'll be running on exponential cost curves, so *really* expensive.

[Tertius]

Why ?

In FFF #376 (the very next FFF after quality) We are introduced to quality research, take that and the boost from max level quality modules and we would very quickly reach the incredibly low cap of 300% productivity

With vanilla Space Age the maximum productivity bonus of an assembling machine 3 with 4 legendary productivity modules is +100%, resulting in 200% productivity or doubling the output. It's not possible to get higher productivity with additional items. The devs give modders the option to again double productivity to 400% with their mods, but that's a hardcoded limit then.
Keep in mind, the 300% is the bonus, resulting in an absolute 400% productivity - 4 items instead of 1 without productivity bonus.

Global productivity bonuses gained from research are nothing more than increasing numbers even more. They boost everything equally, so the only thing they do is increasing every number a bit. The limit in the end is the throughput of inserters and the fastest belts. If the productivity is so high inserters and fastest belts are saturated with even a single machine, that's the real limit. Combined with the speed bonuses from beacons and legendary speed modules and legendary assembling machines, I guess we hit that limit well before the productivity cap becomes relevant.

In my opinion, we need to play the game to be able to judge if these values and limits are balanced. But to outright refuse to accept the limit of +300% is a bit premature.

[Ranakastrasz]

I would be absolutely shocked if the limit of 300% is not itself moddable.

[Nidan]

I would be absolutely shocked if the limit of 300% is not itself moddable.

It is moddable

This is also why we created an overall machine limit on productivity to be +300% (modifiable by mods if needed)