Beacons don't always improve crafting speed

[4wry]

I originally shared this on Reddit, however, I don't believe people fully understood the problem.

To be clear: I understand that there are diminishing returns in 2.0 and that adding more beacons yields a progressively weaker effect. The point is that Adding mixed beacons can actually have net negative effects, but the expectation is that any added beacon should always have a positive effect, even if very marginal.

In the example picture the left setup has 240% crafting speed on the assembler, but it is slightly below the required output to achieve a compressed belt with an even number of units. When adding another beacon with a speed & efficiency module to 'push it over the threshold', contrary to expectations this actually has a net negative effect as the crafting speed drops to 222% in the right setup when adding a beacon.

I believe the problem here is that diminishing returns are calculated based on the total number of beacons. Adding even an empty beacon increases that number and reduces the total effect. Thus unless all beacons have the tier 3 speed modules you will have a net negative effect. Thus, it is a bad idea to intersperce e.g. beacons with efficiency modules to save power etc., because these hurt the active speed module beacons, and slightly incrementing the output (as it would have been in 1.1) cannot be achieved by adding a beacon.

The gripe is that the effect calculation of diminishing returns should not be based on the number of beacons, but the number of specific modules (type). E.g. adding a beacon with efficiency modules should not hurt the effect of speed modules.

• Empty beacons should not lower the effect of active ones

• Adding a beacon with a lower tier should still have a small positive effect

• Efficiency beacons should not diminish speed beacons, and vice versa

The point is that diminishing returns should still apply, but never decrease an existing effect. A notion is that, instead of using the number of beacons, the calculation needs to sum up the total effect of all modules by type, then apply a factor/function of diminishing returns based on that total.

The above would entail that players would never get negative effect when adding beacons, and empty beacons will not have negative effects either. Also mixing speed and efficiency will not be penalized, so efficiency modules/beacons can be added where there is net energy gain.

[Loewchen]

This is working correctly, the effect transmission reduces with each additional beacon.

[4wry]

But is it intended to go purely by the number of beacons? As pointed out in the post the issue is not the diminishing return per se, but the way they are applied that entails negative returns.

[Loewchen]

But is it intended to go purely by the number of beacons?

yes

[4wry]

I read that FFF, but I don't fully agree that using the number of beacons as opposed to the cumulative effect before diminishing returns are applied is the optimal approach. Using beacon numbers alone hurts design freedom since it has negative impact when mixing modules, or using lower tier modules. What the graphs do not capture is them going down when more beacons are added, as in the example.

Do you suggest I post this not as bug, but as improvement suggestion?

[Loewchen]

You can make a suggestion here: viewforum.php?f=6.