Planetary Annihilation's Economy System

I still don't like it, seems too....rewarding for players who lose power.

WTF? Are you serious? They lose buildpower after all.

Can't we just wait for Alpha before discussing this any further?

We should stop any discussion besides fanart and release date guessing games then.

Its fun to speculate!

Don't you guys think a powerstall was just a little too punishing?

Sigh, ok, do I really need to pull up specifics in math to prove this to you guys? It guess it does illustrate how it's hard for people to really understand what's happening in a real time economy system of any kind.

So the question is - will the person who pauses their production during an energy stall be better off (have more stuff) than the person who does not?

First, some assumed unit stats:

Engineer: Builds at 1m/s, uses 10e/s, is busy assisting factory.
Factory: Builds at 10m/s, uses 50e/s
Tank: Costs 50m
Mex: Extracts 1m/s, uses 10e/s
Power Gen: Produces 10e/s. Costs 50m

We ignore rolloff time for the purpose of this example.

Scenario 1: Stable production
5x Engineer
1x Factory
15x Mex
0 Tanks
25 P-Gens

Production begins
Metals: +15m/s, -15m/s, net 0 difference
Energy: +250e/s and -250e/s, net 0 difference

We can produce exactly one tank every (50/15) 3.33 seconds, or 0.3 tanks per second (t/s). In 5 minutes (300s), that is 90 tanks.

Scenario 2: Energy Stall Production, No Management

Imagine we lose 5 power gens to a raid. 20% loss in power. We happen to have 5 extra engineers on hand to rebuild power gens.

+200e, -300e.
+??m, -??m

First thing that happens is energy is prorated.
Then, metal usage is prorated, and then metal extraction is prorated, and metal usage is prorated again based on production.

10x Engineers only get 66% of their energy (-66e)
1x Factory only gets 66% of its energy (-33e)
15x Mex only gets 66% of their energy (-100e)

Metal usage drops from -10 to -6.6 for Engineers
Metal usage drops from -10 to -6.6 for the Factory
Metal Production drops from +15 to +10

HERE IS THE KICKER. Metal usage is now 13.2.... but it gets dropped down to -10 because the metal extractors have been prorated.
Engineering production is now -2.5(construction), -2.5 (assist) and -5(factory)

Power gen replacement takes (50 * 5) = 250m / 2.5 = 100 seconds (lets assume for simplicity that all 5 gens are started at the same time)

Tank production drops from 90 tanks to (-7.5m/s = 0.15t/s for 100 seconds, and then returns to normal for the next 200 seconds. (15) + (60) = 75 tanks

END RESULT:
0 Additional metal. 15 tanks worth of mass lost forever. 50 extra seconds of being without power.


Scenario 3: Energy stall Production, Managed Power

Same as above, we lose 5 gens. But now we have a priority system that automatically pauses less important projects. In this case:
0 - Metal Extraction
1 - Energy Construction Projects
2 - Other Construction Projects
3 - Factories
4 - Engineers assisting factories.
(4 being lowest priority, 0 being highest)

15x Mex get full power (-150e)
5x Engineers begin constructing (-50e)
5x Engineers pause (0e)
1x Factory pauses (0e)

Metal usage for new engineers goes to -5
Metal usage drops from -5 to 0 for engineers assisting
Metal usage for Factory drops from -10 to 0
Metal Production stays at +15

Power gen replacement takes (50 * 5) = 250m / 5 = 50 seconds (lets assume for simplicity that all 5 gens are started at the same time)

Tank production drops from 0.3t/s to 0 for 50 seconds.
Metal storage fills by +10m/s for 50 seconds (500 metal)
Power generation is back online in 50 seconds (half the time), and production resumes as normal for 250 seconds. (0.3 * 250) = 75.
Player now has 500 metal stored, which allows him to build 5 more power gens, which in turn can support 5 more mexes.. and so on.

Even though the production was paused for a certain amount of time, the total number of tanks produced was the same, and no metal was wasted.


ANSWER: Player who pauses his production (manually or by a UI mod) during an energy stall will be better off than the one who does not.

Math isn't my strong point, so do tell if I've screwed up the calculations. It seems pretty conclusive though.

First: give me the right numbers and i will prove you that my distribution is not worse then switching stuff on and off.

Second: Again you are mixing up stuff. You are mixing up energy distribution for a given fixed scenario of production. With the distribution of buildpower for some projects over time. Where the outcome can change the buildpower itself. As i studied math for 4 years, i can tell you that this aint a easy general solveable task. This could be solved by conditional probability, game theory or chaos theory.

What you are asking for is the best possibilty for economy over time.
What i want to solve is the best economy distribution at a given moment.

Edit: Fractal could be a solve to over time stuff.
Edit2: Can you give me for your example the fixed m/e ratio of the enginers and the building costs of a generator?
Also mexes generate 10m and use ?e
All per sec

You are arguing 2 points.

A priority system, and switching off production automatically.

Your factory and 5 engineers are switched off, freeing power for mexes and engineers, automatically.

The difference here is that one requires the player to pause the engineers and the factory to do scenario 2, and rebuild power with the greater build power.

The second automatically switches off the factory and engineers, leaving the player to find them and then do scenario 2.

And the third assumes that the player is building tanks and switches nothing off and simply conducts scenario 1.


Your points indicate not that a priority system nor automatic UI widget is needed, but more a player to actually take direct action to pause stuff to build the power-plants.

So you are arguing against what I believe is a part of the game-play for a player to actively react after a raid on their power.

And secondly that this scenario takes place inside of a completely safe world where for 5 mins after a raid a player can afford to build no combat units at all and that 15 tanks worth of metal is important in the greater scale of things.

I feel that your example is a ship in a bottle, and has no practicality in a real game.

EDIT:

My math is a bit dated, but for the first 160 seconds scenario 1 is more practical.

And even then, scenario 2 still requires no UI or priority change.

EDIT2: Come to think of it, why not just have 1 engineer build power and leave the remaining power and mass to keep building tanks? It will take longer to build all 5 power plants but you'll be in a much, MUCH better position to further attack your enemy.

All in the 'assumed units stats' section. As for where you think I'm arguing for the wrong thing, I have no idea how the 'best economy distribution at a given moment' does NOT equal 'best possible economic outcome over time'. Obviously this is in the realm of studying economy only, with the main goal to produce as much as possible with the given constraints.

I'm not arguing for two points. I'm arguing that when you experience a energy crash, you need to pause / unpause your economy to prevent mass loss. Neutrino has generally indicated he doesn't want to look at priority systems in the first pass, so I'm going to work with that and assume the player manages priority by manually turning things on and off.

The question is - will the person who pauses their production during an energy stall be better off (have more stuff) than the person who does not?

It would be extremely helpful if you matched up your criticisms of a scenario with the numbers I've given. From what I can tell, you're trying to argue that since you select & stop the engineers assisting the factory, you would in the same move tell them to start building power. However, in my scenario I'm talking about having 10 engineers, 5 assisting at optimally balanced resource usage, and 5 standing by (perhaps only just rolling onto the scene at the time of the attack).

Yes I realize this doesn't list all the actions the player COULD take... but it's not a discussion of strategy, it's a discussion of the economic system in controlled circumstances. The principles still remain.

And I am done here as I never reference a scenario 3.

I agree with Pawz's assertion that pausing things can benefit you (same applies to TA, SupCom etc). This is a no-brainer, as it allows focusing of energy and metal to power or produce specific things as required, or get more metal in the long run. I do not agree however that when factoring everything into it, it is always the best thing to do. As there is no automated system that chooses the best unit(s)/structure(s) to pause in 100% of cases, it takes time and attention to get the optimal amount of benefit. In some cases, pausing everything works out best because it's easiest. In cases where you are only just stalling, this is really, really, bad, and it's better to manually switch one or two things off.

Compare this to building an experimental while also constructing units in Sup Com. Sometimes, you want to focus on the experimental, and pause the tank production. This is akin to Pawz's suggestion of pausing the factory to focus on power construction. Other times, you want to keep the factory running, even if it slows everything down, because you are still making tanks in the meantime. A few tanks now can be way more useful than more tanks later, even if you save metal by doing it. Extra metal doesn't help you if you are dead. Furthering your example, say the enemy are still in your base after killing some power - all your factories are now paused, and you're no longer producing defenders. A trivial example of where that system is not optimal.

Essentially, you've just defined "optimal" as some fixed target - the most metal - when in fact it's constantly changing.

I want to know why people think they need a priority system based on previous games of a similar style (IE TA & Supcom). I personally never played so tight with energy that it was ever a problem, but I'm wondering if others did (pro players?). I've heard that supcom's system is extreme punishing of energy crashes, does it have widgets or whatever to turn off everything except mass extractors (not counting metal makers since that's a unique energy issue)? It seems to be a lot of talk about theoretical scenarios that don't really happen very often (especially after a little experience with the game).

I'm impressed with the passion you guys have for this subject.

I still think making mexes take 0 energy is the simple fix for all this. From what I can tell the only real concern people have is long term loss of metal production. If extractors are unpowered you will only lose building / defenses. If they are powered I suppose we could just auto prioritize them. Both of those options are far simpler than building a general priority system which is were you guys bigtime lose me. It's like using a shotgun to kill a mouse.

I actually agree that no energy metal extractors fixes at least 90% of the problem.

I'm fairly sure I've already endorsed a no-energy metal extractor.

+1 from me!

No. where are the actual costs of a power generator all if found was :

And i dont get that tbh.

Well it is different in the case:
Given:
Income Now: 100m/1000e
Production possibility: 100m/ 800e
Question how to distribute energy to produce most

Versus:
Income Now: 100m/1000e

What to do to have most stuff in 5/10/15 minutes
Or what to do to to get Energy up fast again ...
... the first thing doesnt care for this at all its all about the maxing of the build power, wich still leaves the important decissions to the player (example parallel vs seriell production), while the second one is all about strategy and economy mixed up

This depend strongly on how the system handles the distribution.
Use the system as described in OP pausing player will be better.
Use other systems i.e. what i descirbed pausing wont make you better (for raw build power).

Fair enough!

Please re-visit my post on page 13. I proposed an elegant system that is intuitive, doesn't "prioritize" either automatically or through player input, and still enables critical low-demand items like extractors to work in low-power conditions.

Yup, I read that. Interesting idea.