Construction analysis

[Slick]

Ok, you have just colonized that huge breathable and intend to make it a research planet (25 facilities). You intend to build 24 Research Facility III's (700 research points each) and 1 Central Computer Complex III (+30% to existing facilities; each RF III now makes 910 rp).

What is the most efficient way to build?
Do you build 24 RF's and THEN build the computer? or the other way around? or is the best way to build the computer somewhere in the middle of the queue?

There are 2 competing motivations: 1) get the max output at time-now (this one suggests that you should build the computer complex first), 2) get the max total research points possible (this one suggests you should build the computer complex Last).

Being an engineer and remembering those old max/min problems from high school calculus (20 years ago), this really bothered me and I just HAD to figure this out sooner or later. I had this FEELING that the right time to build the computer complex was somewhere in the middle of the queue. So I did some calculations and testing.

Given/assumptions:
base construction rate: 2000 Minerals, 2000 Organics, 2000 Radioactives per turn (settings.txt)
cost of RF III: 2500 M, 0 O, 0 R (facility.txt)
cost of CCC III: 12000 M, 0 O, 3000 R (facility.txt)
planet is in a secure area so don't factor in defense concerns
Emergency build is available, rules in settings.txt

I bet some of you have already done this, and the answer may be obvious to you, but I had to prove it to myself.

Before I post my results, I would like to hear your thoughts on this *evil grin*

Slick

[spoon]

My guess would be to build 24 Research Facilities II (so you can build one per round), then the computer facility, then upgrade to Research Facilites III. And use emergency build when you have 15 turns left of stuff to build/upgrade.

At least that's what I'd do to save myself from having to do math.

-Spoon

[Arkcon]

Originally posted by Slick:
Ok Given/assumptions:
base construction rate: 2000 Minerals, 2000 Organics, 2000 Radioactives per turn (settings.txt)
cost of RF III: 2500 M, 0 O, 0 R (facility.txt)
cost of CCC III: 12000 M, 0 O, 3000 R (facility.txt)

Are you sure about these costs, I thought the computer complex cost some organics?

Anyway, the math was done by someone else a long time ago, and they'll post their results

What I do, with absolutely no math done by me, is build 9 facitities, the computer complex 1, and then the rest. I'll upgrade as I acquire tech.

I notice if I do that for many planets, I'll soon outstrip the AI in production even if it has a small advantage in total planets. And that's all the math I really want to do for a game.

[Jmenschenfresser]

If we are taking into consideration only the time that things are building...assuming you do not use the upgrade route, you build the complex Last.

A gain of 42,000 points.

Correct?

[Wardad]

For Resource Efficeny:
Build a space yard first.
Then build the research III facilitys in one turn.
The space yard will also speed the build on the computer facility.
Build the computer complex when your empire can afford it.

If you would rather burn resources, Then do not build the shipyard.
Build the Research IIs and upgrade later.
The computer complex III can be built midterm.
The upgrades can be done Last on Emergency Build.

Let's hear from the MathaMagicians!

[ October 07, 2002, 20:05: Message edited by: Wardad ]

[capnq]

I never build a level 1 computer facility until the planet has 11 matching production facilities in place. You need that many for the level 1 complex to add more production than another production facility in the same slot would.

Level 2 only needs 6 facilities, and Level 3 needs 4.

[ October 07, 2002, 20:53: Message edited by: capnq ]