Calculated: Trade Ports vs. Refineries

Hehe try culture as well. I always found culture a better investment early game when you couldnt do much econ betetr then trade ports since 1 culture towerd could boost 3-4 worlds quite rapidly if well positioned.

I'd be happy if you could confirm this once you have the files to check. Earn karma if it's true  

The problem is that culture's benefit is even more dependent on the layout of your empire than refineries are.  Trade ports not so much.  In any case, usually you want to balance your trade/refineries in such a way to minimize your use of the black market.  This is a huge black hole of your economy and you'd save tremendous amounts of wealth by avoiding using it.

Hey RA, since you LOVE old threads, here is one where we discussed Refineries and I did some math. 

https://forums.sinsofasolarempire.com/306945

This was back in the 1.03 days when the black market was low...I also have lower numbers for things like trade port income and income for each extractor in a refinery's domain (.06 was what I used back then...did this stuff change on us at some point...I mostly focus on the military side of things, so maybe I missed it). 

Since I don't have a Sins install at the moment, and I'm on dial up so I can't download the dev kit and look at the flatfiles, I will certainly defer to some kind person who feels like getting the actual numbers.  The methodology is still solid once new numbers are plugged in -- this is how I developed my strong appreciation for a well placed refinery with current black market prices.

My in game experimenting back then found that refineries were unaffected by culture.

I don't really recommend bumping the thread because this was about 16 months ago and I think some of the numbers have changed, so it will just confuse people who don't read the dates on the posts.

Hmm..  Its nice that someone has done this.  Well, Amish, you are getting some much deserved  from this.  I don't give it out often and sort of keep a mental record in decimals...  Each thing someone does earns them +.09 Karma in my book..  I only deal with integers and don't round, so only once someone gives me 10-12 reasons will they get it.  In other words, you deserve it.

Thanks for the link Cykur. Although this thread is quite old it still does the math and explains the occasions when refineries do make sense.  

Woot Cykur. I miss ya bro. I will attempt use my Amish powers of prayer to bring you good fortune in the future.

I don't dispute the math you've done RA, but I think there is one factor missing in your analysis: the cost in terms of logistics slots.  Not the cost of developing, but the fact that the refinery or trade port sits there and sucks up slots.

Your analysis compares return on investment -- which one pays off quicker in terms of creds and resources only.  This is entirely valid if we assume that you haven't filled all the logistics slots in your empire.  Once you have run out of places to build things, though, the absolute income from the building comes into play. 

Looking at it another way, the trade port may provide a better % return on investment, but the refinery allows you to invest more total resources in your logistics infrastructure (possibly at a lower % return).  Using this analysis, in the very long term (ie VERY long games), you are better off producing whichever structure provides more ABSOLUTE income --the price of the building becomes less relevent as it amortizes over more time.

Using the above assumptions, I would tend to use RA's calculations in shorter games, or games where I didn't expect to max out development, with a slow shift to a different way of valuing things as the game drags out.

bdgackle

I fail to see were you are getting at. Both trade ports and refineries use up 4 logistic slots. All civilian buildings use 4 slots. So their is no slot advanatge to be spoken ofve. Thus why it is not covered sicne it is irrelevent.

As for the advantages over time in short and long term it also doesn't matter. What the calculations take concideration ofve is how fast doe sthe building pay for it's self and how much income it will produce. For trade ports we can easily predict and increase the income each port will generate. With refineries predicting the income is more of a gamble and it will not increase like the income of ports will. So ports income will always be superior to refinery income because of the way it works with expending trade lines.

Now if refineries also had trade lines they would kick trade ports asses. But it does not.

In addition you need to calculate the trade in value of minerals vs credits. With the black market you need on average 5credits to buy 1mineral. How ever you will sell 1mineral for 2.5 credits. If you have credits you can buy the mineral you want/need directly, with a refinery you will be producing 1 mineral more most likely and selling that mineral to buy teh other. Thing is you will need sell 2x the ammounts of mineral that you buying. But refineries in general don't produce mineral in such an excest like trade ports can poduce. So you lose out on the black market.

yep. you fail.

ok... if you have a long chain, and another trade port generates... 3 credits per second..

and say... a refinary at that spot would generate .96 resources per second (*4.5)~4.32 credits per second... (10 mines connected... (thats an 4 roid planet, a 3 roid planet, and a 3 roid planet... not that hard to find)  which would you rather have hoging up your logistics slots?

lets see... I only have 4 slots left...

 if i build 4 trade ports... thats 12 credits per second...

if i build 3 refinaries... and one port for my trade chain thats 15.96 credits per second... costs more... but the limit as time goes to infinate... the initial cost does not matter...  the refinarys make more $$$ per slot, as long as you have enough resoruces connected to them.

SO... there IS a slot advantage.

It's hard to argue against tradeports because the longer the line gets, the stronger it gets. This is why when you're Advent and TEC, tradeports make the most sense first. Ok, Advent don't have "real" refineries, and TEC refineries are tier 4. 

With Vasari, it actually makes sense to build refineries first. Yeah. I said it. As Vasari you end up trading a LOT of your minerals away for creds, and while you only get a fraction back (1/2?) on the black market, it's worth noting that you can sell minerals at your own price. Go to the black market page. Look in the lower corners. You'll figure it out. Someone has to buy your minerals, however, so make sure you have time to wait for someone to do it.

The biggest deal with Vasari is that you need 4 labs for trade ports, but only 3 for refineries. If you're stuck on a small number of planets, or somewhere that you just can't get a decent 4 planet trade line going, get refineries. Really. You'll probably end up making more if you find the right spot to put them.

All bdgackle is saying is that a well placed, more profitable refinery is a better long term slot investment than a tradeport for very long games.  I said the same thing 1.5 years ago in the other thread I linked.  The longer the game, the more the construction / research considerations become negligible compared to the long term profit potential.  The only time this doesn't hold true is when you have very long trade chains...then a tradeport could actually become more valuable than a well placed refinery.  Well placed refineries are very profitable...assuming you were going to buy the resources in question, and they usually cover a mix of resources.  As I have said in multiple threads, you want both tradeports and refineries for a well balanced economy.

Your missing the biggest factor of all, refineries trump trade by the fallowing behavior.

You may have 200 creds per second coming it, but it's all null and void if I have the resource income to match it.  I put my resoucres up for sell, you buy like crazy and all the money goes into my account.  Resouce prices inflate (they fall back to default at too fast a rate IMO), and I get more and more credits for every 100 metal/crystal I sell you.  Remember this next time metal or crystal prices are hovering above 600 for long periods of time.

Consider two situations (I'm totally fudging the numbers to make the point as clear as possible):

1) Trade port that pays for itself in 15 mins, and generates 2 creds / minute of value

2) Refinery that pays for itself in 30 mins, and generates 3 creds / minute of value

Short run, the trade port is a better deal here, per RA's ROI argument. 

Long run, and assuming all logistic slots are full, Refinery is the better deal here -- because it allows me to make better use out of the limited logistic slots at my disposal.  I understand that the slot cost is the same, my point was that slots are a limited resource, and maximizing resource ROI doesn't always maximize economic potential.

What if you had a "minor" trade port that produced half the income of the real one, still took 4 slots, and was FREE.  Would you not still choose the regular port if your time horizon was long?  Think about that for a minute.  Your reason for choosing the "real" trade port in that hypothetical example is the same reason you might choose a refinory (or TP) when the ROI said to do otherwise.

RA -- I certainly wasn't saying you are wrong -- you'd wipe the field with me in a real game, no doubt, and I've learned most of what I know about strats for this game from your posts.  Just trying to inject another factor into the discussion that I thought was missing.

to me tradeports need to bring in alot more money then what they already are. unless 1 credit is like the equivilant of a 1,000 dollar bill.

Matching credits so they are directly compairable to modern currency shouldn't be a high priority IMO.

These are undefined units, make them equal to whatever you want.

That's silly. Reminds me of people who see foreign currency exchange rates and go something like: Wow 1 dolalr is 35 xxx. Like it even makes sense to have that kind of comparison.

My preference is to build trade ports early to get the cash rolling in. Only mid game when I need to start building a sizable fleet do I build refineries. Even then I place them strategically so they bring the maximum resource income rather than just putting them in every available slot.

I'm not sure if saying this swings the vote for people one way or another, but my take is this.

If you have the ability to make a 4+ (preferably 6+) trade port line, then that is the more sensible economic movie early, especially if you're TEC or Advent.

I build refineries first only when I have that crap set up where there is no chain of planets BUT there are a ton of neutrals that I can get resources from with my refineries. This is primarily with Vasari, because with TEC the odds are you're going to have some viable way (say by getting +4 logistic slots so dead asteroids aren't useless) to get a trade line.

It's in this wierd set-up that refineries are the more sensible economic investment.

To me, tradeports are NOT worth it if you can't get a decent chain going. This is why in 1v1 my technique tends to be to spot the longest trade line I can get early, and build 1 tradeport at each point in the chain. This is the fastest method to payoff tradeports. If you'd like to me to explain the math behind it, sure, I'll back it up, but I'm lazy.

RA, so you are saying in games where you neither have opportunities for a solid trade chain nor good spots for refineries your are building none of the two? Doesn't that screw your mid- to end-game economy as you are probably the only player who has no "extra income" of any sort?

Not at all. There's a pretty good rule of thumb. If I don't have good starting position, my enemy does. It's rare that there won't be at least ONE sensible place for refineries, but it's fairly common that your starting positon doesn't support a trade line....but your enemy's does. In this spot I gear everything for rushing. I want a knock out blow early against said opponent.

If you don't move early and "rush" when you have bad economic starting position, you will lose purely by macro. Your enemy will produce more than you if you don't move your ass. Once I kill the dude, THEN I start getting an ECO started.

Makes total sense. I've noticed whenever this is tried online though, whatever teamates I have keep going eco. rather than feeding me or attacking. Of course the enemies allies don't do that leaving me with an achy butthole. So I thought "gee, I guess I'm just dumb then." makes e smile, seeing that typed out by an amish dude. I thought you weren't allowed to use computers?

So the takeaways, as I undertand them, are:

1. Culture is (usually) a better investment than either refineries or TPs.

2. If we price resources at their "buy" price--because we're cash rich/reource poor--refineries (usually) beat trading posts.  If we price at the "sell" price--because resource rich--TPs are (usually) better.  So build refineries if/when that will get you from cash rich to neutral.  Build TPs if needed to get from resource rich to neutral.  Note that, because in this case we price resources at the "sell" price we do not follow RA's calculation in the OP.  The TP costs 1256.25 and pays back in ~10.5 minutes.

3. If/when resource neutral, build refs and TPs in balance w/each other--or not at all to save slots for labs.

4. None of the above calculations, nor RA's, include the cost of the enabling tech, so the real payoffs are slower in all cases.

5. Value of resource enhancing techs depends on empire size and on whether you are short or rich in that resource.  In my (limited) experience, for about 6 planets one 7% bump is worth 0.2-0.3/second.  Take the example of the first Advent ore tech (level 2): 600+50+100. 

If ore poor, ore=4.5, crystal=2.25.  Cost = 600+225+225 = 1050.  Return = 0.9 (~20 mins) to 1.35 (~13 mins).  Not great.  If ore rich, cost = 600 + 112 + 450 = 1172.  Return = 0.45 (~43 mins) to 0.67 (~28 mins). 

Should you ever improve your rich resource?  Best case is ore for TEC.  1st level tech costs 512, returns 0.45 (~21 mins) to 0.67 (~14 mins).  So not quite never.

6.  Should one prefer resource tech or refieneries?  Whether to improve your poor resource depends on empire size, how good is your best refinery opportunity (if not Advent), and how many refs you would build.  Best case for refs, you are TEC and poor in both resources.  Refineries cost 2800 (tech) + 2850 each.  So 3 are way more efficient than 2 or 1.  Let's say 3 refineries x 8 extractors = 2.4/second (worth 10.8) vs total cost of 11250.  Pays off in 1040 seconds ~ 17 minutes.  But it is much worse if building 3 refs tips you over into being rich in one resource--in which case you probably don;t build all three.

If resource tech will produce 0.3/second/level, then level 1 and 2 resource techs--for the poor resource only--will almost always be a better bet than refineries, as well as saving slots.  Even level 3 and 4 could be better than refs barring a map especially favorable for refs.

Don't forget how maximum all. affects income... people always tend to brush that under the rug.

Maximum what?